The Bitcoin Field Manual

Part 01Ground truth

Chapter 01The nine pages that reset my career

In 2011 I was an electrical engineer at a robotics company outside Toronto. I sold control systems and haptic devices into university research labs. MIT, Stanford, Georgia Tech. My professional world ran on transfer functions, sampling rates and stability margins, and I liked it, because in that world a claim is either true or the hardware falls over.

Someone sent me a nine page PDF called "Bitcoin: A Peer-to-Peer Electronic Cash System." I read it at a coffee shop on a Tuesday afternoon when I was supposed to be reviewing a mechanical spec.

What caught me was not the money. It was the control problem. Distributed systems people had spent decades on the question of how a set of machines that cannot trust each other agree on a single ordered history. The clean version of that problem is called Byzantine fault tolerance, and every solution I knew of assumed you could enumerate the participants in advance. This paper did not enumerate anyone. It made lying expensive instead. It attached a physical cost to the act of proposing history, and then made honesty the cheapest strategy available.

That is a control loop. Measure, compare, correct, repeat, with the correction applied through incentives rather than through a supervisor. I had spent eight years designing loops that had a supervisor. This one did not need one, and it had been running for two years without stopping.

It took me a long time to move from that recognition to conviction. Curiosity is cheap. Conviction takes months of reading things you did not plan to read: monetary history, the Bretton Woods arrangement, what actually happened in August 1971 when the dollar stopped being redeemable for gold, and what happened to the price of a house and a year of university afterwards. By the time I finished I was not interested in a clever database. I was interested in the fact that for the first time since coins, somebody had built money that no committee could dilute.

I left engineering. My friends thought I had lost it. Building anything bitcoin related in 2013 was like trying to explain the internet in 1993, except with worse press. I co-founded India's first bitcoin exchange, and we were in a fight almost immediately.

The fight is the part worth telling. India's central bank issued a circular that cut regulated banks off from anyone dealing in digital assets. Not a ban on bitcoin. A ban on the banking rails underneath it, which is more effective and harder to argue with. It nearly killed the company. We fought it, and the case went all the way to the Supreme Court of India, and we won. That decision opened the market for everybody who came after us, including our competitors, which is the correct outcome and also an expensive one.

The exchange went on to serve more than two and a half million users. I have since worked in exchange business development at a global venue, in regulated fund structures in Canada, and I now run a family office and venture studio, and sit on the board of a publicly traded humanoid robotics company. Different rooms, same underlying job: take something new and make it legible to people who control capital.

That is the job this book is about. Not trading. Not predicting price. The unglamorous, durable work of building a business whose foundation is sound money, at a moment when the number of people who genuinely understand that foundation is still small relative to the number of institutions that need to.

I am going to be direct about what this book is not. It is not a guide to getting rich from price appreciation. It contains no forecast, no target, and no timeline, because I do not have one and neither does anyone who tells you they do. It is not written for people who want a passive position. It is written for operators, and every chapter is built to be executed rather than admired.

Chapter 02Bitcoin is a monetary bet, not a technology bet

The single most common analytical error I see, and I have watched intelligent people make it in board meetings, is treating bitcoin as a technology company you can value with technology comparables. People ask what its revenue is, who its competitors are, and what happens when someone builds a faster version. All three questions assume the wrong category.

Bitcoin does not compete with Visa. It does not compete with a payments startup. It competes with the practice of monetary expansion, which is to say it competes with the ability of an institution to create new units of account and spend them before the effect shows up in prices.

Once you hold that frame, the properties stop looking like features and start looking like the entire product.

Twenty one million units, enforced by every node independently rather than by a promise. That is the whole thesis compressed into a number. Not scarce in the way a limited edition print is scarce, where the issuer promises not to print more. Scarce in the way that any attempt to print more produces a chain that other people's software simply refuses to accept.

A fixed issuance schedule that steps down roughly every four years and does not care what the price is doing, what an election produced, or what a central bank would prefer. Supply that does not respond to demand is a strange object. Every other commodity answers a price rise with more production. Bitcoin answers a price rise by making it harder to mine, which is the opposite reflex, and it is the reflex that makes the schedule credible.

Settlement that finalises without a counterparty. When a block containing your transaction is buried under enough subsequent work, no institution can reverse it, because there is no institution positioned to. That is not a small convenience. It changes what a contract can assume.

Verification available to anyone with a laptop and a few hundred gigabytes. You do not have to trust an auditor's opinion about the supply. You can check it yourself, in about a day, on consumer hardware. There is no other monetary system in history where an ordinary person can independently verify the total outstanding.

Here is the business consequence, and it is the reason this chapter sits at the front. If you think you are in a technology market, you will position on features, and features get copied. If you understand that you are operating inside a monetary transition, you position on trust, custody, jurisdiction, and time, and none of those get copied on a quarterly release cycle.

I learned this by watching competitors. While we were arguing our case in court, several venues in the same market were running acquisition campaigns and adding coin listings, on the theory that the market wanted more choice. What the market wanted was to be certain their money would still be there in the morning. The listings did not survive. The legal precedent did.

I want to give you the version of this argument that works in a boardroom, because the version that works online does not, and watching somebody deploy the online version to a group of directors is painful.

Do not start with the monetary system. Start with their balance sheet. A company holding a large cash position has made an investment decision whether or not it thinks of it that way, and the decision is to hold an asset with an expanding supply for an indefinite period. Nobody signed off on that. It happened by default, and the reason it never appears as a decision is that the erosion is denominated in the same unit as the measurement, so it never shows up as a loss on any statement they read.

That framing does something specific. It moves the conversation from "should we do something unusual" to "we are already doing something, and it was never chosen." Boards respond to the second framing because it is a governance question, which is their actual job, rather than an allocation question, which feels like speculation to them.

Then give them the comparison in their own terms. Over a ten year horizon, what is the expected purchasing power of a unit whose supply is set by a committee responding to political conditions, against a unit whose supply is fixed, published, and independently verifiable by any participant. You do not need a forecast to make that comparison. You need only the difference in how the two supplies are determined, which is a fact rather than a prediction.

Then stop, and let them raise the objection, which will be volatility. Do not pre-empt it. When it comes, agree with it immediately and move it to where it belongs: this is a question about sizing, horizon and policy, not about whether the asset is sound. What proportion, over what horizon, under whose authority, with what custody. Those are questions a board knows how to answer, and answering them is the engagement.

The mistake I watch people make is arriving with the ideological case and treating the room as unconverted. Directors do not need to share your worldview and mostly will not. They need a governance-shaped question they can process, a policy they can approve, and a person who will still be reachable in three years. Give them those three things and the worldview becomes irrelevant, which is the point.

Chapter 03You do not need to be technical. You need to be legible.

Every week someone tells me they cannot build in this space because they are not a developer. It is the single most common self-disqualification I hear, and it is close to backwards.

Look at who actually gets paid. The pipeline of demand runs through chief financial officers, treasurers, general counsel, compliance officers, board members and family office principals. Those people have a specific problem, and the problem is not cryptography. The problem is that they are being asked to make an irreversible decision in a domain where they cannot evaluate their own advisors.

They do not need a lecture on elliptic curves. They need somebody who can sit in a room, hold both vocabularies at once, and say what happens if this goes wrong.

I have watched brilliant protocol engineers lose that room in ninety seconds. Not because they were wrong, but because they answered a governance question with a technical answer, and the person across the table concluded, correctly, that the engineer did not understand what was being asked. The question was never "how does multisignature work." The question was "who gets fired if this is lost, and can I put that person's name in a memo."

So the scarce skill is translation, and translation has a price because it removes career risk from a decision maker. That is what you are actually selling. Not information, which is free and abundant. Judgment under uncertainty, delivered in language the buyer's board will accept.

Being non-technical is not a handicap here, provided you fix the one thing that actually disqualifies you: shallowness. There is a floor of real understanding you have to clear, and below that floor you get found out, usually in front of the exact person you wanted to impress. Chapter four is the floor.

What a non-technical background gives you, if you have it, is the ability to notice which parts are confusing, because they were confusing to you six months ago. People who have been technical their whole lives lose access to that. They cannot remember not knowing, which makes them poor guides.

My own version of this is that I came from control systems, not from finance and not from cryptography. It meant I spent my first year translating everything into loops, feedback, stability and failure modes, which felt like a handicap until I noticed that operations people and risk people think in exactly those terms and had never heard anyone describe bitcoin that way.

Whatever domain you came from is a translation layer nobody else has. A tax partner sees basis tracking and reporting obligations. An insurance underwriter sees a novel loss category with no actuarial history. A supply chain operator sees settlement finality applied to trade documents. An emergency physician sees triage protocol, which is genuinely a better frame for incident response than most of what the security industry publishes.

Find yours. It is not a marketing angle. It is the reason you will see something the people already here have stopped being able to see.

Chapter 04The floor: five things you actually have to understand

There is a real threshold of competence, and I want to be specific about it, because "learn bitcoin" is useless advice. Below the floor you are a liability to your clients. Above it you can operate for years without ever writing code.

Here is the floor. Five things, and you should be able to explain each one to a sceptical CFO without notes, and defend it against the obvious objection.

First, supply and issuance. Not "there are twenty one million." That is trivia. You need to know why the cap is credible, what the halving schedule does to miner revenue, what happens to security when the subsidy approaches zero, and what the honest uncertainty is around that transition to a fee driven security budget. That last part matters. The person testing you will ask the hard version, and if you have never thought about it, you fail.

Second, custody. This is where most of the money and nearly all of the catastrophe lives. You need to understand private keys, seed phrases, hardware signing devices, multisignature quorums, the difference between a service holding keys for you and a service holding one key of several, and the specific ways each arrangement fails. Not theoretically. Concretely: what happens when a signer dies, when a device is lost, when a company holding one key goes out of business, when a spouse cannot get in.

Third, the transaction lifecycle. What a UTXO is and why it makes bitcoin accounting different from balance accounting. What a confirmation actually means, and why "six confirmations" is a convention rather than a law. What fee estimation is doing and why it fails during congestion. Why a transaction can sit unconfirmed and what to do about it. This is the layer where operational embarrassment happens.

Fourth, the settlement and payment layers built above the base chain. Lightning primarily: what a payment channel is, what liquidity means in that context, why inbound liquidity is the constant operational problem, and what class of businesses it genuinely unlocks versus what gets oversold. Being able to say clearly what Lightning is bad at will earn you more credibility than any amount of enthusiasm about what it is good at.

Fifth, the regulatory surface in your jurisdiction. Not global regulation, which nobody knows. Your jurisdiction, specifically: what triggers a money transmission or money services licence, what the tax treatment of a disposal is, what reporting obligations attach to a business holding the asset on its balance sheet, and where the boundary is between advice you may give and advice that requires a licence you do not have. Knowing where your boundary is, and saying so out loud, is one of the strongest trust signals available to you.

That is the floor. Notice what is not on it: mining economics in depth, cryptographic primitives, protocol development history, and the entire universe of other digital assets. You can build a large practice without any of them.

I would give yourself three months of serious reading to clear the floor, and I mean serious. Ten hours a week, with notes, with the whitepaper read three times rather than once, and with the practical exercises done rather than skipped. Run a node. Build a multisignature wallet with three keys and then deliberately destroy one and recover. Send a transaction with a fee that is too low and watch what happens. You will learn more from an afternoon of that than from a month of podcasts.

Chapter 05The objections worth taking seriously

You will be tested. Usually by one person in the room whose job is to find the flaw, and usually with a question they have asked before and had answered badly. How you handle that moment decides the engagement, and the way to handle it is not to have better enthusiasm. It is to have already thought harder about the objection than they have.

So here are the ones that actually come up, with the honest answer rather than the rehearsed one. Notice that in two cases the honest answer includes a genuine unresolved question, and saying so is the thing that wins the room.

Energy. The version you will hear is that this consumes an absurd amount of electricity for no productive purpose. The first half is a fact and the second half is the claim, and you should concede the first half immediately rather than arguing about the number, because arguing about the number signals that you think the number is the problem. The substantive answer is that the energy is what makes history expensive to rewrite. It is not overhead attached to a payment system, it is the security mechanism itself, and a version that used less energy would be proportionally cheaper to attack. Then make the second point, which is the more interesting one: this is a buyer of last resort for energy that has no other customer, it is interruptible on a timescale that almost no other industrial load can match, and that combination makes it unusually useful to a grid operator managing intermittent supply. If the person opposite is technical about power systems, that second point is where the conversation gets good.

No intrinsic value. The honest response is that intrinsic value is not a coherent property of any monetary good, including the one they are paid in. Gold's monetary premium was never about its industrial uses. A national currency has no intrinsic value either; it has legal tender status, tax demand, and a network of people who accept it. Bitcoin's value comes from the same place any monetary good's does: from the combination of properties that make it good at the job, and from the number of people who have concluded the same. That is not a weaker foundation. It is the same foundation, held up more explicitly.

Governments will ban it. I have a specific answer to this one because I have been on the receiving end of the nearest thing to it. India's central bank did not ban the asset, which would have been difficult to enforce. It cut off the banking rails, which is what a state actually does and is far more effective. It nearly killed our company. We fought it in court for years and the Supreme Court struck it down. So my answer is not that it cannot happen, because it happened to me. My answer is that it happened, the network kept producing blocks throughout, and the legal position afterwards was better than before. States can make access difficult in their jurisdiction and can impose real costs on businesses. They have not demonstrated an ability to stop the ledger, and the recent movement in most major jurisdictions has been toward frameworks rather than prohibition.

Quantum computing. Two separate questions get merged here and you should separate them. The signature scheme would eventually be vulnerable to a sufficiently capable quantum computer, and the same is true of the cryptography protecting essentially every bank, government and messaging system in the world, which is why standards bodies have been publishing post-quantum algorithms. The mining function is far less exposed. The practical answer is that this is a migration problem on a long timeline, faced by the entire digital economy simultaneously, and bitcoin has a mechanism for protocol upgrades and a strong incentive to use it. Anyone presenting this as a bitcoin specific extinction risk has not thought about their own bank.

The security budget. This is the good one, and if somebody raises it properly you should tell them so. The block subsidy halves roughly every four years and eventually approaches zero, at which point the security of the network is paid for entirely by transaction fees. Whether fee revenue at that point is sufficient to sustain the current level of security is genuinely unresolved. There are reasonable arguments in both directions: fee markets during periods of congestion have produced substantial revenue, and settlement of high value transactions is exactly the sort of activity that supports high fees. But it is not proven and the transition is decades out and dependent on adoption patterns nobody can forecast. The right answer is to describe the mechanism, describe both cases, and say that it is open. Do not manufacture certainty. The person asking usually knows it is open, and is testing whether you will pretend otherwise.

Why not a faster or cheaper alternative. Because throughput was never the constraint being solved. The base layer optimises for verifiability by ordinary participants on ordinary hardware, and every design that increases throughput at the base layer does so by increasing the cost of running a node, which concentrates verification, which removes the property the whole thing exists for. Throughput belongs in layers above. That is not a compromise, it is the architecture, and it is the same reason no serious settlement system in the world processes retail volume at the settlement layer.

One rule about all of these. If you do not know, say you do not know, and then say what you would need to find out and by when. I have never lost an engagement by saying "I do not know, let me come back to you on Thursday." I have watched people lose engagements by guessing, once, in front of somebody who knew the answer.

Chapter 06What changed by 2026, and what did not

I want to update the map, because a lot of bitcoin business advice was written for a world that no longer exists, and following it now will put you in the wrong place.

Here is what genuinely changed.

Spot exchange traded products in the United States were approved in January 2024, and equivalent regulated access has broadened elsewhere since. This mattered more than the price reaction suggested. It converted "I cannot hold this because my mandate does not permit it" into "I can hold this in the instrument my mandate already permits," which removed the single largest structural blocker for allocators. The consequence for operators is that the first meeting is no longer an education meeting. Your prospect's board has already had the conversation. They are now asking implementation questions, and implementation questions are billable in a way that awareness questions never were.

Corporate treasury allocation stopped being one company's eccentricity and became a recognised category with recognised specialists, accountants who have done it before, and an emerging body of practice. That is good news and bad news. Good, because the buyer no longer has to be convinced the category exists. Bad, because "we can help you think about bitcoin on your balance sheet" is no longer differentiated. You now need a specific competence inside the category.

Accounting treatment improved materially. For years the intangible asset treatment in the United States meant a company recognised impairment when the price fell and nothing when it rose, which made the reported financials of any holder look absurd. Fair value measurement fixed the asymmetry. If you advise companies, this is a change you must be able to explain, because it altered the internal politics of the decision at a lot of firms.

Regulatory posture moved from hostility to frameworks in most of the jurisdictions that matter. Europe built a comprehensive regime. Several Asian financial centres built licensing paths. The United States moved from enforcement-led ambiguity toward legislated categories. I am deliberately not summarising the detail, because it changes and because the detail in your jurisdiction is your job, not mine. The structural point is that the era of "nobody knows if this is legal" is over, and with it the era of businesses whose entire moat was a willingness to operate in fog.

Now here is what did not change, and this list is more important.

Custody is still the hardest operational problem, and people still lose coins in the same six ways they did a decade ago. The failure modes are not novel. They are the boring ones: single points of failure, undocumented recovery, one person who knows everything, backups nobody has ever tested, and an heir who does not know the assets exist.

Education is still the bottleneck inside organisations. The board approving an allocation and the operations team executing it are not the same people, and the second group has usually been given no training at all. That gap is where most implementation failures live and where a lot of honest work sits.

The supply of people who can hold both vocabularies is still small. It has grown, but demand grew faster, because every institution that entered created a whole internal population of people who now need to understand something they never asked to learn.

And the fundamental case did not change at all. Twenty one million. Independent verification. Settlement without a counterparty. Everything above it moved. The base did not, which is exactly what a base is supposed to do.

If you internalise one thing from this part, make it that split. The surface moves constantly and you have to track it. The base does not move at all, and it is the only thing worth putting a decade of work on top of.

Part 02The operator's business model

Chapter 07Expertise is the only asset that starts at zero cost

If you want to build something in this space and you do not have capital, a team or a product, there is exactly one starting position that works, and it is selling your own judgment by the engagement.

I am not romantic about consulting. It is a job with a ceiling, it does not compound while you sleep, and if you stay in it forever you have bought yourself an expensive job rather than a business. Part five is about getting past it. But as a starting position it has properties that nothing else has.

It requires no inventory and no infrastructure. It requires no outside capital, which means no dilution and no board and no obligation to a growth curve you did not choose. It produces revenue in weeks rather than years. And most importantly, it is the fastest known method for finding out what the market actually wants, because a client paying you is a far stronger signal than a survey, an interview, or your own conviction.

Every durable product business I know of in this space was preceded by somebody doing the work manually for money. Custody products came from people who had personally set up quorums for wealthy families and got tired of the parts that repeated. Compliance software came from consultants who had filled the same forms out enough times to know the shape of them. Treasury advisory practices became treasury platforms. The manual version is not a stepping stone you tolerate. It is the research.

Here is the mechanism that makes the first year work, and I want to state it plainly because it is the thing that gives people permission to start.

You do not need to be the world's leading expert. You need to know materially more than the person paying you, about the specific problem in front of them, and you need to be honest about the boundary of what you know. That is it. That is the entire qualification.

I have never once been the smartest person about bitcoin in a room I got paid to be in. I have frequently been the only person in the room who had done the specific thing before and could say what went wrong last time.

The obvious objection is impostor syndrome, and I want to deal with it properly rather than dismiss it. If your internal experience is "I am not qualified for this," the correct response is not affirmation. It is to check whether you have cleared the floor from chapter four. If you have not, the feeling is accurate and you should go clear it. If you have, the feeling is a calibration error, and the fix is reps, not confidence. You will feel qualified after the fourth engagement, and there is no shortcut that skips the first three.

The second objection is that the market is crowded now. It is more crowded than it was, and it is still not crowded. Look at any mid-market company with real cash on the balance sheet and ask who inside it can answer a custody question. Usually nobody. Then ask who their advisors are, and you find a general accounting firm with one partner who read a summary. That is the actual competitive landscape at the size of business you can realistically reach in year one.

The third objection, and the honest one, is that consulting income is lumpy and you are the product. Both true. Chapter nine and chapter ten are about turning lumps into something that resembles a schedule.

One more thing about the first year, because the arithmetic of it catches people out and the surprise is avoidable.

Year one is not a year. It is roughly seven earning months, and you should plan on that number rather than twelve. The first quarter produces almost no revenue by design, because you are building the position, the evidence and the offer. The second quarter produces discounted work. Real money starts arriving somewhere in the third quarter and there is a further lag between an engagement being agreed and cash being in the account, which for corporate buyers is routinely thirty to sixty days after an invoice that was itself issued at the end of the work.

So the practical requirement is that you can survive nine months without meaningful income from this. That is not a motivational statement, it is a constraint, and the way people handle it determines a great deal about how the year goes.

If you have savings, define how much of it is committed to this and hold the rest back, so that a bad month is a bad month rather than an existential event. If you do not, keep the income you have and build this alongside it, at ten to fifteen hours a week, and accept that the ninety day plan becomes a hundred and fifty day plan. That is a completely legitimate path and I would take it over the alternative every time.

The alternative, which I want to name because it is common and it is a trap, is quitting with three months of runway on the theory that pressure produces results. It does produce results. It produces the wrong ones, because a seller with three weeks of runway asks smaller questions, discounts before being asked, chases prospects who have already said no in every way but the words, and takes the engagement that pays now over the engagement that builds the position. The buyer can hear all of it. Financial runway is not a personal virtue, it is an input to how you sell, and treating it that way is the difference between choosing your clients and accepting them.

Chapter 08The six lanes of the work

"Bitcoin consultant" is not a positioning. It is an absence of one, and it reads to a buyer as somebody available for anything, which is the same signal as somebody good at nothing.

There are six distinguishable lanes. Pick one. You can add a second in year two, but pick one now.

Strategic advisory. You sit with executives and a board and help them decide whether, how much, under what policy, and with what governance. The deliverable is usually a written policy and a board memo, not an implementation. This lane has the highest fees and the highest credibility bar, because the buyer is senior and their downside is personal. It also has the longest sales cycle. If you have executive or board experience in any industry, this is the lane where that experience transfers most directly.

Implementation. You do the hands-on work: setting up the custody arrangement, integrating payment acceptance, connecting the accounting, writing the runbooks the operations team will actually use. Lower per-hour than strategy, but far more repeatable, and it produces a defined artifact that either works or does not, which makes it easy to sell to buyers who are sick of consultants. This lane also converts into a product faster than any other, because you will build the same seven documents for every client and eventually you will just sell the seven documents.

Compliance and regulatory. You help firms map their obligations, build their procedures, prepare for examination and stay inside their licences. Enormous need, high recurring revenue, and a hard prerequisite: you must be clear about where advice ends and where regulated legal or accounting advice begins. Stay inside your line and say so in writing. Cross it and you will have a very bad year.

Security. Key management, quorum design, incident response, recovery testing, inheritance planning. Existentially important and chronically underbought until something scares the buyer. The unlock in this lane is not persuasion, it is the recovery drill. Run a client through a live simulation where a key is unavailable and they cannot recover, and you will never have to explain the value again.

Education and training. Board sessions, executive workshops, team curricula, ongoing internal programmes. Underrated as a lane. It is the easiest first engagement to sell because it is low risk for the buyer, it puts you in front of everybody who matters at once, and roughly half the time somebody in the room raises the problem that becomes your next, larger engagement. I have had a single half day session generate three years of work.

Technical. Node infrastructure, wallet integration, Lightning operations, architecture review. Requires actual engineering skill, narrows your market considerably, and pays accordingly. Only enter this lane if you can genuinely build.

Now the harder move, which is picking a vertical to cross with your lane.

"Security for bitcoin holders" is a lane. "Custody and inheritance design for family offices with an operating business" is a position. The second one lets the buyer recognise themselves in your description, which is the entire job of positioning. It also means that when somebody at a family office conference is asked who does this, there is a name that comes up, and vague generalists do not get named.

Cross your lane with the industry you already have credibility in. If you spent fifteen years in insurance, you are not a bitcoin consultant. You are the person who can talk to insurance carriers about a novel asset on their books, and there are perhaps six of those people.

Chapter 09The arithmetic of a one person practice

Most people who fail at this fail at arithmetic, not at bitcoin. They never actually work out what the business has to do, so they cannot tell whether they are ahead or behind until the year is over.

Let me do the arithmetic in public.

Start with the number that matters, which is billable capacity. A full time solo operator does not bill forty hours a week. Nobody does. Between selling, admin, learning, writing and the work that recovers a client relationship after something goes wrong, a realistic sustained billable load is somewhere between fifteen and twenty five hours a week, and the upper end of that is not sustainable across a full year without a pipeline that runs itself.

Call it eighteen hours a week, forty four working weeks. That is 792 billable hours in a year, and I would treat anything above 900 as a claim requiring evidence.

Now the rate. At two hundred an hour, 792 hours is about $158,000. At five hundred, about $396,000. At a thousand, about $792,000. Those are gross revenue figures before tax, before tooling, before travel, and before the fact that year one will not be a full year.

Three things fall straight out of that arithmetic.

The first is that the rate is the lever, not the hours. Going from eighteen billable hours to twenty five is a 39 percent increase and it costs you your evenings. Going from three hundred an hour to six hundred is a 100 percent increase and it costs you a harder conversation. People consistently choose the evenings, because the evenings do not require them to say a number out loud and wait.

The second is that a seven figure practice cannot be built on hours alone by one person. At 792 hours, a million dollars requires $1,263 an hour, which exists but is a narrow market. So the path past the ceiling is not a higher hourly rate. It is a different unit of sale, and that is chapter nine.

The third is the one people miss. Capacity is fixed but the pipeline required to fill it is not. If your average engagement is $12,000 and you want $300,000, you need 25 engagements. If your close rate on qualified conversations is one in four, you need 100 qualified conversations, which at a 20 percent qualification rate from initial contacts means 500 initial contacts, spread across a year. That is roughly ten a week, every week, forever.

Ten a week is the actual job. Nobody says this at the start because it sounds like drudgery, and it is, and it is also the only part of the model that has never once failed me when I did it and never once worked when I did not. Every flat quarter I have ever had, in every business I have run, has traced back to a period eight to twelve weeks earlier when I stopped having conversations because I was busy delivering.

Track two numbers weekly and you will be ahead of most people in this field.

Conversations started. Not emails sent, not posts published: actual two way conversations with somebody who could plausibly buy. If that number goes to zero for two weeks, your revenue goes soft roughly a quarter later, with the delay being the entire problem.

Effective rate, which is revenue divided by all hours worked, including the unbilled ones. Not your list rate. Your list rate is a story you tell. Effective rate is what is happening. Watch it move when you change how you package the work, and you will learn more about your own business in one quarter than in three years of reading.

Two footnotes on the arithmetic, both of which I got wrong for longer than I should have.

Booked is not collected, and the gap will hurt you in the first year specifically. Corporate procurement moves at its own speed regardless of how quickly you delivered, and a thirty day invoice at a mid-sized company routinely becomes fifty. Take a deposit. Thirty to fifty percent on signature is standard, it is not rude to ask for, and I have never had a serious buyer object to it. The people who object to a deposit are correlated with the people who will be slow to pay, which makes the question useful in itself.

And your list rate is not your effective rate, which is why chapter nine exists. Take last quarter, add every hour you worked including the proposal you wrote for free, the call that did not convert, the scope you revised three times, and the client email you answered on a Sunday. Divide revenue by that number. The result is frequently forty to sixty percent of the number on your website, and the gap is not waste, it is the actual cost structure of a service business. The reason to compute it is that it tells you which changes matter. Raising the rate moves it. Reducing scope creep moves it. Adding hours does not.

Chapter 10Price for the outcome, not for the hour

The hourly rate is a trap disguised as a fair arrangement, and I stayed in it far longer than I should have.

Three things are wrong with it.

It caps you at your calendar, which we just proved arithmetically. It punishes you for getting faster, because the more expert you become the fewer hours the same job takes, which means your income falls as your competence rises. And it puts you in the wrong category in the buyer's mind. Hourly is what you pay a resource. Fixed fee for a defined outcome is what you pay a firm.

The move is to sell defined engagements with defined deliverables at a fixed fee.

A custody architecture review. A treasury policy with a board memo. A team curriculum with an assessment. An examination readiness assessment. An implementation with a runbook and a live recovery drill. Each one has a scope, a start, a finish, an artifact, and a price that does not move because you were quick.

Setting that price is the part people get wrong, so here is the method I use.

Estimate the hours honestly, including the parts you will not enjoy. Multiply by the rate you want. Then look at the number next to the client's downside, and ask whether it is obviously small. A treasury policy that takes forty hours at four hundred is sixteen thousand dollars. The decision it governs might be eight figures. Sixteen thousand is not the expensive part of that transaction and both of you know it.

Then say the number flat, with nothing after it. The number, and then silence. Most people say the price and then keep talking, and every sentence after the number is a discount you volunteered.

Two more rules I hold to.

Never price as a percentage of anything the client gains. This matters most on anything touching capital raising, and I will be explicit because the boundary is not a style preference. Charge a flat fee for work performed. Never a success fee, never a percentage of an amount raised, never compensation contingent on a transaction closing. That structure makes you a participant in a securities transaction rather than a service provider, which is a licensing question in nearly every jurisdiction and a fast route to a problem that outlives the fee. Flat fee, paid for the work, whatever the outcome.

And never guarantee an outcome you do not control. You control the quality of the work, the timeline, and the artifacts. You do not control whether a board approves, whether a fund allocates, whether a raise closes, or what price does. Say what you will do. Do not say what will result. Anyone in this industry promising you a result is either inexperienced or selling you something, and the difference does not matter much to your downside.

The retainer is the destination. Once a client has been through a project and the relationship works, propose a monthly arrangement: a fixed number of hours, a standing call, priority access, and ongoing monitoring of whatever moves in their world. Retainers turn a lumpy consulting income into something you can plan around, and they are far easier to sell to an existing client than any new project is to a stranger.

Chapter 11The minimum viable offer

Do not design your service. Design one offer, sell it eight times, and let the eighth one tell you what your service should be.

A minimum viable offer has five properties, and if any one is missing the offer will not sell.

It solves one specific problem, expressible in a sentence the buyer would use themselves. Not "bitcoin advisory." Something like: your company holds bitcoin and there is no written procedure for what happens if the person with the keys is unavailable.

It produces a defined artifact. A document, a configuration, a curriculum, a report. Something that exists when you are finished and can be handed to somebody who was not in the room. Engagements that produce only conversation are hard to sell, hard to price, and impossible to refer, because the buyer cannot show anybody what they got.

It has a fixed price and a fixed timeline. Both stated up front. Two weeks, this number, this is what you receive.

It is deliverable inside about thirty hours. This is the constraint people resist and it is the important one. Under thirty hours you can complete engagements while still selling, you fail cheaply if the offer is wrong, and you get to eight repetitions inside a year instead of two.

And it opens the door to something larger. The custody review reveals the training gap. The board session reveals the policy gap. The policy reveals the implementation. Design the first engagement so that doing it well makes the second one obvious to the client rather than to you.

Your first three should be discounted, and I mean genuinely discounted, at something like half of where you intend to land. Not because your work is worth less, but because you are buying three things you cannot otherwise get: proof the offer is coherent, a written account of a real result, and the experience of having done it. Be explicit with those first clients about the trade. You are getting a lower rate, I am getting your candid feedback and, if the work is good, permission to describe it. People say yes to that arrangement readily because it is honest.

Then raise the price. Not gradually. After the third engagement, quote the full number to the next prospect and find out. The worst case is they say no and you have learned the ceiling, which is information you needed anyway.

One warning from my own history. Do not build the thing before you sell it. I have watched capable people spend four months building a beautiful bitcoin treasury framework, complete with templates and a website, before speaking to a single buyer, and then discover that the buyers they could reach did not have the problem the framework solved. Sell first, in a conversation, with a scope you sketched on a call. Build the artifact while you are being paid to build it.

That is the whole model in one picture. One offer, small enough to fail cheaply, deliberately built so that finishing it well surfaces the next problem. Everything in part five about scaling assumes you have run this ladder at least once with real money on it.

Chapter 12The first conversation

An offer that nobody has heard is a document. The thing that turns it into revenue is a conversation, and most people run that conversation badly in a specific and fixable way: they talk.

I have sat in on a lot of these, my own included, and the pattern is consistent. The seller is nervous, the nervousness expresses itself as filling silence, and the filling consists of explaining bitcoin to somebody who did not ask. Forty minutes later the buyer knows a great deal about the seller and the seller knows nothing about the buyer, and there is no basis on which to propose anything.

Here is the structure I use. It is thirty minutes and roughly eighty percent of it is them.

Open by framing, in two sentences, then stop. Something like: I want to understand what is actually in front of you and whether this is something I can help with. If it is not, I will say so and point you somewhere useful. Is that alright. That framing does two things. It sets the meeting as a diagnosis rather than a pitch, and it puts the possibility of a no on the table immediately, which lowers the temperature for both of you.

Then ask, and keep asking, in roughly this order.

What made you take this call now. The word that matters is now. Something changed: an auditor asked a question, a board member raised it, a person left, a competitor did something, a number got large enough to be noticeable. Whatever that thing is, it is the actual driver of the purchase, and everything else in the conversation should be attached to it.

Walk me through how it works today. Get the current state in operational detail. Who does what, on what system, with what documentation. You are listening for the gap between what they describe and what they think they described, and it is usually large.

What happens if this is still true in a year. This is the cost question without asking about money directly. If the honest answer is "nothing much," you have discovered that the problem is not urgent, which is more useful than a polite conversation about scope.

Who else has to agree. Ask this early rather than late. The number of proposals that die because a person nobody mentioned had an objection nobody heard is very large, and it is entirely preventable by asking a plain question in the first meeting.

What have you already tried. This tells you what they will not buy again, and what they were burned by, and how they evaluate. It also stops you proposing the exact thing that failed for them last year.

Then, and only then, say what you would do. Briefly. Two or three sentences, in their words, aimed at the thing they said made them take the call. Not your methodology. Not your background. What you would do and what they would have at the end.

Then the number, said flat, followed by silence.

That last part is the part people cannot do, and it is worth practising on low stakes calls until you can. Say the price and stop. Do not add a justification, a comparison, a payment option or a softener. Every sentence after the number is something you gave away without being asked. The silence will feel much longer to you than it does to them, because you are the one experiencing it as a verdict.

A few things I have learned to do differently over the years.

Ask the disqualifying question. Somewhere in the middle, ask something whose answer could end the engagement: is there a budget for this at all, is this actually a priority this quarter, is there someone internally who thinks this is unnecessary. People are afraid to ask these because they might get an answer they do not want, which is precisely why they are worth asking. A no in the first meeting is far cheaper than a maybe that consumes six weeks.

Take notes visibly and read them back. "So the situation is this, the concern is this, and the thing that would make it worth doing is this. Have I got that right." Hearing their own situation described accurately by somebody else is a stronger trust signal than any credential you could produce.

Send the scope within twenty four hours, in their language, using the words they used. Not a template with their name in it. The scope should be short enough to read on a phone and specific enough that a third person could tell whether it had been delivered.

And follow up more than twice. Most engagements I have closed were closed on the third or fourth contact, not the first, and the reason people do not follow up is that it feels like pestering. It is not, provided each contact contains something. Send the note you wrote about their exact problem. Send the relevant thing you read. Silence from a buyer is almost never a decision, it is almost always their week having happened to them.

Part 03Authority, and how it is actually built

Chapter 13Authority is evidence, not volume

There is a version of authority building that is really just posting, and it does not work. I watched a lot of people try it between 2017 and 2021, and the ones who got followers and no clients all made the same mistake: they optimised for reach when the buyer was screening for evidence.

Think about what a chief financial officer actually does when your name comes up. They do not count your followers. They open a browser and spend about eleven minutes trying to answer one question: has this person done the thing before, and is there any proof.

Everything you publish should be aimed at those eleven minutes.

That reframe changes what is worth making. A single detailed write-up of how a custody quorum should be structured for a company with three signatories, two jurisdictions and a succession problem is worth more than three hundred short posts, because it survives the eleven minutes. The short posts do not, because they are indistinguishable from the short posts of somebody who read the same article you did.

Here is the hierarchy of evidence as buyers actually weigh it, roughly strongest to weakest.

A result you produced for someone comparable to them, described specifically, ideally in that person's own words. Nothing beats this and nothing is close.

A piece of work that demonstrates judgment rather than knowledge. Judgment shows up as an opinion with a cost attached. "Here is why I would not use that arrangement, and here is what it costs you if you do" reads as expertise. "Here are the five benefits of multisignature" reads as a summary of somebody else's page.

Something durable with your name on it in a context you did not control. A conference programme, a podcast that is not yours, an industry publication, a panel. The credibility is not in the audience, it is in the fact that a third party made a selection and you were in it.

A body of writing with enough depth and consistency that it is obvious you have been paying attention for years rather than months.

And at the bottom, social metrics, which correlate with nothing your buyer cares about.

The practical protocol I use, and have used since long before it was fashionable, is to publish the thing I wish had existed when I was stuck. Every real engagement leaves behind a problem you had to think properly about. Write that up. Strip the client entirely, keep the mechanism. Do it after every engagement and in a year you have twelve pieces of genuine evidence, which is more than almost anyone in this field has.

One rule I hold absolutely. Never name a client, a prospect, or their numbers without explicit written permission for that specific use. Not in a post, not in a deck, not in a talk, not as a throwaway line on a podcast. This is not only ethics, though it is that. It is commercial. The buyers with the largest problems are exactly the ones most alarmed by a consultant who talks about other people's holdings, and the moment somebody hears you describe a named client's custody arrangement, they have learned everything they need to know about what you would say about theirs. Anonymise or leave it out.

Chapter 14The learning budget

Your ability to produce anything on that evidence ladder is capped by what you actually know, so the learning budget is not self improvement. It is capital expenditure, and it should be scheduled like one.

For the first six months, ten hours a week, minimum. If you can find twenty for the first three months, take it, because the early returns are steeper than they will ever be again.

Ten hours a week sounds like a lot until you compare it to the alternative, which is charging for judgment you do not have. It is roughly the time most people spend on things they would struggle to name at the end of the week.

Here is how I would spend it, and the proportions matter more than the specific sources.

Four hours on primary material. The whitepaper, read properly, three separate times across the six months, because you will notice different things each time. Monetary history, which is the part almost everyone skips and the part that produces conviction. The mechanics of the current monetary system: how credit creation actually works in a commercial banking system, what a central bank balance sheet does, what happened structurally in 1971 and why the effects took a decade to become visible. Read the serious economics rather than the internet version of it, including the arguments against your position.

Three hours on practical work with your hands. This is the part that separates people who can talk from people who can be trusted. Run a node and keep it running. Build a two of three multisignature wallet, then simulate the loss of one key and complete a recovery. Open a Lightning channel and route a payment and then try to receive one, which is where you learn what inbound liquidity means in a way no article will teach you. Break something on a test network on purpose. The confidence you gain here is not psychological, it is earned, and it comes through in a room.

Two hours on the industry as it is now. Regulatory developments in your jurisdiction specifically. What the institutions you want as clients are actually doing and saying publicly. What the operators near you are shipping. Read primary sources: filings, consultation papers, official guidance. Secondary commentary is a way of learning what other people concluded, which is what everyone else already has.

One hour writing. Not publishing necessarily, writing. Writing is how you find out whether you understand something. I have started a great many paragraphs confident about a mechanism and discovered halfway through the second sentence that I could not actually explain it. That discovery is the whole value.

Two things to avoid.

Do not spend the budget on price commentary. It is the most abundant content in the space and the least useful to anyone building a business. You are not being paid to have a view on the next six months. You are being paid to help somebody design something that survives being wrong about the next six months.

Do not spend it on other digital assets. I say this as a matter of focus rather than tribalism. The entire thesis in this book rests on absolute scarcity and independent verification, and it is a property that either holds or does not. Diluting your attention across things that do not have that property costs you the one advantage a specialist has, which is that when you speak about your subject people can tell you have thought about nothing else for a long time.

After six months, drop to five hours a week and hold it there permanently. The field moves. The moment you stop, your knowledge begins expiring, and the expiry is invisible to you and obvious to your clients.

Chapter 15Cornerstone content, and the machine that multiplies it

Almost everyone gets the sequence of content backwards. They start with the small pieces and hope volume accumulates into authority. It does not, because a hundred small pieces on a hundred topics reads as a person with a hundred shallow opinions.

Start with one large thing. I call it the cornerstone, and everything else in your first year is derived from it.

The cornerstone is a genuinely comprehensive treatment of the single most important unresolved question in your chosen lane. Two to four thousand words. Written for the buyer, not for peers. It should take you a week, and it should be the best thing that exists on that question anywhere, which is a high bar and also an achievable one because most of what exists is thin.

Picking the question is the hard part, so here is the test. It should be a question your buyer has actually asked out loud, that has no good published answer, and that you have a defensible opinion on. If it fails any of the three, keep looking.

Some shapes that work: what a company should do about custody when the person holding the keys leaves. How a board should structure the governance around a treasury allocation, including who is allowed to sell and under what conditions. What a family office should put in writing so that heirs can actually access assets. What a regulated institution should expect during examination on this subject. Each of those is a real question with real consequences and a shortage of good answers.

Once the cornerstone exists, you multiply it. Not by rewriting it, by cutting it apart.

The core argument becomes a short piece of writing for the platform where your buyers actually are, which for most of these audiences is a professional network rather than a social one. The strongest mechanism inside it becomes a diagram, and a good diagram travels further than any paragraph. The objections section becomes a series of short posts, each taking one objection seriously. The whole thing becomes a talk, which becomes a recorded talk, which becomes something you can send. It becomes a podcast conversation because you now have something specific to say rather than general enthusiasm. And the reference sections become a downloadable checklist, which is the thing people actually keep.

That is one week of real work turned into a quarter of material, all of it consistent, all of it pointing at the same competence.

Then repeat quarterly. Four cornerstones a year, each on a different question inside the same lane, is a body of work. After two years you have eight, and at that point you have effectively written a book, and the people in your market who have not done this cannot catch up quickly, because they would need two years.

On the format question, my honest read after a lot of years. Written material does the persuading, because it survives the eleven minutes and it can be forwarded to a board. Audio and video do the trusting, because people decide whether they want to be in a room with you long before they decide whether you are right. You want both, and if you can only sustain one, write.

The compounding here is the point. One cornerstone is a good article. Four is a position. Sixteen, over four years, is a reputation that somebody starting today cannot assemble inside a decade, because the constraint is not effort, it is elapsed time and the accumulation of specific work.

Chapter 16Write so it can be forwarded

There is a specific test that separates writing which produces business from writing which produces compliments, and it is this: can the person who read it forward it to their boss without editing it or explaining it.

Almost nothing published in this industry passes that test. Most of it is written for peers, which means it assumes vocabulary the forwarding recipient does not have, adopts a tone the recipient will find unserious, and contains at least one aside that would embarrass the person who sent it. So it does not get forwarded, and forwarding is the entire distribution mechanism inside an organisation.

Write for the third reader. The first reader is the person you had in mind. The second is their colleague. The third is the person who has to approve something, has never heard of you, has eleven minutes, and is looking for a reason to stop reading. Write for that one.

Some rules that follow from it.

Lead with the conclusion. Not with the background, not with the history of money, not with why this topic matters. State the finding in the first two sentences and spend the rest earning it. Executives read the top and the bottom, and if the top does not contain a position they will not reach the bottom.

Define nothing that does not need defining, and define everything that does, once, in a clause rather than a paragraph. The instinct to explain the basics comes from insecurity about whether the reader knows them, and it is exactly what signals that the writer is used to talking to beginners.

Use their units. If you are writing for a treasurer, express things in the terms a treasury operates in: policy, limits, authority, reporting, counterparty. If you are writing for a compliance function, express it as obligations, controls, evidence and examination. The identical mechanism described in the reader's own vocabulary reads as expertise, and described in yours reads as a translation problem they now have to perform.

Take a position and attach the cost to it. "There are several approaches, each with trade-offs" is the sound of somebody protecting themselves. "I would not use that arrangement for a company your size, and here is what it costs you if you do" is the sound of somebody who has done this. The second one is riskier and it is the only one that produces business.

Make the structure visible. Headings that state the finding rather than the topic. Short paragraphs. A summary at the top that could stand alone if that is all anyone read, because for many readers it will be. If the piece contains a decision, put the decision in a box of its own.

Cut the throat clearing. Almost every draft I have ever written has a first paragraph that can be deleted entirely, and almost every draft anybody sends me has the same. The real opening is usually the second or third paragraph, where the writer stopped warming up and said the thing.

And be honest about what you do not know, in writing, explicitly. This feels like weakness and it is the single strongest credibility signal available in a written document. A piece that says "here is what is established, here is what is contested, here is where I have a view and here is where I do not" is unusual enough that people remember who wrote it.

One structural suggestion for the format that has worked best for me with institutional readers. Situation, in their words. Finding, stated plainly. Reasoning, with the counterargument taken seriously. Recommendation, specific enough to act on. What would change my mind. Five sections, two pages, and it is forwardable to anyone.

Finally, a note about the voice. Do not write in the register of the industry. The house style of this sector is loud, certain, a bit adolescent, and it is fine among people who already agree. To the third reader it sounds like something they would be embarrassed to circulate. Write plainly, at normal volume, in complete sentences, the way you would write to a colleague you respect. That alone will put you in a small minority.

Chapter 17The rooms

There is a category of opportunity that content cannot produce, and it is being in rooms.

I have traced back the origin of most of the significant relationships in my working life, and very few of them started online. They started because I was physically present somewhere, usually somewhere slightly inconvenient, and I said something specific in front of people who were paying attention.

Speaking is the highest leverage version of this, and the ladder into it is more accessible than people assume.

Start with local meetups. Every city of any size has a bitcoin meetup and most of them are chronically short of speakers, which means the barrier is not quality, it is offering. Email the organiser with a specific title and a two sentence description. Not "I would like to speak sometime." A title, a description, and a date you are available. That single change in how you ask converts at a rate that surprises people.

Then industry events in your vertical rather than bitcoin events. This is the move most people miss and it is the one that pays. A bitcoin conference is a room full of people who already agree with you and are mostly not your buyers. A treasurers' association meeting, a family office gathering, a compliance conference, an insurance industry event: those rooms contain your actual buyers, they have almost nobody credible on this subject, and you will be the only person on the programme who can speak to it. I have had more business come from being the only bitcoin person at a non-bitcoin event than from every bitcoin event I have ever attended combined.

Then the larger stages, which come from having done the smaller ones and having recordings to show.

Podcasts are the parallel track, and I say this as somebody who hosts one. Being a guest on other people's shows is the single most efficient credibility mechanism available, because it is a third party selecting you and an hour of unedited exposure to how you think. Pitch specifically: name the episode you would do, say why their audience specifically needs it, and make it obvious you have listened to the show. Most pitches fail because they are clearly form letters, which tells the host everything.

Starting your own show is a longer game and a different one. It is slow, it does not produce clients directly for a long time, and its real value is somewhere else entirely: it gives you a legitimate reason to have a substantial conversation with almost anyone. That is the asset. I have had conversations with people I could never have reached with a meeting request, because a meeting request asks for their time and an interview offers them something.

Two things about how to behave in these rooms.

Say the specific thing. Generic enthusiasm about bitcoin is noise in every room now, including bitcoin rooms. Specific competence, ideally including something you are willing to say that others in the room might disagree with, is what gets remembered. The people who get invited back are the ones who said something that cost them something.

And give without a ledger. Make introductions with no expectation. Answer questions from people who will obviously never hire you. Send the thing you promised to send within a day. Almost every good thing that has happened in my career arrived through somebody I helped for no reason, on a timeline of years, in a way I could not possibly have engineered. The people who network transactionally are visible from a long way off and they get exactly the network they are building.

Chapter 18Distribution, and the part nobody wants to do

Content and rooms build authority. Neither one reliably fills a calendar on a schedule, and if you rely on them alone you will have a business whose revenue arrives when it feels like it.

The missing piece is deliberate outbound: you, contacting specific people who have the problem you solve, directly, in a way that earns a reply.

I know the reaction, because I had it. Outbound feels like the opposite of authority. It feels like the thing you do when nobody is calling you. That framing is wrong, and it costs people years. Outbound done badly is spam. Outbound done properly is research plus a specific, useful, relevant message to a person who genuinely has the problem, and the difference between the two is entirely in the work you put in before you write.

Some things I hold to, from having run this at scale and having got it wrong first.

The list is the whole thing. A hundred correctly chosen people beat ten thousand scraped ones by a margin that is difficult to overstate. Correct means: they plausibly have the exact problem your offer solves, they are senior enough to buy, and there is something specific and true you can say about their situation. If you cannot say something specific and true, they are not on the list.

The message is short and it is about them. One observation about their situation that demonstrates you looked. One sentence about the problem you suspect they have. One low friction ask. That is the whole email. Not your background, not your credentials, not a paragraph about bitcoin. Nobody has ever replied because the sender explained themselves at length.

Volume without measurement is guessing with extra steps. Track sends, replies, and conversations. If the reply rate is low, the problem is almost always the list or the relevance of the message, not the volume, and adding volume to a bad message just burns the market you were going to sell into.

The follow up is where the results live. Most replies do not come on the first message. They come on the third or fourth, and the follow up has to add something rather than ask again. "Just checking in" is not a follow up, it is an admission that you have nothing to say. Send the thing you wrote about their exact problem instead.

Now the boundaries, and I want to be precise because this is where people get themselves into trouble and it is worth being explicit even in a book that is free.

You own every relationship, always. If you use technology or hire help for the mechanics of sending, the account is yours, the domain is yours, the inbox is yours, and every reply comes to you. Any arrangement where someone else stands between you and the person who replied is an arrangement you will regret, and if you are ever the one providing this service to others, hold the same line: you provide technology and service, the client owns every relationship that results.

Never work for a percentage of an outcome. If you help another company with distribution, charge a flat fee for the work performed. Not a share of revenue, not a success fee, and never anything contingent on a financing closing. The moment compensation is tied to a transaction outcome you have changed what you are, legally and practically.

Promise activity, never results. You can commit to the number of messages, the quality of the research, the responsiveness, the reporting. You cannot commit to replies, meetings, or deals, because you do not control the other side of the conversation. Anyone promising you a specific number of outcomes is either about to disappoint you or is selling something that will not survive contact with a regulator.

And send from your own identity, to people you could defend having contacted, about something they would genuinely want to know. That is the whole ethical test and it is also the whole deliverability test, which is a convenient alignment.

Part 04The inner game

Chapter 19Conviction is built, not felt

Everybody in this industry talks about conviction, usually as though it were a temperament. It is not. It is an artifact of specific work, and the work is identifiable, which means conviction is available to anyone willing to do it.

I did not have conviction when I read the whitepaper. I had interest. Interest is what makes you read a second thing. Conviction is what makes you say a number out loud in a board meeting and not soften it, and it took me about two years to get there.

Three components, and you need all three. Two out of three produces someone who sounds convinced and folds under a good question.

The first is monetary history. Until you understand what money was before it was issued by states, you cannot evaluate the claim that it should not be. Money emerged as a market phenomenon. Gold won its role not by decree but because it had the best combination of properties available: hard to produce, hard to destroy, divisible, portable, recognisable. Understanding that sequence changes what you think you are looking at. Bitcoin stops being a novel invention and becomes a return to a category that existed for most of human history, with the properties improved and the physical constraints removed.

Then read the twentieth century properly. The gold exchange standard, its abandonment in 1971, and what happened to the ratio between wages and assets afterwards. Not the polemic version. The actual sequence, with the arguments that were made at the time by people who thought they were doing the right thing. You will come out understanding that the current arrangement is roughly fifty years old, was adopted as an emergency measure, and has never been submitted to anyone for approval.

The second is technical appreciation, which is different from technical skill. You need to understand why the design is hard to improve on. Why proof of work is doing something that a cheaper mechanism cannot, namely importing a cost from outside the system so that history cannot be rewritten by whoever holds the most units inside it. Why the difficulty adjustment makes the issuance schedule robust to any amount of hardware. Why the conservatism of the protocol, which frustrates people who want features, is the property that makes the whole thing credible as money. Once you see the design as a set of deliberate refusals rather than a set of missing features, it reads completely differently.

The third is economic logic. Why an asset with fixed supply behaves differently from one with elastic supply. Why monetary expansion is not neutral in its effects, and why the people who receive new units first do systematically better than the people who receive them last. Why an unforecastable money supply makes long horizon calculation difficult, which is a subtle cost that never shows up as a line item and shapes everything.

Here is the test for whether you actually have conviction, and it is a behavioural test rather than an emotional one.

Can you hold your position through a sixty percent drawdown without changing anything you do? Not without feeling anything. Feelings are not the measurement. Without changing your behaviour: your allocation, your business plan, your public position, your hiring.

If the answer is no, you do not have conviction, you have a position that has been going your way. And the honest response is not to talk yourself into it. It is to go back to the three components and find the one you skipped, because that is where the vulnerability is.

Chapter 20Volatility is a measurement problem

There is a conversation about volatility that I have had perhaps four hundred times, and almost all of it is confused because both people are using one word for two different things.

Price movement measured in fiat is not the same thing as risk in the asset, and conflating them produces bad decisions in both directions.

Start with what is actually being measured. When you say bitcoin moved thirty percent, you are describing the ratio between two things, both of which move. One of them has a fixed supply and a fixed issuance schedule known to everyone. The other has a supply set by committee, adjusted for reasons that are political as often as economic, and reported with a lag. Attributing all of the movement in that ratio to the side with the fixed rules is an odd convention. It is the convention we have, because pricing in the local unit is how everyone thinks, but it is worth noticing that it is a convention.

Then think about what the movement actually is. An asset in the middle of being monetised, moving from near zero recognition toward whatever its eventual role is, cannot get there smoothly. Every step of that repricing is a market changing its mind about what something is for. That process generates large moves by construction, and it would generate them for anything undergoing the same transition. The movement is a feature of the transition, not a property of the endpoint.

None of that makes it disappear, and the operator's question is not philosophical. It is: what do I actually do.

Here is what I do, and what I recommend to anyone running a business alongside a holding.

Separate the operating business from the holding, completely and structurally. Your business has obligations denominated in your local currency: payroll, rent, tax. Those obligations do not care about your thesis. Hold enough of the local unit to cover them for a defined horizon, and treat that as an operating requirement rather than an investment decision. Everything above the operating requirement can be held in the asset you actually believe in. The failure mode that destroys people is not the drawdown, it is the forced sale during a drawdown, and the forced sale is always caused by not having done this separation in advance.

Match your time horizon to your obligations, not to your conviction. If you need money in eighteen months, that money should not be in a position that can be down materially in eighteen months, however strongly you feel about the decade. This is not a statement about the asset. It is arithmetic about your calendar.

Write the policy while nothing is happening. What triggers a sale, who authorises it, what the rebalancing rule is, what happens if something falls by half. Write it down when you are calm, and then follow it when you are not. Every catastrophic decision I have watched somebody make was made in a week when they were improvising, and improvising is a guaranteed outcome if there is no document.

And take the accounting seriously, because for any business with external stakeholders the reported numbers matter independently of the economics. Understand how the holding will appear in your financial statements and make sure your board, your auditor and your lender understand it before it appears rather than afterwards.

The part I want to be blunt about. Drawdowns are when this industry sorts itself out. Businesses that existed only because capital was cheap disappear. Talent that was there for the upside leaves and becomes available. Attention that was on twenty things concentrates on the few that are real. Every meaningful advantage I have ever built was built during a period when the general mood was that this was over. If you are operating with a decade horizon and a properly separated balance sheet, those periods are the best working conditions you will ever get.

There is a second order effect here that I want to name, because it shapes hiring and it is rarely discussed.

Drawdowns change who is available to you. In the periods when the general mood is that this is finished, capable people become reachable who were not reachable eighteen months earlier. Some of them are leaving companies that should not have existed. Some of them are simply tired of explaining their job at dinner parties. Either way, the cost of a good hire falls and the competition for them evaporates at the same moment that everyone else has frozen hiring, which is the definition of an inefficient market.

The same is true of attention. In an exuberant period, every meeting you want is competing against forty other meetings and every piece you publish is competing against a torrent. In a quiet period, the serious people are still there, still working, and much easier to reach, and a genuinely good piece of writing has far less to compete with.

The only requirement for exploiting this is that you have to still be operating, which brings the whole thing back to the balance sheet separation. The operator who kept a properly funded operating reserve gets to hire during the quiet and publish into a clear field. The operator who did not is selling assets to make payroll at exactly the moment the opportunity is largest, and the reason is a decision made two years earlier about how much of the reserve to hold in the local unit.

That is why the boring chapter is at the front of this part and not the back. The interesting opportunities in this industry are all available to people who are still standing, and staying standing is almost entirely an arithmetic exercise done in advance.

Chapter 21Decades and quarters

The most valuable structural advantage available to an operator in this field is a longer time horizon than the people you compete with, and almost nobody takes it, because it requires tolerating the appearance of being behind.

When you are building for ten years, different decisions become obviously correct.

You invest in relationships that will not pay for years, because at a ten year horizon they will pay. You build systems rather than applying fixes, because you will be the one living with the fix. You develop expertise instead of taking whatever work is available, because expertise compounds and opportunism does not. You write things down. You turn down engagements that are good money and wrong direction, which is the hardest one, and the one that separates people most.

The paradox worth understanding is that a long horizon makes you faster in the short run, not slower. This is counterintuitive and it is reliably true. When you are not required to produce a visible result this quarter, you can make a decision in a day that a quarterly-optimising competitor needs a month of internal justification to make. You can move on something unproven. You can say no immediately. Most of the slowness in organisations is the cost of having to explain, and a long horizon buys you out of a great deal of explaining.

I have watched founders paralysed by needing to show progress. They take the safe engagement, chase the visible win, avoid the thing that would take eight months, and three years later they have a business that works and cannot grow, because every decision was selected for legibility rather than for value.

Two practices that make a long horizon operational rather than aspirational.

Keep a decision journal. When you make a decision that matters, write down the decision, the reasoning, what you expected, and what would have to be true for you to conclude you were wrong. Do it in four sentences, not four pages. Then read the old entries once a quarter. This is the single highest return habit I have, and the reason is that memory quietly edits your reasoning to match the outcome. You will discover that you were right for reasons you did not have at the time, which is more useful and more unsettling than being wrong.

Run a quarterly review with the same three questions every time. What compounded this quarter. What did I do that will not matter in five years. What did I avoid because it was uncomfortable. The third question is the one that produces the value, and the reason to write the questions down in advance is so you cannot quietly swap them for easier ones.

One caution, because long term thinking has a failure mode and it is a common one. It can become an excuse for not shipping. "I am building for the decade" is sometimes true and is sometimes a story told by somebody who has not had a hard conversation in six weeks. The test is whether the long horizon shows up as investment in things that compound, or as an absence of anything happening. Compounding requires a principal. If nothing is being deposited, the horizon is not doing any work.

Chapter 22Your network compounds, and it cannot be built late

Of everything in this book, the asset with the longest lead time and the highest eventual value is the set of people who would take your call.

I want to be specific about why, because "network" is one of those words that has been used until it means nothing.

Concretely, the network does four things. It brings you opportunities before they are public, which is where nearly all the good ones are. It gives you an early warning system, because you hear that a regulator's posture is shifting or that a category is dying from people inside it, weeks or months before it is written up anywhere. It gives you access to judgment you do not have, so you can make a decision in a domain you are weak in by making one call. And it is the primary source of clients, forever, in every business I have run.

The mechanics of building it are unglamorous and they work.

Give first, consistently, without a ledger. Make the introduction. Send the document. Answer the question from the person who will never hire you. Do it for years. The people who do this end up with networks that function as force multipliers, and the people who show up only when they need something end up wondering why their network is inert. The difference is visible to everyone except the person doing it.

Be specific about what you can offer, because vague generosity produces nothing. "Let me know if I can help" is a phrase that has never once caused anything to happen. "I know two people running exactly that problem, want an introduction to either" causes something to happen every time.

Follow up when you said you would. This is the lowest effort and highest signal behaviour available to you. Most people do not. Doing what you said you would do within the timeframe you said it puts you in a small group, and people notice, and they remember it years later when somebody asks them who to call.

Stay in touch with no agenda. Send the article. Note the milestone. Ask how the thing went. A relationship that only activates when you need something is not a relationship, it is a transaction with a long setup, and everybody can tell the difference.

And earn your place through work rather than talk. This community, more than most I have been in, has a functioning immune system against people who are performing. It is generous with builders and merciless with promoters, and it is unusually good at telling them apart. Contribute something real, publicly, and the doors open. Show up with a personal brand and no substance and you will find the room polite and permanently closed.

The reason this must start now, before you need it, is that the lead time is measured in years and it cannot be compressed with money or effort at the point of need. Somebody who begins building relationships when they need a client is already too late for this year, and probably next. Somebody who has been giving for three years has a network that produces opportunities they did not ask for, which is the whole point.

Which is why the correct time to start is the least convenient one: now, while you do not need anything from anybody, and while every interaction you have is therefore uncontaminated by an ask.

Chapter 23The contrarian discount

Everything worth building in this space was, at the moment somebody started building it, obviously a bad idea to most informed people. That is not a coincidence, it is the mechanism, and understanding it properly is worth more than most strategy frameworks.

The reasoning is simple. If an opportunity is visible to everyone and considered sensible by everyone, it is priced. The excess return, in business as in anything else, comes from being right about something the consensus has not yet accepted. Which means the feeling of doing something that looks foolish is not a warning sign to be overcome. It is the entry condition.

When we started building an exchange in India, everyone said it was impossible. The regulatory position was unclear and getting worse, the banks were hostile, the market did not understand the product, and the infrastructure did not exist. All of that was accurate. It was also exactly why the opportunity was available, because if it had been easy somebody with more capital would have already done it.

I want to be careful here, because there is a stupid version of this idea and it is popular.

Being contrarian is not being different for its own sake. Most consensus is correct. Most of the time, when everybody thinks something is a bad idea, it is a bad idea, and the people who cultivate contrarianism as an identity end up systematically wrong in an expensive way.

The useful version is narrow. It is: think from primary sources, follow the reasoning wherever it goes, and do not discount your conclusion because it is unpopular. Sometimes that lands you with the consensus, and you should be entirely comfortable with that. Sometimes it lands you outside it, and then you have something.

The practical test I use has three questions.

Do I have a reason for this position that does not depend on other people holding it? If the answer is that smart people are doing it, that is not a reason, that is a citation.

Can I state the strongest version of the opposing case, in its own terms, well enough that somebody holding it would agree I had represented them fairly? If not, I do not understand my own position, I have only rehearsed it.

Would I hold this if it stayed unpopular for five years? This is the one that matters, because most contrarian positions are actually bets on a fast vindication, and a fast vindication is exactly what you cannot count on.

The timing element is where people fail even when the analysis is right. The best moment to be contrarian is when it feels worst, which is by construction the moment you are least inclined to act. When the mood is euphoric, be careful. When the mood is that this is finished, move. Everyone knows this and almost nobody does it, because the social cost of being early is paid immediately and the reward is paid later, and humans are not built for that trade.

The one advantage available to an operator here is that you can make it structural instead of emotional. Decide in advance. Write down, while nothing is happening, what you will do if the whole sector is written off again: what you will buy, who you will hire, what you will build, what you will stop. Then when it happens, execute the document rather than the mood. This is the same mechanism as the treasury policy in the volatility chapter, applied to strategy instead of the balance sheet, and it works for the same reason. It moves the decision to a moment when you were thinking clearly.

Chapter 24When you are wrong

Everything in this part has been about building conviction and holding it. This chapter is the counterweight, and without it the rest is dangerous, because the same machinery that lets you hold a position through a drawdown will let you hold a wrong position through everything.

I have been wrong about a lot of things in this industry, and the expensive ones were never the calls where I was uncertain. They were the calls where I was certain and had stopped checking.

The specific failure mode is that conviction is load bearing for your identity as well as your portfolio. Once you have spent three years publicly explaining why something is true, being wrong about it is no longer an update, it is a cost to your standing and your self-description. So the mind does the economical thing: it stops sampling evidence that would produce that cost. This does not feel like avoidance from the inside. It feels like having settled the question.

A few practices that keep this from happening, all of which I use and none of which are comfortable.

Write the disconfirming condition down in advance. Whenever I take a position that matters, the decision journal entry includes a line that starts "I would conclude I was wrong if." It has to be specific and observable. Not "if the thesis breaks." Something like: if the fee market fails to develop over the next two halving cycles while the subsidy falls, my view of the long term security budget was wrong. Writing that sentence at the moment of decision is easy. Writing it later is impossible, because by then you know which way things have gone and you will unconsciously set the bar where it cannot be reached.

Keep somebody around who will tell you. Not a contrarian for its own sake, which is just noise with a different sign. Somebody competent, who understands the domain, and who has demonstrated that they will disagree with you in front of other people. Most founders systematically remove these people over time, because disagreement is friction and the removal happens one hiring decision at a time and never feels like a policy. Notice when your last three meetings contained no disagreement, and treat that as a finding about your organisation rather than a compliment to your judgment.

Separate the thesis from the implementation, and hold them at different confidence levels. I hold the monetary thesis with very high confidence and I hold nearly every specific opinion about how the industry will develop with much lower confidence, and I try to be explicit about which is which when I speak. The people who get into trouble are the ones who apply the confidence of the base layer to their own predictions about companies, timelines and adoption curves. Those are not the same class of claim and they do not deserve the same certainty.

Update publicly when you update. This is the one that costs something and it is the one that pays. If you have written that something is true and you come to believe otherwise, say so, in the same place, with the reasoning. The industry punishes this less than people fear and rewards it more than they expect, because it is rare, and because everyone reading has watched people quietly delete positions instead. I have changed my public view on more than one thing over the years and it has never once cost me a client.

And separate being wrong from being wrong yet. This distinction matters enormously in a field with long timelines, and it can also be used to avoid ever updating, so it needs a rule. The rule is that "early, not wrong" is only available to you if you wrote down a timeline in advance. If you did not, and you are now explaining that your call needs more time, you are not being patient, you are moving the goalposts, and the fact that it feels the same from the inside is exactly why the written timeline exists.

One more thing, on how to handle being wrong in front of a client, because it will happen. Own it immediately, plainly, without a performance of contrition. Say what happened, say what you are doing about it, say what it means for them, and do not attach a paragraph of apology. In my experience a client relationship survives a mistake handled directly far more often than it survives a mistake handled defensively, and a meaningful number of my longest relationships have a moment like that somewhere near the beginning of them.

Part 05Building the company

Chapter 25Regulation is a moat, not a wall

Most entrepreneurs treat regulation as an obstacle: a cost, a delay, a department that says no. I understand the instinct, and I want to argue against it, because in this industry the instinct is expensive.

Here is the thing I learned the hard way, in a courtroom, over years. Regulatory clarity is the precondition for the capital you want. Not an inconvenience on the way to it. The precondition.

Think about who is not yet allocating. Pension funds. Insurance general accounts. Corporate treasuries at conservative companies. Sovereign funds. Endowments. Those pools are enormous and they are not sitting out because the people running them lack imagination. They are sitting out because a fiduciary has a legal duty of prudence, and moving into an asset with an unresolved regulatory status is a personal career decision dressed up as a portfolio decision. The moment the status is resolved, the decision stops being personal, and that is the entire unlock.

You can watch this happen. Every time a regulated wrapper appeared, capital followed, not because anybody changed their mind about the asset but because the form of access changed. The asset was identical the day before.

The business consequence is that the firms which built for compliance before it was required are the ones positioned when the money moves. It is a slower, more expensive path, and it produces a moat that a faster competitor cannot cross quickly, because the moat is made of time. You cannot buy a three year track record of clean examinations. You cannot buy relationships with regulators. You cannot buy an audit history.

Four things to build before you need them.

Institutional grade operations. Written procedures, real segregation of duties, actual reporting, evidence that things happened. Not because it feels good but because at some point somebody will ask you to prove a control was operating on a date eighteen months ago, and the answer is either in a system or it is not.

Professional relationships. Auditors who have done this before, counsel who is genuinely specialised rather than a generalist reading up, banking relationships built before you urgently need one. Building a banking relationship while under stress is close to impossible, and I have watched otherwise good businesses die of exactly that.

Regulatory expertise as a first class internal function. Not a box ticked by an outside firm once a year. Somebody inside who reads the consultation papers, tracks the changes, and can tell you what a shift means for your product before it is a problem.

Products designed for the constraints your buyers actually live under. Reporting formats an institution can consume. Controls a board can approve. Documentation an auditor will accept. This is where most bitcoin native companies lose institutional deals: the product is technically superior and organisationally unusable.

One line about jurisdiction, because there is a version of this conversation that goes wrong. Structuring a business across jurisdictions to take advantage of clearer frameworks, better banking access and more workable licensing is legitimate and ordinary and every serious multinational does it. Structuring a business to escape rules that would otherwise apply to your customers is a different activity with a different ending, and the people who did it are mostly not in the industry any more. Choose your jurisdictions with counsel, comply everywhere you operate, and never make regulatory avoidance the load bearing part of your model.

The mindset shift I would ask you to make is this. Compliance is not a tax on the business. In a market where trust is the actual product, compliance is the product's packaging, and buyers at the size you eventually want cannot purchase without it.

A practical note on licensing, because the question comes up in year one and the wrong answer is expensive in both directions.

Most of what is described in this book requires no licence at all. Advising a company on how to structure a custody arrangement, running a training programme, writing a treasury policy, reviewing security architecture: these are ordinary professional services in nearly every jurisdiction. What triggers registration is generally taking custody of client assets, transmitting value on behalf of others, dealing or arranging deals in regulated instruments, and giving investment advice as defined by your regulator, which is usually narrower than the everyday meaning of the phrase and occasionally much broader.

The two failure modes are symmetrical and both are common. The first is drifting across the line without noticing, usually by being helpful: holding a key for a client as a favour, moving funds on their instruction once because they were travelling, giving a view on whether they should buy. Each of those is a small accommodation and each of them changes what you are. The second failure mode is the opposite, paralysis, where somebody spends a year and a large amount of money on a licensing question that their actual business never triggered.

The fix for both is the same and it costs a few thousand at most. Early, before you have clients, get a written scoping opinion from counsel who is genuinely specialised in your jurisdiction, describing exactly what you intend to do and confirming what it does and does not trigger. Then write your own boundary into your scope documents, in plain words, so that clients can see it: this engagement covers design and documentation, it does not include holding keys, executing transactions, or advice on whether to acquire the asset.

Saying that out loud is not a limitation on your business. It is one of the strongest trust signals available to you, because the buyer has met the other kind of consultant, and a person who volunteers the edge of their own competence is the person they will call again.

Chapter 26The impossibility test

There is one question I use to evaluate every bitcoin business idea, mine and other people's, and it has saved me more time than any framework I know.

If bitcoin disappeared tomorrow, would this business still make sense?

If yes, you are building a normal business with a bitcoin feature. That is allowed and it can work, but understand what you have: your moat is whatever it would have been anyway, and the bitcoin part is marketing, which competitors can copy in a quarter.

If no, you are building something bitcoin native, and the properties of the base layer become part of your defensibility.

The distinction is not about how much bitcoin is involved. It is about whether the value proposition depends on a property that only exists here.

Consider what those properties actually are, because this is where the ideas come from.

Absolute scarcity with independent verification. Not "there is a limited amount," which many things have, but "there is a limited amount and any participant can check without trusting anyone." That enables business models built on provable reserves, on collateral with no counterparty, on obligations that can be publicly verified rather than attested.

Final settlement without an intermediary. Once buried, done. No reversal, no chargeback, no institution with the ability to unwind it. This changes what a contract can assume, and it enables businesses that could not previously bear the cost of settlement risk.

Bearer property with cryptographic control. The asset can be held directly by a person or a quorum of people, with no registrar. That creates an entire category of problems that did not exist before 2009: quorum design, inheritance, incapacity, corporate governance over keys. Every one of those problems is a business, and most of them are underserved.

Global, permissionless, always on. A settlement network that does not close for weekends, does not require a relationship with a correspondent bank, and does not ask your nationality. That is a genuine structural difference from every rail that preceded it, and the businesses built on it are not faster versions of existing ones, they are things the existing rails cannot do at all.

Programmability at the payment layer. Conditional, streaming and automated payments at sizes below what any card network can process economically. This is the least mature of the properties and the one I would be most careful about, because the gap between what is demonstrated and what is in production is still wide. It is real, and it is not yet where the enthusiasm suggests.

Now the discipline. Having a list of properties tempts people to invent a business for a property rather than find a problem. That is backwards and it produces the specific failure I see most often in this industry: a technically elegant product that nobody wanted.

So run it in the other direction. Start with a problem that costs somebody real money today. Ask whether any bitcoin property makes it materially better, not marginally. Then apply the impossibility test to confirm you are building something defensible rather than a feature. Three filters, in that order, and most ideas die at the second one, which is the point.

The strongest examples in this space all came from the same place, which is that somebody had the problem personally. Collateralised lending against a holding came from people who did not want to sell. Collaborative custody came from people who had watched somebody lose everything to a single point of failure. Institutional custody came from people who could not get an allocator over the line without it. None of it came from a whiteboard.

Chapter 27Treasury: holding the asset on your own balance sheet

At some point, if you believe what you are telling clients, you will face the question of whether your own company should hold bitcoin. I want to treat this properly, because it is where a lot of otherwise sensible operators get themselves in trouble.

The case for it is straightforward and I find it convincing. Corporate cash held in a currency with an expanding supply loses purchasing power at a rate that is not visible on the income statement, because the unit of measurement is the thing that is moving. Over a five or ten year horizon that erosion is material. Holding a portion of long horizon reserves in an asset with a fixed supply is a defensible response to a real problem, and it has moved from eccentric to established practice.

The case against doing it badly is much longer, and it is the part worth your attention.

Start with the sequencing. Do not put the asset on the balance sheet until the operating business generates enough cash that you are not depending on the position. A company that needs its holding to go up in order to make payroll has not made a treasury decision, it has made a bet with other people's employment attached to it. That is the single most common way this goes wrong.

Define the operating reserve first and hold it in the local unit. Payroll, rent, tax, vendor obligations, for a defined number of months. That number is a policy decision, and I would not go below twelve months for a business with employees. Only what sits above it is available.

Write the policy before you buy anything, and get it approved by whoever governs the company. It should say what percentage of reserves may be held in the asset, who is authorised to transact, what the custody arrangement is, what triggers a sale, and what happens if the position falls by half. The purpose of the document is not compliance theatre. It is to move the decision to a moment when you were calm, because the decisions you will otherwise make are made in a week when you are not.

Understand the accounting before you act, not after. Know how the holding is measured in your jurisdiction, how it will appear in your statements, what it does to any covenant you are subject to, and what your auditor will need to see. If you have a lender, tell them before rather than after. If you have a board, the memo comes before the purchase.

Get custody right, and treat it as the hardest part rather than an afterthought. For a company holding meaningful reserves this means multiple signatures held by different people in different places, with at least one key outside the organisation, with a written and tested recovery procedure, and with a documented answer to what happens if any single person is unavailable permanently. If your company's holding depends on one person and one device, you do not have a treasury, you have an incident waiting for a date.

And communicate before, not after. Employees, investors, lenders and auditors should learn about this from you, in advance, with the reasoning attached. Every reputational problem I have seen around a corporate holding came from a stakeholder finding out from a filing.

One more thing, on the version of this that has become a whole category. There are companies whose primary activity is now acquiring the asset using capital markets instruments, and whose equity trades as a leveraged expression of it. Whatever you think of that as an investment, understand it as a business model: it depends on continued access to capital markets on favourable terms, which is a condition, not a constant. If you are considering anything in that direction, model what happens when that access closes, because it has closed before for every category of issuer that has ever existed.

Chapter 28Security is the whole business

In most industries a security incident is a bad quarter. Here it is often the end, because the transactions are final and there is no institution positioned to reverse them. I want to be direct about this: if you handle other people's bitcoin, or your own at any scale, security is not a department. It is the business, and everything else is packaging.

The threat model is broader than people assume, and it is worth naming the categories because defences tend to be built for only one of them.

Technical attacks on your software and infrastructure. This is the one everyone thinks of and it is not usually how people lose money.

Social engineering. This is how people lose money. A convincing message, an urgent request from someone who appears to be an executive, a support call that arrives at exactly the right moment. Your controls have to assume that a competent attacker will successfully impersonate somebody your team trusts, because eventually one will.

Physical attacks, on hardware, facilities and people. This is real at the top end and it is why geographic distribution of keys is not paranoia. It is also why nobody in your organisation should be publicly identifiable as the person who controls the assets.

Insider risk, which is the one nobody wants to design for because designing for it feels like an accusation. Build the controls so that no single person can move funds, and then say plainly to your team that the controls exist to protect them as much as the company, because they do. A person who cannot unilaterally move funds cannot be coerced into moving them.

Now the controls that actually matter, in rough order of how much loss they prevent per unit of effort.

Multiple signatures for anything above a trivial threshold, with the keys held by different people, stored in different physical locations, on different device types. Every serious loss I know of traces back to a single point of failure that somebody knew about and had not got around to fixing.

A recovery procedure that has been tested by executing it, not by reading it. Written, followed end to end, with the key holder who is meant to be unavailable actually not participating. Untested backups are not backups, they are an assumption, and I would guess more than half of the recovery procedures in this industry have never been run.

Documented succession and incapacity. What happens if a signer dies, resigns badly, or becomes unreachable. Who is contacted, in what order, with what authority. For family offices and closely held companies this is frequently the largest actual risk and it is almost never written down.

Verification out of band for anything that moves value. Any instruction to transact, however it arrives, gets confirmed through a different channel with a person the confirmer knows by voice. This one control defeats the majority of social engineering attempts and costs almost nothing.

Access control and least privilege everywhere, reviewed on a schedule rather than when somebody remembers.

Training that is specific rather than general. Not an annual video. A quarterly session where you show the team the actual message that nearly worked on somebody in your industry last month.

And an incident plan written in advance: who is called, in what order, what is said publicly, who decides. Incidents are chaotic and the plan is what stops the chaos from producing the second, larger mistake.

If you are advising clients rather than operating, security is the lane I would push you toward hardest, for a slightly cynical reason. It is chronically underbought until something frightens the buyer, which means the sales problem is real, and the way to solve it is not persuasion. It is the drill. Sit a client down and ask them to recover from a simulated loss of one key, live, with you watching. Roughly two thirds cannot. That experience sells the engagement, and more importantly it fixes the thing that would have destroyed them.

Chapter 29The team, and the curriculum you have to build yourself

At some point you stop being a practice and start being a company, and the constraint changes from your calendar to your ability to create people who can do the work.

The talent market here is not one you can win by competing on salary. The people with deep experience are scarce, expensive, and mostly already committed to something they believe in. So the strategy that works is not recruitment. It is manufacture: hire capable, motivated people and build them into experts internally.

That means you need an actual curriculum, not a reading list. There is a difference and it is the difference between people who have opinions and people who can be trusted with a client.

Here is the shape that has worked for me.

The first thirty days are the same for everyone, regardless of role. Engineers, marketers, operations, support, all of them. The purpose is a shared vocabulary, because most organisational failures in this field are two people using one word for two things.

That month contains required reading, which is short and non-negotiable: the whitepaper, one serious book on the monetary case, and enough monetary history to understand what preceded 1971. It contains required practical work, which is the part that matters: every new hire sets up a node, creates a wallet, sends and receives a real transaction with real fees, participates in building a multisignature arrangement, and completes a recovery. Not a demonstration they watched. Work they did with their hands.

And it ends with an assessment, which people resist and which is the reason the programme works. Each new hire presents a topic to the team and takes questions, and writes a short reflection on what they got wrong on the way in. Presenting to peers is a much harder test than a quiz, and it surfaces the shallow understanding that a quiz lets people hide.

Days thirty one to ninety go role specific. Engineers into protocol detail, key management, and the failure modes of the libraries they will use. Client facing people into objections, the regulatory surface, and the specific vocabulary of the vertical you sell into. Operations into procedures, controls and incident response. Pair everyone with somebody more experienced and give the pair a real piece of work rather than an exercise.

Then continuous, forever, because the field moves. A weekly session where somebody presents what actually mattered this week and why, which trains the far more valuable skill of separating signal from noise. A budget for conferences and outside training that is real rather than nominal. And a quarterly session where you revisit the company's own thesis and ask what you now believe that you did not three months ago.

On compensation, one thing I feel strongly about. Offering people the option to take part of their compensation in the asset, if they want it and if the legal and tax position in their jurisdiction supports it, does something that no equity grant achieves. It aligns them with the thesis of the business in a way they experience directly. Some will take it, most will not at first, and after a year or two more will. Do it as an option, never as a default, never as a substitute for competitive cash, and get it right with counsel and an accountant before you offer it, because the tax treatment varies enormously and getting it wrong hands your team a problem instead of an alignment.

On culture, the shortest version I can give. This industry has a functioning immune system, and it will detect the difference between a company that believes what it says and one that has adopted it as positioning. Every person you hire either strengthens or dilutes that, and the dilution is not recoverable. I would rather run understaffed for a quarter than hire somebody who thinks this is a sector rotation.

Chapter 30Capital, and how to think about funding what you build

Most of what you have read about raising money was written for software companies with predictable metrics and a well understood category. Some of it transfers. A meaningful amount does not, and I want to give you the honest version rather than the encouraging one.

Start with the question people skip: should you raise at all.

If you are building a service business, the answer is usually no. Services fund themselves from client revenue, and outside capital in a service business mostly buys you an obligation to grow faster than the work allows, which degrades the work. I have never once wished a consulting practice had taken money.

If you are building a product with real infrastructure requirements, custody, regulated operations or a licensing path, the answer is probably yes, because those things cost money before they produce any, and you cannot bootstrap a licence.

If the honest answer is that you want validation, do not raise. You will be buying a permanent obligation with a temporary feeling.

Assuming you should raise, a few things about this market specifically.

The investors who understand the category are a small, identifiable group, and they are selective in a particular way: they are testing your conviction as much as your metrics, because they have watched people arrive for a cycle and leave. Generalist investors who have developed a thesis are a larger group and a slower process, and they will need you to make the category legible before they can evaluate you within it. Strategic investors, meaning operating companies that want exposure or capability, can be excellent and come with commitments you should read carefully.

What they will actually diligence, beyond the usual, is security architecture and key management, which is unique to this sector and where inexperienced founders get exposed; regulatory position across every jurisdiction you touch; and whether you personally understand what you are building at a level below the pitch. That last one is tested by a single unscripted technical question, and it is the reason chapter four exists.

I am going to be explicit about a few boundaries, because this is the area where people cause themselves lasting damage.

Do not issue a token. I am not going to hedge this. If your business model requires creating and selling a new asset to fund itself, you have a securities problem, a focus problem, and an alignment problem, and the historical record on this is not ambiguous. Raise equity or raise against revenue.

If you engage anyone to help with a raise, the compensation is a flat fee for work performed, never a percentage of the amount raised and never contingent on the financing closing. Success based compensation for introducing investors is broker activity in most jurisdictions and requires a registration that most service providers do not have. This is not a preference. Both sides carry exposure, and the fact that it is common does not make it safe.

You own every investor relationship, always. Whatever technology or service sits underneath your process, the research, the list, the sending infrastructure, the materials, the relationship belongs to you and every conversation is yours to run. Anyone positioned between you and an investor is a structure you should decline.

Nobody can promise you a raise, a term sheet, or a number of meetings, and anyone who does is telling you something about themselves. What can be committed to is work: the quality of the research, the materials, the volume of outreach you personally send, the responsiveness, the reporting. Judge providers on the work, because the work is the only part anyone controls.

And treat your own runway as an operating input rather than a background fact. A founder with six months of runway negotiates differently from a founder with three weeks, and the difference is not character, it is arithmetic that the person across the table can hear. The best thing you can do for the terms of your next raise is to need it less, and the only way to need it less is revenue.

Hold those four boundaries and you will be a rare thing in this market: a provider whose incentives a sophisticated buyer can read off the contract in thirty seconds and find nothing hidden in.

Part 06The build

Chapter 31Days 1 to 30: become someone who can charge

The first thirty days are not about revenue. They are about becoming a person who can credibly ask for money, and if you compress this phase you will spend the next six months being found out.

I am going to give this to you as a sequence with dates, because in my experience the people who succeed at this are not the ones with better ideas, they are the ones who did the boring thing on the day it was scheduled.

Week one is about commitment and raw input.

On day one, tell somebody. Not a social media announcement necessarily, though that works: tell three people whose opinion of you matters that you are building something specific in this space and that you will have your first paying client within ninety days. The mechanism here is not motivational. It is that a stated commitment to a specific person on a specific timeline changes what you do on day nineteen when you do not feel like it.

Days two and three, read. The whitepaper, slowly, twice. One serious treatment of the monetary case. Take notes by hand, because typing lets you transcribe without understanding and handwriting does not. Aim for twenty pages of notes in your own words, and if that sounds excessive, that is the point.

Days four and five, go where the problems are. Sit in the places your future buyers complain: professional forums in your vertical, industry discussion groups, the question sections of relevant events, wherever practitioners talk to each other. Do not participate yet. Build a list of thirty specific problems, written in the words the person used. This document will be the most valuable thing you produce this month.

Days six and seven, audit yourself honestly. Write down every industry you have worked in, every process you understand from the inside, every category of person whose calls you would get returned, and every thing you know how to do that took you years. Then cross it against the problem list. You are looking for intersections: problems where your existing background gives you a genuine advantage over a generic entrant. Write one page: the problem, the buyer, and why you specifically.

Week two is positioning, and it is where most people rush and pay for it later.

Days eight to ten, take the top three intersections and research them properly. For each: who exactly has this problem, roughly how many of them are reachable by you, what they currently do instead, what it costs them to keep doing that, and who else is already selling into it. One page each. If you cannot fill a page, you do not understand the problem yet, which is itself the finding.

Days eleven and twelve, choose. One lane, one vertical, for the next seventy eight days. Write it as a single sentence and put it where you will see it. The commitment is the point. You can be wrong about the choice and recover. You cannot recover from spending the quarter choosing.

Days thirteen and fourteen, study the people already selling into your chosen position. What they offer, how they price, how they describe themselves, where they publish, and specifically where they are weak. Note that "there is nobody" is usually wrong and means you have not looked hard enough, and "there are many" is usually also wrong and means you have counted people adjacent to your position rather than in it.

Week three is the foundation of your evidence.

Days fifteen to seventeen, write your cornerstone. Two to four thousand words on the most important unresolved question in your position. This will take longer than you expect and it will be the single highest return three days of the ninety. Publish it somewhere you control.

Days eighteen and nineteen, build the smallest possible home. Five pages: what you do, who for, what it costs, who you are, how to reach you. Do not spend a week on design. Nobody has ever declined to hire a specialist because the typography was ordinary, and I have watched people spend a month on a website during a quarter when their real problem was that they had spoken to nobody.

Days twenty and twenty one, make one useful thing you can give away. A checklist, a template, a short assessment. It should be genuinely useful on its own, which means somebody could use it and never contact you and still be better off. That is the standard.

Week four is your first offer.

Days twenty two to twenty four, design the minimum viable offer from chapter ten. One problem, one artifact, fixed price, fixed timeline, under thirty hours. Write the scope document as though you were sending it to a client tomorrow, because you are.

Days twenty five to twenty seven, find your first five prospects. Not a list of five hundred. Five specific named people you can reach, who have the problem, who could say yes. Contact each one personally, referencing something true and specific about their situation, offering the offer at a stated early rate in exchange for candid feedback and, if the work is good, permission to describe it.

Days twenty eight to thirty, review. Read the month back. You have a position, a substantial piece of published work, a home, a giveaway, a defined offer and five live conversations. That is more than most people in this field accomplish in a year, and you are not close to done.

Chapter 32Days 31 to 60: get proof

The second month is about converting a plausible position into demonstrated competence, and the currency is completed work.

Week five, multiply what you have.

Days thirty one to thirty three, cut the cornerstone apart as described in chapter thirteen. The core argument, the mechanism diagram, the objections series, the talk outline, the checklist. Schedule the pieces out across the next six weeks so you are not producing under pressure while delivering. Batch the making, spread the publishing.

Days thirty four and thirty five, install a daily contribution habit and hold it for the rest of the ninety days. My version is three things a day: answer one real question in a place where your buyers are, respond substantively to one person doing interesting work, and send one message to one specific new person. It takes about forty minutes and it is the engine underneath everything else. The temptation, once delivery work arrives, is to drop it. Do not drop it. This is precisely the mechanism by which people produce a good month followed by a dead quarter.

Week six, deliver, and over deliver deliberately.

Days thirty six to forty, do the work for your first clients and do it visibly better than what they paid for. Finish early. Include the thing you noticed that was not in scope. Write the summary they can forward to their board without editing it. This is not generosity, it is the cheapest marketing available to you, and the reason is that the person who received it will describe the experience to somebody else in specific terms.

Days forty one and forty two, ask for the account of it. Do this while the work is fresh, because in three weeks they will remember that it was good and not why. Ask specific questions: what was the situation before, what was the concern going in, what changed, what would you tell somebody considering this. Get it in writing. Ask explicitly for permission to use it, and be precise about where and how, and if the answer is no, respect it completely and permanently. An anonymised account of a real result, used with permission, is worth more than any claim you can make about yourself.

Week seven, go to full price.

Days forty three to forty five, prepare properly. Write the offer page, the outreach message, and the scope document at full price. Set the number at roughly twice your early rate. Have the payment mechanism working before you need it, because the worst possible moment to discover an invoicing problem is the moment somebody has said yes.

Days forty six to forty eight, launch it. Tell everyone who has been reading you. Contact everyone who expressed interest and did not buy. Reach the next twenty five names on your list. Do not be coy about it. You built something, it works, people should know it exists.

Days forty nine and fifty, handle the first full price outcome, whichever way it goes. If somebody buys at full price without a prior relationship, that is the moment the business becomes real and you should notice it. If nobody does, go back to the people who looked and did not buy and ask them, directly and without defensiveness, what stopped them. That conversation is worth more than the sale would have been, because it tells you whether the problem is the price, the scope, the proof, or the fit, and those have four different fixes.

Week eight, start writing things down.

Days fifty one to fifty three, document your delivery. Onboarding, the questions you ask, the sequence you follow, the artifacts you produce, the way you close an engagement. Not for a future team. For you, next week, so that quality does not depend on your energy level on a given Tuesday.

Days fifty four to fifty six, build the feedback loop. Ask every client the same three questions at the end: what was most valuable, what was missing, what would you have paid more for. The third one reshapes your offer faster than any amount of thinking about it.

Days fifty seven to sixty, look at the numbers. Conversations started, conversations converted, engagements delivered, revenue, effective rate including unbilled hours. Do the arithmetic honestly. This is the first month where you have real data rather than plans, and the data will contradict at least one thing you believed at day one.

Chapter 33Days 61 to 90: build the machine

The third month is where you stop being a person doing projects and start being a business with a process, and the difference is almost entirely whether the next client is a surprise or a consequence.

Week nine, make revenue repeatable.

Days sixty one to sixty three, write down your actual system end to end. Where prospects come from, how they become conversations, how conversations become engagements, how engagements are delivered, how clients are retained or referred. Write it as it currently is, including the parts that are luck. Then mark every step that depends on you being motivated that day, because those are the steps that will break first.

Days sixty four to sixty six, test the price again. You have delivered work now and you have accounts of results. Quote the next prospect twenty five percent higher than the last. Say the number flat and stop talking. The most common outcome is that nothing happens, which is information: your price was under the market and you have been paying for it.

Days sixty seven to sixty nine, build the referral mechanism, which is the highest margin channel that exists and the one most people leave entirely to chance. At the end of every engagement, ask a specific question rather than a general one. Not "let me know if you know anyone." Instead: "who else in your position has this same problem." A specific question gets a specific name. Make it a step in your documented close, so it happens whether or not you feel like asking.

Week ten, get on programmes and into rooms.

Days seventy to seventy two, pitch three speaking opportunities. One local meetup, one industry event in your vertical, one podcast. Use the specific pitch format from chapter fourteen: a title, a description, a date. Expect one yes from three, which is a fine rate and the reason you send three.

Days seventy three to seventy five, approach two publications or newsletters your buyers actually read with a specific piece, not a general offer to contribute. The piece should be derived from your cornerstone, which means it already exists and you are reshaping rather than writing.

Days seventy six to seventy eight, sketch the larger work. Not to publish now. To structure the next two years of cornerstones so that they accumulate into something rather than scattering. Four questions a year, sixteen over four years, and at some point that is a book and the people who did not start cannot catch up.

Week eleven, add the second rung.

Days seventy nine to eighty one, design the offer above your current one. If you have been selling assessments, design the implementation. If you have been selling education, design the policy work. You already know what the next problem is, because your clients told you when you did the first engagement.

Days eighty two to eighty four, design a one to many version of the thing you have now delivered several times. A workshop, a cohort session, a structured programme. The test for whether you are ready is whether you have delivered the one to one version at least four times, because before that you are packaging a guess.

Days eighty five to eighty seven, define the retainer. Fixed monthly hours, a standing call, priority access, ongoing monitoring of whatever moves in their world. Offer it to every client who has completed a project. The conversion rate on this is far higher than any new business activity you will do, and it is the thing that turns lumpy income into a floor.

Week twelve, look up.

Days eighty eight to ninety, write down what you want this to be in ten years. Not a plan, a direction. Then look at everything you built in ninety days and ask which parts point at it and which parts are drift. Cut the drift now, while it is small.

Then look at what actually happened. Ninety days ago you had a position you had not chosen, no published work, no offer, no clients and no process. If you did the sequence you now have all five. That is not a transformation, it is a foundation, and the difference between people who compound from here and people who do not is entirely in chapter thirty.

Chapter 34The weekly loop

Everything in this book fails without one thing, and it is the least interesting thing in it: a repeating weekly cycle that you run whether or not the week went well.

I have run some version of this for over a decade across several businesses, and every period of drift I have had traces to a stretch where I stopped running it. Not to a bad decision. To an absence of the review that would have caught the bad decision in week two instead of week nine.

It takes about forty five minutes and it happens on the same day every week.

First, count. Five numbers, written down, in the same place every week so you can see the series rather than the point.

Conversations started this week, meaning genuine two way exchanges with somebody who could plausibly buy. This is the leading indicator and it is the only one that predicts anything. When it goes to zero, revenue goes soft about eight to twelve weeks later, and the delay is what makes it dangerous, because by the time you feel it the cause is two months behind you.

Proposals or scopes sent. This is the conversion of interest into a decision point, and a gap between high conversations and low proposals means you are having pleasant chats rather than asking for the business.

Revenue booked, which is different from revenue collected and both are worth tracking separately.

Hours worked, all of them, billable or not. You need this to compute your effective rate, and the effective rate is the number that tells you whether your packaging is working.

Something published or shipped. One thing. If this is zero for three consecutive weeks, your evidence base has stopped growing and your pipeline will follow it.

Second, ask three questions and write the answers down.

What worked this week that I should do more of. Specific, not general.

What did I avoid this week. This is the one that produces the value, because the thing you avoided is almost always the thing that would have moved the business: the price conversation, the follow up on the deal that went quiet, the client who is unhappy, the piece of work that requires real thought. Avoidance is the actual bottleneck in most businesses and it is invisible unless you name it weekly.

What did I learn that changes something. Not what did I read. What changed.

Third, set the next week. Three outcomes, not a task list. The three things that, if they happen, make the week a success regardless of what else does. Then put them in the calendar as time blocks, because an outcome without a block is a wish.

That is the whole loop and it is deliberately short, because a review process that takes two hours does not survive a busy quarter, and a review process that does not survive a busy quarter is worthless, since busy quarters are exactly when drift happens.

Two additions on longer cycles.

Monthly, read your decision journal entries from three months ago. You will find that you remember your reasoning inaccurately, in a direction that flatters you, every single time. This is the cheapest calibration available.

Quarterly, ask the three questions from chapter eighteen. What compounded. What will not matter in five years. What did I avoid because it was uncomfortable. And once a year, ask whether the position you chose is still the right one, which is a question you should be genuinely willing to answer with no, and almost nobody is.

Chapter 35The shelf

I am going to keep this short and opinionated, because a resource list that includes everything is the same as no list at all.

For the monetary case, read the whitepaper first, three times, spaced out. It is nine pages and almost nobody who has an opinion about bitcoin has read it properly. Then read a serious treatment of the case for hard money, and then read the strongest available argument against your position, deliberately, because you will meet it in a room and you should not meet it there first.

For monetary history, go older than the industry. The history of money as a market phenomenon, the classical arguments about the denationalisation of money, the anthropological work on debt and credit, and a properly researched account of the twentieth century monetary arrangements and their collapse. This is the part people skip and it is the part that produces conviction rather than enthusiasm.

For the technical layer, there is one canonical technical treatment of how bitcoin works under the hood, and it is worth reading even if you never write code, because it will let you follow an engineering conversation. Supplement it with the protocol's own improvement proposals, which are public, and with the technical newsletters produced by the development community, which are the fastest way to know what is actually being worked on rather than what is being discussed.

For primary sources on regulation, read the actual documents. Consultation papers, official guidance, published enforcement actions and legislative text in your jurisdiction. It is slower than commentary and it is the difference between knowing what happened and knowing what somebody concluded about what happened. If you build a habit of reading primary regulatory sources, you will routinely know things before the people commenting on them.

For practical work, the tools you need are all free or cheap: a full node implementation, a hardware signing device or two, a wallet that supports multiple signatures, and a test network. Budget a few hundred for devices and a weekend for setup. This is the single highest return spend in the entire book.

For staying current, I would keep the daily input small and the weekly input substantial. Thirty minutes a day of primary sources and industry filings beats three hours a day of commentary, and the compounding difference over a year is enormous. Deliberately exclude price commentary. It is the highest volume category of content in this industry and it is close to useless for anybody building something.

A word on what is not on this shelf. There is no course you need. There is no certification that a serious buyer will care about, and I say that as somebody who has hired for these roles. The credential in this field is demonstrated work, and there is no shortcut around producing some.

And there is nothing here about other digital assets, deliberately. The thesis of this book rests on a specific property, absolute scarcity independently verifiable by anyone, and the number of things that have it is one. Spreading your attention removes the only advantage a specialist has.

Chapter 36What to do Monday

I want to end without a summary, because summaries let people feel finished, and the failure mode of a book like this is a reader who enjoyed it.

So here is the honest position. Everything in these pages is either something I have done or something I have watched closely enough to be confident about. None of it is complicated. The frameworks are simple, the arithmetic is arithmetic, and the ninety day sequence is a sequence of ordinary days. The reason most people who read this will not do it is not that any step is hard. It is that step one has to happen on a specific Monday and nobody is checking.

So make Monday small and specific.

Do not choose your position on Monday. Do not build a website, do not design an offer, do not plan a content calendar. Those are week two problems and treating them as day one problems is the most common way this stalls, because they are all things you can do alone in a room, and things you can do alone in a room feel like progress and produce nothing.

On Monday, do two things.

Write the list of thirty problems, in the words the people having them actually used. Go where those people talk and take it down verbatim. Two hours.

Then contact one person. One. Somebody who plausibly has one of those problems, who you can reach, with a short message that says something true and specific about their situation and asks a question you genuinely want the answer to. Not a pitch. A question.

That is Monday. Everything else in this book is downstream of doing those two things and then doing the second one again on Tuesday.

I want to say one last thing about why I think this work is worth doing, beyond the money, because the money is a reason but it is not a sufficient one and you will need something else at month seven.

The monetary arrangement most of the world lives under is about fifty years old. It was adopted as a temporary measure, was never put to anyone for approval, and has a structural property that most people have never had explained to them: the units can be created, and the people who receive them first do systematically better than the people who receive them last. Most of the political anger of the last two decades is downstream of that mechanism, and almost none of the people who are angry have been given the vocabulary to name it.

Bitcoin is the first credible alternative that does not require anyone's permission and cannot be diluted by anyone's decision. Whether it ends up as the base layer of a new arrangement or as a permanent check on the old one, the transition needs people who can build the boring, necessary things: the custody, the compliance, the education, the infrastructure, the businesses that make it usable by ordinary institutions run by ordinary people who have no interest in the ideology.

That is the work. It is unglamorous and it is genuinely important, and there are still not enough people doing it well.

Ninety days from Monday you can be one of them.