the question is the skill

a data lead who built analytics teams from zero told me ai is commoditizing the dashboard. what stays valuable is knowing which question to ask.

share x linkedin facebook

a lone figure holding a lantern stands on a path facing a pale purple city of towers, the lantern light reaching low across empty ground

friday afternoon i had a guest on the show who built data teams from zero at two big companies in the digital asset world. now he teaches people how to read raw onchain data, the public ledger that sits under every transaction.

i expected a technical conversation. i got one of the clearest takes on ai and careers i've heard all month.

his first point was about vanity metrics. i asked him what question he wished more teams were asking, and he went straight there. teams love the big totals, like the value locked in a protocol or the trading volume, because those numbers look great in a deck. he said they can hide a lot. a handful of wallets can be most of the total. volume can be the same money going back and forth. the signals he trusts are the boring ones, like whether users come back and how many leave.

that applies to your business too, whatever you build. the number that makes you feel good is seldom the number that tells you what's going on. revenue can look healthy while one customer slowly turns into half of it. signups can climb while nobody returns in week two.

his second point is the one i want you to sit with.

he's watching ai and agents take over basic dashboard work. you can describe a chart in plain english and get it in seconds. so he's pushing his students away from the parts that are getting automated and toward the harder layers, the data engineering underneath. and he said the durable skill is knowing the business well enough to ask the right question, because every analysis is about the business in the end.

i've seen this in my own work. ai will write the query, build the chart, summarize the result. it won't tell you which question is worth an afternoon. that part still comes from someone who knows the customer.

the third thing he said was practical. when the two biggest free places to learn his craft shut down or put up a paywall this year, his first reaction was worry, because that's how he learned too. then he saw the gap. a whole generation of analysts had lost their practice ground. so he built one, with a free tier and a new question every week.

that's a founder move. the market removed something people needed, and he showed up with a replacement before anyone else did.

his advice to new analysts translates to anyone trying to break into a field right now. pick one niche and one chain. build in public. take small freelance gigs. the teams that see your work start to think of you as the person for that problem, and the gigs turn into jobs.

long term, i think this is the shape of most knowledge work in the ai era. the tool does the producing and the human does the asking, so anyone who can frame a good question gets more expensive every year.

if you want more on where i think ai pays off, it's in the ai book.

so what's the vanity number in your business... and what's the boring one you've been avoiding?

share x linkedin facebook
the newsletter

the machine economy brief

one email when it matters: bitcoin, ai, robotics, and what founders should do about it. unsubscribe anytime.

no spam, no list-selling. your email goes to sunny, nowhere else.

this is the thesis i'm building on.

i write the machine economy brief: bitcoin x ai x robotics, where it's all going, once a week.

join the machine economy brief →