how to grow a live stream audience from zero when every episode starts over

a few days ago charles and i sat down and did the thing most creators avoid. we looked at the numbers without flinching.

youtube was growing about one or two subscribers a day. twitter was a trickle.

that's not nothing. but it's not a media company either. and if you're trying to grow a live stream audience from zero, you know the feeling. you go live, you wait, and the counter says three. one of them is you on your phone.

so we asked what had worked, even a little.

the first thing was live tweeting the questions while the interview was happening. one question at a time, in real time, so people could see the conversation and jump in. by the end of a thirty minute show that pulled thirty to fifty people. on longer sessions it got up to two hundred.

the second thing was bigger and simpler. one guest tweeted the interview out himself. "it was because the guy tweeted it out, and he had seventy thousand followers."

that was the spike. not a better thumbnail. not a better title. a guest with an audience who decided to share.

and that's when it hit me. every weekly show has a hidden problem. every episode starts at zero.

you build a crowd for thirty minutes. then the stream ends and the crowd goes home. next week you start from nothing again. the audience never compounds, it resets.

so we landed on two levers.

lever one... the guest's audience is the real distribution. your audience is small. theirs might not be. so make it easy and natural for them to bring their people. push guests to stream the conversation on their own channels at the same time. give them the clips. give them the questions to tweet. the show becomes a co-production, not a favour they did you.

lever two... stop ending the stream.

there's a show called tbpn that runs like a live tv channel for tech. it doesn't reset because it doesn't stop. so we sketched our own version. charles and i alternating, guest after guest, the stream just keeps rolling. if thirty people show up in the first half hour and some of them stay, by hour two we figured two hundred plus was realistic. later you hand the mic to co-hosts in other time zones and it runs around the clock.

the audience compounds instead of evaporating.

now the honest part. both of us were running at about ninety eight percent bandwidth. a twenty four hour stream sounds great on a whiteboard and brutal on a calendar.

so we made a smaller promise first. we agreed to stop canceling the weekly shows.

that's the unsexy truth about growth from zero. the clever lever only works if the boring thing is already happening every week. consistency is the floor. the guest's audience and the always-on stream are what you build on top of it.

if you're starting from zero right now, don't wait for the always-on version. do the two cheap things this week. tweet the questions while you're live, so the conversation has a front door people can walk through. and before every taping, ask your guest one simple thing... will you share this with your people when it's done? most will say yes. most were never asked.

long term i think every founder ends up running some kind of show, whether they call it that or not. the ones who win won't be the best talkers. they'll be the ones who stopped starting from zero.

and if you've got an audience of your own, that makes you the lever for someone else. come on the show and bring your people with you.

so when your next stream ends... where does the crowd go?