GTM Zero To One: Focus 100% on Testing One Thing To Get Signal
Powerful insights from TBH and Gas founder Nikita Bier ⛽️🔥
Day 41 of writing.
The thing you need to be good at is testing and learning and making it really quick.
~ Nikita Bier
I just listened to this interview and knew I HAD to write about — thanks to the YouTube algorithm for serving me up the perfect video at the perfect moment (even if the pod on Lenny’s newsletter was released late August, and ‘1 month ago’ on YouTube).
Side note: There was this a tongue-in-cheek moment from a recent interview indie.vc’s Bryce Roberts did with Michael Dempsey, where he states that there are only three founder archetypes at the moment — Sams, Delians and Nikitas.
In one of my favorite frameworks, and to dramatically simplify:
Lenny Rachitsky did an in-depth interview with Nikita Bier who (in tech circles, famously) built and sold TBH to Facebook as an ‘overnight success’ (although, not surprisingly as we all know, overnight successes take a lot to get there, as he tried 15 other consumer mobile apps, every type you can think of according to him —> and this was after originally building something that was selling B2G), and then ‘rebuilt’ it again as Gas, which sold to Discord.
Well, I’ve had respect for and caught parts of the TBH and Gas story over time (including the story about its geofenced release and approach with high schools, of following students who then follow back —> one little nugget from this pod was they launched TBH and Gas in the same Georgia high school, on the same day, 5 years apart.
But candidly, I also had so many assumptions and judgments, and while I think there may be some doublespeak or hand-wavey type thinking/justification (or basically some fundamentally differing world views), I think a lot of my assumptions were blown away by the:
clarity of thinking
conviction Nikita has in his worldviews
🎧 I learnt a lot from this 1 hour 40 min pod (listened on 1x speed because the density of insights was high). I’m going to put a bunch of miscellaneous notes in footnotes, you should watch the whole thing whether you’re B2C or B2B.
There are things I don’t want to sacrifice or do right now (e.g., work until 3 am, get 2-3 hours a sleep a night). But that doesn’t mean I still cannot take some important and useful lessons from him.
One of the biggest ones which I still think you can only really learn to do right and calibrate through experience, but nonetheless I’m going to continue to try and impart everywhere is (slightly paraphrased, but for clarity in written vs. spoken form):
A great way to do 0 -> 1 product development is to execute 100% on testing the ONE thing that you’re trying to validate at that specific stage of the product development cycle.
You need to focus on this one thing, and sort of half-ass everything else.
Compartmentalizing those other things is important because otherwise you’ll have too much scope creep if you try to solve too many things at one time, and you’re not going to get signal unless you test one thing at a time.
When testing for consumer product development, I have a heuristic/think about it like this:
If this is true, what next needs to be true for this thing to work out.
With multiple layers of these ‘if, then’ conditional statements.
The more layers you have the more risk you have.
So you should try to condense it to like 4 things that must be true for the thing to work. (I advise everyone)
So for example, for him, with TBH and Gas:
Is the user experience, flow perfect - for them the polling experience, questions, notifications
Then, does it get shared by peers (next stage was getting it shared and virality, the flywheel going)
Then, does that sharing hop peer groups? (e.g., from school to school and beyond)
To be honest, this applies EQUALLY to B2B startups, so it was for me fascinating and validating that this framework does work for both, even if my own personal expertise and zone of genius is in B2B.
So, whether you’re:
my students in Penn GSE’s Masters of Education Entrepreneurship
the next cohort of Transcend Fellows I’ll be coaching
any other founder I talk to
myself, when considering what I’m build now
Remember these words of wisdom: A great way to do 0 -> 1 product development is to execute 100% on testing the ONE thing that you’re trying to validate at that specific stage of the product development cycle. Compartmentalizing everything else is important because otherwise you’ll have too much scope creep if you try to solve too many things at one time, and you’re not going to get signal unless you test one thing at a time. (even if you set everything up as a bundle: pain point/problem/JTBD, ICP, value prop, positioning)
Additional, misc unorganized notes/nuggets in the footnotes4
Listen to him talking about the billions, or trillions it’ll take to build AGI —> plus raising $6.6 billion on a $157 billion post-money valuation; if that’s not the ultimate latest hustle in a string of hustles for a company that is expected to incur $14 billion in losses in 2016, I’m not sure what is —> building a nonprofit, raising a killer round from Microsoft, getting ousted then coming back in dramatic fashion to the said ‘nonprofit’ to say nothing of his previous stint at YC…
Partner at Founders Fund. He is also Co-founder and President of Varda Space Industries, which is building the world's first space factories; you know you’ve made it when you can use the following in your LinkedIn about section:
world's first space drugs and arms dealer @VardaSpace
village idiot @foundersfund
The more I listened to him, the more I could see he’s a designer, and he shares a lot of similar traits as Brian Chesky of Airbnb; they both share similar sentiments about product managers given how Paul Graham triggered a whole discussion from his talk on ‘founder mode.’
Sharing (virality) reduces 20% every year someone is older from age 12/13 onward —>
Consumer apps esp social media is hard for adults, etc
The high school thing was not growth, only testing - simultaneous release to get enough density and test engagement and messaging - to know core works esp inside of a school/specific social graph sharing works
Human trafficking - anyone who builds a popular social app usually happens —> because whoever posts this gets tons of followers (such as powerful insight on why fake news is so powerful - it has pull, gives you cache whether it’s true or not; people respond to incentives)
Relaunched/ rebranded gas once on other side of country, started going viral again
Story re-emerged because on friend who was on the app in one state told the other friend it was for human traffickingDespite trying to escape it, Nikita finally got energized and decided to fight it on multiple fronts (so the company wouldn’t die), SnapChat founder, police chiefs and school sups, wapo headline gas is not for human trafficking
300,000 downloads per day required to be #1 in App Store; at peak TBH or Gas 360,000
TBH
Gas
At Facebook
Resources, academic science of social networking at Meta / Facebook
Product at large companies, cannot be completely intellectually honest (want to use app for flirting)
Product glorified secretaries, sit far away from from design, data and users | cannot work for 0 -> 1
Second with Gas, really wanted to test monetization
Never raised
Once he saw unit economics, how it was growing, monetization conversion
He negotiated with AWS, Mixpanel, all the credits to get cost as tightly controlled as possible (aligns with the idea of driving satisfaction, growth, then efficiency per First Round PMF Levels)
It was just all cash flows 10 mm users, $11 million
Didn’t expect to sell but then people came calling (3 offers, took Discord’s)
Using Transcend’s PMF framework:
Pain point: People want to feel good about themselves - high schoolers / teens
Magic - share a poll, get anonymous compliments on yourself
Retention - keep doing it, send a bunch, react to a bunch because it makes you feel good/better
(though we won’t really every know about a ‘durable’ social app because he keeps getting acquired)Growth - vitality and sharing
Had to rebuild growth system from TBH to Gas 5 years later, could no longer just use Twilio and invite friends from app, regulation made it so have to send msg from phone, so takes you to compose then send, that added step introduced friction and significantly reduced sharing
Monetization - did they monetize on revealing who sent (that was a hugely requested feature, people would pay $1 to see who sent them from TBH), or?
Nikita app meme
When relaunched different names rebrands: crush, Melt,
Thought crush was great, had a great domain
Tested it
Invitations dropped significantly under crush name, invite to app named crush
Boys invite boys -> boys didn’t want to invite. A friends to pap called crush with pink icon
Girls invite girls to apps
App was 60-65% women
Hence, why made it gas with flame design and black background —> and invites rates jumped ⛽️ 🔥