The Four-Step Distribution Platform Cycle
Every new growth channel opens then closes in the same four predictable steps — learn to see where you are.
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 4
- Confidence
- 95%
A repeatable lifecycle model for how new distribution platforms (Facebook, Google, iOS, LinkedIn, and — Balfour predicts — ChatGPT) emerge, open up free organic reach to third parties, then close down to monetize. Recognizing which step a platform is in tells you whether there is still a land-grab window and when to plan your exit. Balfour's core claim: the steps repeat every time, and the cycles are getting shorter.
Origin
Brian Balfour's synthesis from living the full Facebook social-gaming platform cycle at his first startup and watching Google, iOS, and LinkedIn repeat it. He explicitly credits Alex Rampell (a16z) for the 'escape velocity — get distribution before the incumbent copies you' framing (a ~2015 blog post) and Casey Winters for the insight that a technology shift and its distribution shift arrive separately, the distribution shift lagging.
Core principles
- 01A technology shift (AI) and its distribution shift arrive at different times; the most powerful shifts eventually spawn a new distribution platform.
- 02Startups exploit new platforms faster than incumbents can, which is the whole opportunity to reach escape velocity before being copied.
- 03Platforms open organic distribution to gather a moat, then close it for monetization and self-protection — this is capitalist incentive, not villainy.
- 04The window between opening and closing is shrinking with each successive platform, so you have less time to act than the last cycle suggested.
How to run it
- 0
Market conditions met (consensus, no winner)
A new category has clear consensus it will be huge, but there is no clear winner — typically five to seven major players fighting fiercely because these markets end in monopolies or duopolies. Identify that a category is in this state (AI chat: OpenAI, Claude, Gemini, Meta all battling with huge capital).
Pro tip High capital + consensus + no winner + 5-7 fierce players is the signature of Step Zero. When you see it, a distribution platform is likely coming.
- 1
A moat gets identified and pressed
One player identifies the source of defensibility — the flywheel that will let them hit escape velocity and become the monopoly/duopoly — then races to gather it as fast as possible. For Facebook it was the friend graph; for ChatGPT, Balfour argues, it is context + memory (more usage feeds more memory feeds more personalized context feeds better output).
Pro tip To spot the likely winner, look for the moat flywheel plus the best retention and depth of engagement — not the biggest current user base. Google and Facebook both won on retention, not initial scale.
Watch out The player with the most distribution today is often not the winner; watch for 'flyby' vanity users inflating a rival's MAU.
- 2
Platform opens to third parties
The moat can't be gathered alone, so the player opens a third-party platform for developers, content creators, and businesses. The value exchange: build on my platform and add use cases/engagement, and I give you distribution, plus (for AI) context and memory. This is the land-grab window where you can grow like crazy. Facebook's 'canvas' let developers monetize freely in exchange for sidebar ad real estate and access to notification/feed distribution.
Pro tip Preferred-partner deals with big brand names (the 'guinea pigs', an initial 10-20) are the credibility signal that the open phase is starting — watch for them as the go-early trigger.
Watch out Being early beats being late here, but 'early' still requires anticipating Step 3 from day one.
- 3
Platform closes for control and monetization
The platform locks down via one of three moves: (1) shut the platform down entirely, (2) build first-party apps that absorb the highest-value use cases, or (3) artificially depress the organic distribution it gave you to push you toward paid mechanisms. Google filled real estate with ads and first-party answers; Facebook absorbed events/photos; Apple imposed the 30% tax; Udemy cut creator rev-share from ~80% down toward 15-30%.
Pro tip Companies that survive built their exit before this step — owning a workflow, accumulating proprietary data, or creating micro network effects the platform can't replicate.
Watch out Companies whose only channel was the platform's organic distribution (e.g. SEO-loop businesses) get destroyed when Step 3 hits. Never let a platform be your single channel.
In the wild
Facebook was ~1/4 the size of MySpace/Friendster in 2007 but understood the direct-network-effect moat. It opened the 'canvas' platform giving developers free monetization plus notification and feed distribution, triggering a viral social-gaming gold rush. Then it peeled back the value exchange: took a percentage of canvas revenue, suppressed organic channels, and absorbed top use cases into first-party apps.
→ Companies built purely on Facebook distribution were 'basically shut down for dead,' while Facebook rode the developer-fueled growth so far ahead that competitors no longer mattered. Zynga, which moved fast, became a massive company.
Both mobile ecosystems opened developer platforms. Balfour notes that aligning your bets only to Android tended to lose, while betting on iOS — even with a smaller user base — could win, because you had to have iOS in your betting strategy.
→ iOS captured ~70% of market dollars on ~30% of devices; the exact flip of Android, proving retention/monetization quality beats raw user count.
Common mistakes
Refusing to play ('screw them, I'm not doing it')
The instinctive reaction to the closing phase is to opt out. But it's a prisoner's dilemma: if you don't join, competitors will, and customer expectations shift to expect you in the new experience. There is no opting out of the game.
Treating the platform as evil rather than as a game
The cycle is driven by competitive and capitalist incentives, not malice — markets don't reward flat companies, so platforms must keep growing and protect against self-disruption. Moralizing wastes energy you should spend learning the rules and arbitraging them.
Entering without an exit plan
Most companies that got killed failed to anticipate Step 3. You must plan your exit the moment you enter, not when the platform starts closing.
Is it for you?
Best for
Founders and growth leaders deciding whether and when to build on an emerging platform (currently AI chat / ChatGPT agent platforms)
Not ideal for
Businesses in mature, stable channels with no new-platform dynamics, or teams that can't tolerate turning strategy on a dime when the platform opens
From the transcript
“one is like I call a step zero. It's it's the the conditions of the market have been met. Um, step one is about a…”
“They either shut it down entirely, two, they develop their own firstparty applications to absorb the the highest uh you know use cases, or three,…”
“the broad trend is that the cycles seem to be getting shorter and shorter and shorter and shorter. So you actually have a smaller amount…”
“startups is a game of trying to get distribution before the incumbent can copy”
From the episode
Why ChatGPT will be the next big growth channel (and how to capitalize on it)
Brian Balfour (Reforge)