How to Win the Next Distribution Cycle

Brian Balfour

Founder & CEO, Reforge

2024

The Moment

New Distribution Emerges Every Cycle

"Building a great product is necessary but not sufficient. The separation between winners and losers is those that build really great distribution."

Every major distribution channel, SEO, paid ads, social, eventually saturates and declines
New platforms emerge with open ecosystems that favor early movers
The window to gain escape velocity is shrinking as incumbents copy faster
AI technology shift is paired with emerging distribution platforms (ChatGPT, agents, etc.)

The Framework

Every Platform Follows a 4-Step Cycle

Step Zero: Consensus, No Winner

Multiple players (5-7) battling for dominance in a new category. Everyone agrees the category is huge, OpenAI vs Claude vs Gemini vs others, but no clear winner yet. Stakes are massive, competition fierce.

Step One: Build the Moat

Winner identifies their defensibility advantage and opens a third-party platform. Value exchange: developers build on your platform, you give them distribution. The platform becomes an ecosystem.

Step Two: Gather the Moat

Creators, developers, and businesses flood the platform to access that distribution. The more useful content and apps, the stickier the platform, the larger the moat.

Step Three: Close & Monetize

Platform locks down. Either shut down competing apps (Vine, Periscope), launch first-party alternatives, or depress organic distribution to push toward paid (Google, Facebook).

Historical Proof

Facebook Shows the Full Cycle

Step 0 (2004-2007): MySpace, Friendster, and others had MORE users than Facebook, but Facebook won the network effects game
Step 1-2 (2007-2009): Facebook launched its platform. Social gaming exploded (Zynga, FarmVille). Developers got early distribution access. Facebook got stickiness and engagement
Step 3 (2010-2024): Facebook closed the gates. First, they shut down Vine and Periscope directly. Then they depressed organic reach, forcing brands to buy ads. Viral organic distribution nearly killed

The prisoner's dilemma

Facebook platform developers made millions. Then Facebook changed the economics. By 2024, their rev share moved from 70% down to 15-20%. You can't opt out, your competitors are already on the new platform, so customer expectations shift.

Same cycle ahead for AI

Cursor, Claude, and ChatGPT will likely all build agent platforms. They will start open to attract developers, then close down as they mature. Early moats built on data, user feedback, and network effects.

Playbook

Where to Place Your Bets

Startups: Choose one platform and go all-in. You don't have resources to play multiple bets. Early bet on iOS vs Android in 2007 was life-or-death
Late-stage companies: You must play multiple platforms. You have scale and capital to hedge bets across emerging channels
Timing: The next 6 months are critical. ChatGPT may launch a full third-party developer platform. Watch for: agent mode rollout, preferred partner announcements, then full platform opening
The trap: Don't assume it's too early. Your competitors are already placing bets. By the time it looks obvious, you're late

Competition erodes everythingEven outcome-based AI agent pricing ($1 per solved issue) will compress as competition increases. Build a durable moat, data feedback loops, specific use cases, network effects, not just pricing power.

Contrarian

Distribution Myths You Need to Break

✗SEO and paid ads are foreverINSTEAD →✓ Every channel saturates. Organic reach is dying on LinkedIn, Twitter/X, and search. AI and agent platforms will be the next escape hatch.

✗You can wait and see if a platform winsINSTEAD →✓ By the time it's obvious, you're already late. The cycles are getting shorter. First movers gain escape velocity before incumbents can copy.

✗Great products don't need distribution strategyINSTEAD →✓ Product quality is table stakes. The winners are those who understand which distribution channels to dominate and when to place bets on new ones.

✗You have time to figure this out laterINSTEAD →✓ There is no opting out. Your customers' expectations are set by what your competitors do. If they bet on ChatGPT agents and you don't, you lose.

Based on Brian Balfour's episode on Lenny's Podcast. All ideas on this page are from the episode.Watch on YouTube Follow @bbalfour on X

GO DEEPER IN THE EPISODE

04:13 The changing landscape of product growth 05:09 The importance of distribution 08:14 The role of new distribution platforms 09:45 The four-step cycle of distribution platforms 17:38 Examples of platform cycles 30:01 The rise of ChatGPT 44:47 The future of AI agents 46:01 Preferred partners and platform credibility 47:18 Monetization mechanisms and free tiers 48:14 Betting strategies for startups 01:04:34 Adopting AI tools: challenges and strategies 01:08:41 The importance of hard constraints 01:14:23 Effective AI adoption in companies 01:19:05 Lightning round and final thoughts

Every sketch here began as a transcript.

Send us yours and we’ll draw it. $1 a transcript.

Sketch mine →

0:00