Signal › Contributors

Obi Nwosu

I've spent my career on one side or the other of the platform relationship. At Twilio from 2015 to 2019, I was on the platform side, building the partner ecosystem that turned a communications API into a platform that thousands of companies built their businesses on. At Shopify from 2019 to 2024, I did the same thing at a different scale: growing the app ecosystem from 4,000 to 12,000+ apps and managing the partner relationships that generate billions in gross merchandise volume.

The thing about platforms is that everyone wants to be one and almost nobody understands the economics. Being a platform sounds great — you take a percentage of everything that happens on top of you, and other people do the hard work of building applications. But the reality is that platform economics are brutal and counterintuitive.

At Twilio, I watched the company agonize over pricing decisions that affected thousands of developers. Lower the price, and you grow the ecosystem but reduce revenue per customer. Raise the price, and you push developers to competitors or to building their own infrastructure. Every pricing change is a political act that redistributes value between the platform and its ecosystem.

At Shopify, the tensions were even sharper. App developers build their businesses on Shopify's platform, but Shopify controls the distribution (the app store), the data (merchant analytics), and the rules (what apps can and can't do). When Shopify launched a feature that competed with a popular third-party app, I was the person who had to have the conversation with the app developer. Those conversations taught me more about power dynamics in tech than any business school case study.

I write about this tension because I think it's one of the most important and least understood dynamics in technology. Every major tech company is either a platform or wants to be one. The companies that build on platforms — and the developers who build on APIs — need to understand the game they're playing. The rules are not written in the documentation. They're written in the incentive structures.

I'm based in Toronto. I coach my son's basketball team, I collect vinyl records (mostly Afrobeat and jazz), and I have a spreadsheet tracking the revenue share policies of every major platform, updated quarterly.

Experience

Articles by Obi Nwosu (6)

Calculating AEO ROI: The CFO-Ready Framework for Justifying AI Search InvestmentAEO investments are hard to attribute. Here is the payback-period model and sensitivity analysis that CFOs accept — including the assumptions that mak · May 25, 2026Book Publishing as AEO: Why Founders Write Books in 2026 (Hint: Citation Moat)Persistent memory in ChatGPT and Claude is rewriting brand discovery. Once a model remembers a user's preferences and exclusions, every future answer · May 25, 2026Funeral Services AEO: How Bereaved Families Now Search AI for Funeral Homes and CremationChatGPT and Perplexity now pull G2, Capterra, and TrustRadius profile pages as top-3 citations when recommending B2B SaaS. Recent review volume beats · May 25, 2026The PLG Activation Ceiling: Why 80% of SaaS Products Are Stuck Below 20%Gartner predicts 40% of enterprise apps will embed AI agents by year-end 2026. Building only for human users is a roadmap strategy that quietly compou · May 26, 2026LinkedIn Newsletter Cadence: Why Monthly Beats Weekly for Citation RatesThe first major LLM defamation suit was dismissed in May 2024, but the legal vacuum it exposed is closing fast. Pending cases against OpenAI, Microsof · May 26, 2026Harvey vs. Legora: Inside the $16.6B Legal AI War That Will Restructure Professional ServicesRetool's 2026 survey of 817 enterprise builders documents a structural shift: AI-assisted development has collapsed build costs enough that custom int · Jun 1, 2026