two & a half gamers – 🧠 What You Send Is What You Get: Signal Engineering and the Future of Attribution with Core Plan

Summary

The old way of buying an MMP took weeks — sales calls, demos, contracts, CSM onboarding, documentation. Airbridge just collapsed that into a couple of hours.
Core Plan: https://abr.ge/xqaqqlu
Use code Matej26 for bonus attributed installs!

Matej Lančarič sits down with Roi Nam, CEO and founder of Airbridge, to unpack Core Plan — a self-serve, pay-as-you-go MMP that comes with 15,000 attributed installs free for a year. They get into why now (AI has driven a ~60% year-over-year jump in app releases, and those founders need measurement fast), who it’s for (founders under $10M ARR, teams of 1-20, mostly consumer and subscription apps), how the AI-native onboarding works (MCP and an AI pilot that installs the SDK and builds reports for you), what got stripped out to keep it “core,” and the roadmap — instant pre-SDK analysis from your ad accounts, easier web-to-app, and built-in signal engineering. On that last point: Roi shares how one sleep-tracking app cut CPA 27% with the simplest signal-engineering tactic — delaying the cancellation signal to Meta.
The throughline: measurement should be as fast as the AI tools founders already use.

📌 KEY TAKEAWAYS
— Airbridge launched Core Plan, its first-ever product-led-growth motion in 10 years as an MMP. It’s self-serve and pay-as-you-go, with 15,000 attributed installs free for one year, then 5 cents per attributed install (installs, not conversions) after that. Founders can sign up on the dashboard and be running in 2-3 hours.
— The “why now” is AI. App releases are up roughly 60% year-over-year (per a16z) as founders use AI to research, pick a segment, and ship apps in days. But while AI tools deliver value in minutes, MMPs historically took weeks to two months to onboard — sales calls, contracts, CSM, documentation. Core Plan closes that speed gap.
— The target is small-to-mid-market founders under $10M ARR, teams of 1-20, who grow primarily through UA. Currently ~85% apps (vs games), ~60% subscription apps, with more than half of users from the US, UK, and Europe but sign-ups genuinely global.
— Core Plan is ~65% of the enterprise feature set. It keeps the essentials — cohort analysis, raw data export, deep linking, up to two integrations (RevenueCat, Adapty, AdMob, Amplitude, Mixpanel, Braze) — and strips advanced reporting, multi-touch attribution, incrementality, and agency/security management. The 15K free installs are calibrated to get an early app to ~$3-5K MRR and prove product-market fit.
— The AI-native onboarding is the core differentiator. An MCP integration (drop it into Claude Code or Codex) plus an embedded AI pilot lets founders install the SDK, set up taxonomy, place tracking codes, and generate reports by asking — instead of reading documentation for hours. Tech-savvy founders are getting the SDK live in 2-3 hours vs the 2-month enterprise timeline.
— The roadmap is ambitious. Pre-SDK instant analysis (sign in your Meta/Google/TikTok/Apple ad accounts and get recommendations before installing anything), streamlined web-to-app setup guided by the AI pilot, and built-in signal engineering — all targeted for the coming months.
— Signal engineering is the sleeper topic. Everyone talks about it; almost nobody implements it. The idea: ~20% of trial subscribers cancel within 20-30 minutes, and sending that signal to Meta/Google can mislead the algorithm about your real audience. The fix can be as simple as delaying the cancellation signal by ~2 hours. One sleep-tracking app (a category leader in Japan) cut CPA by 27% with that single tactic. Airbridge wants to make this a toggle on the dashboard rather than a custom engineering project.
— Roi’s bigger bet is the “thick tail.” AI is unlocking a flood of new, weird, niche apps across every vertical, and the long-tail app market is actually a thick-tail market. The strategy is to meet those founders early, make measurement simple and fast, and grow with them — they don’t care what an MMP is, they just want to turn ad spend into revenue and measure it correctly.