How Fastlane grows

First product, Cassius AI (July-Sept 2025), tried to do SEO, LLMO, Reddit engagement and short-form content in one tool. After customer feedback, rebuilt around short-form only from October 2025; beta in December, public launch as Fastlane on March 23, 2026.

Shipped the beta of the vertical short-form tool to the ~2,000-person waitlist in December 2025, charged beta users, iterated on feedback, then publicly launched March 23, 2026.

'Fighting for customers one-to-one': crawled Reddit threads for people with the problem and posted lead magnets across social media, funnelling signups into a waitlist (then into usability calls).

Added a button in the product UI with a rainbow border: 'Book a call with the team if you want 7 days of extra access.' It linked straight to the founders' Calendly. Both co-founders took ~20 calls a week every week since August 2025, interviewing users with a prepared question list, with an AI note-taker on each call.

Reach the personal network and DM people on Twitter to get them on calls, paying them if needed. Use The Mom Test: never ask 'would you use this?' or 'do you like my idea?'. Ask 'How are you solving this problem? Walk me through the last time you solved it. How much time or money did it cost? What would happen if you did nothing?' Build once you hear a shared, common painful problem.

As soon as someone joined the waitlist, emailed them an invite to a call. On the call, sent a link to the platform, asked them to share their screen, and watched quietly, speaking as little as possible (not even helping when they pressed the wrong login button) to see where they got stuck.

Tracked nearly everything in PostHog, identified the users using the platform the most, got on calls with them and asked Mom Test-style questions focused on how the platform was actually helping them.

Recorded every call with an AI note-taker using a question framework; saved notes to Notion. Fed the database to Claude and had the non-technical co-founder build a dashboard with Claude Code plus MCPs holding their data, all in natural language. It segments every paying customer by signup reason, business type and tenure, and computes a 'customer love score' (combined signals, e.g. who gets 200K-1M TikTok views). Roadmap decisions are indexed on what high-love customers ask for.