How Revid doubled its ChatGPT-sent signups from 355 to 781 per week in 10 weeks
tibo_maker · x · 2026-11-17
tibomaker shares the full playbook Revid used to grow weekly new users arriving via ChatGPT from 355 to 781 in 10 weeks.
Key insight: when asked for product recommendations, ChatGPT doesn't answer from memory — it searches the web, reads a handful of pages, and you can inspect exactly which ones.
Actionable steps covered:
- A copy-paste prompt to audit your product's standing in ChatGPT recommendations in 5 minutes
- Finding the listicles ChatGPT reads and winning top spots via Outrank
- A second outreach round per target keyword with Squad
- Climbing the Reddit threads ChatGPT pulls from using Bazzly
The author shows week-by-week results and argues the stack compounds over time.
More from coding & agent
- FormMCP: one MCP connection to a family of desktop apps, tools loaded on demand — nodemode_ai · 2026-11-17
- Pi harness adapter ships as a LoRA runnable on vLLM, llama.cpp, and MLX — yb2698 · 2026-11-17
- Pi harness adapter via RL: SWE-bench Pro 29.4%→62.7% with 55% fewer tool calls — yb2698 · 2026-11-17
- Real SWE-Bench Pro task: base model hits 100-step limit, adapter solves it in 8 calls — yb2698 · 2026-11-17
- Constrained Decoding for Function Calling Cuts Endpoint Throughput by 17%-90% — remilouf · 2026-11-17
- Chamfer: open-source 2D/3D CAD workbench with MCP server and analytical DRC for coding agents — Weak-Cow1889 · 2026-11-17