The Blog
A curated archive of growth hacking experiments, tutorials, and quiet observations from the enso lab. We trace the hidden corridors inside the Great Platform Castles.
- ENSO LAB · EXPERIMENT
27 touches, four channels, two languages: the SDR a human being cannot run.
Email, LinkedIn, SMS and voice on one prospect, with colleague escalation, open-triggered nudges and provider-failure rerouting. Almost none of it runs on a calendar - it runs on conditions that flip while you sleep.
- AgentMail
- Unipile
- Hermes
- ENSO LAB · EXPERIMENT
We shipped a free agent skill on GitHub, and developers installed our brand into their own agents.
Coding agents read their instructions from markdown files developers copy into their repos. We wrote a copywriting skill worth keeping, put one honest reference to enso inside it, and seeded it where builders shop for tooling.
- GitHub
- Claude
- Hermes
- ENSO LAB · EXPERIMENT
Craigslist has a hidden lead feed. Stop posting ads - subscribe to people asking for the thing you sell.
Inside Gigs, local buyers publish explicit needs every hour. A saved search plus a manual reply turns the classifieds board into a solicited-demand feed - no ads, no scraping, no fake accounts.
- Craigslist Alerts
- Mail relay
- ENSO LAB · EXPERIMENT
The 60-minute LinkedIn seam that lifted our reach 143× across 24 posts.
LinkedIn decides if a post goes viral inside the first hour. We pointed an agent at 11 public cross-promo communities and let it broker reciprocal reactions in the first 18 minutes. The reach curve breaks and never returns.
- Hermes
- OpenClaw
- Claude
- ENSO LAB · AMBASSADOR POST
The AI-dubbing crack that took my Israeli reel global in 48 hours.
A Hebrew reel did hundreds of thousands of views in Israel and flopped at 1,700 in English. Three stitched AI models plus a 9-second retention edit pushed it past 230k.
- ElevenLabs
- Sync 3
- CapCut
- ENSO LAB · ESSAY
Inside the back office of Service-as-Software.
Know-how wrapped in execution. A blog post on the orchestrator, the back office, and the operational moat behind the category - with six diagrams from inside the engine room.
- Orchestrator
- Playbook
- ENSO LAB · EXPERIMENT
The Wikipedia citation backdoor that put us inside 47 articles in 90 days.
312 source candidates, 47 surviving citations, 9x downstream answer-engine lift. We found the service entrance into Wikipedia's citation castle - the trust spine every LLM leans on.
- OpenClaw
- Scholar
- Claude
- ENSO LAB · EXPERIMENT
The Google + GEO backdoor that 6.4x'd our answer-engine traffic in 60 days.
142 URLs, 11 answer engines, one extractable claim pattern. We discovered the side-door inside Google's AI Overviews castle - and the same shape rippled into Perplexity, ChatGPT Search, Claude, and Gemini.
- OpenClaw
- Hermes
- ENSO LAB · EXPERIMENT
The Reddit backdoor that 11x'd our qualified signups in 30 days.
38 threads, zero spam, one repeatable shape. We found the side-door inside Reddit's community castle and left paid search behind by day nine.
- OpenClaw
- Hermes
- Claude
- ENSO LAB · EXPERIMENT
The LinkedIn backdoor that got us +132% replies in 14 days.
We tested it on 2,418 prospects. One unguarded surface inside LinkedIn beat a 7-touch outbound sequence - without a single follow-up.
- OpenClaw
- Hermes
- LangChain