vishal@gtm:~/case-studies — zsh cd ..

Case Studies

What actually worked · what I'd skip · the honest version

Six years in startups across GTM, marketing, and SEO. Two tiers below: direct wins I owned end-to-end, then the agency portfolio with Torcl Labs where I was part of the team across 11 shipped projects.

Personal wins — owned end-to-end

Uncle Rudy · 150 users · 2 days · beta cohort of a few people globally

Uncle Rudy — Reddit-only launch for an iOS 27 beta app

iOS 27 was still in beta when we launched Uncle Rudy — the testable universe was tiny, maybe a few thousand developers and testers worldwide. No ads, no newsletter, no founder audience. I ran a Reddit-only distribution play — the right subreddits, native-voice posts, and comment-first engagement before any link drop.

What moved the needle: picking three subreddits where iOS 27 beta users actually hang out, posting the problem first (not the product), and waiting for the question before the link.

What I'd skip: cross-posting. It burned goodwill faster than it drove installs.

GTM · Reddit · Beta launch · 2026
Enterprise GTM motion · built from zero

Conscious Engine — full GTM from scratch

Enterprise B2B with no existing GTM. I built the whole motion from scratch — ICP definition, positioning, outbound infrastructure, content engine, and the hand-off playbook between marketing and sales. Everything from "who do we sell to" to "what does the SDR say on the second call."

What moved the needle: narrowing the ICP twice before writing a single email. Enterprise targeting fails because teams go broad; this one worked because we picked one sub-vertical and owned it.

What I'd skip: the first content calendar. Written before we had the voice right — had to redo most of it after the first five sales calls.

Enterprise GTM · ICP · Outbound · Content

Agency portfolio — with Torcl Labs

Honest framing: I worked on all 11 projects below as part of the team at Torcl Labs. Credit is shared with the agency and the clients — I'm linking Torcl's own case study page so the numbers are verifiable at source.

See the Torcl Labs case study page →

Want one written up in depth?

Tell me on LinkedIn which case you'd want long-form — I'll move it up the queue.

← back to desktop · full /read view · llms-full.txt