GTM advice from my experience at 2 YC Startups

3 points

a year ago

I’ve been lucky enough to lead GTM for two YC-backed startups and ship multiple AI products from 0 to real adoption. I say “real” because there’s a huge difference between people signing up and people actually using your product (I have seen both)

Getting your first 100 real users is brutal. Especially in AI right now. Everyone's building something. Most users have seen it all. They don’t care unless your product solves a very specific, painful problem.

Here’s what actually worked for us, not theories, not growth hacks, real dirt-under-the-nails strategies:

1. Small, Weird Communities > Broad Outreach Reddit threads with 2 comments. Slack groups with 200 nerds. Discord servers that feel like a nerdy group chat. That’s where early users lives. We used redditsearch.io + Pushshift to find conversations our ideal users were already having.

2. Trend-Hacking + Contextual Insertion We didn’t launch on a random day. We launched during the FIFA World Cup and framed our sports commentary AI accordingly using our product. For another tool, we anchored it to a debate on the new Indian cricket captain — got crazy engagement.

3. Public Series (not just threads) We ran daily series like “Day 1 of building X with our tool” on LinkedIn and X. We weren’t promoting, we were storytelling — and it worked. By Day 4, we had DMs asking “what tool is this again?”

4. Show up where competitors are We didn’t try to be the hero of every story. We just showed up. Commented on related posts. Messaged people who commented on competitor launches with “hey, we’re building something like this but for [niche] would love to show ?”

5. Lazy-Friendly Content = Growth People don’t want to dig. We did the digging and published curated lists like “Top 10 AI tools for [persona]” and subtly included ours. These lists traveled far — way more than a “launch” post ever did.

One final thing: GTM isn’t one launch. It’s 1:1 conversations, comment by comment, for months. But if your product is good and you really listen to user pain, things do start to flow.

Also building Varnan(https://varnan.tech) now with all this knowledge for early stage high growth startups.

Happy to answer questions — AMA.