I built a tool that finds quiet quitting signals in Slack/Teams — before someone hands in their notice
Hey IH! 👋
Solo builder here — bootstrapped, no funding, dev work is fully AI-assisted. Just shipped ExitSign.
The problem: By the time someone says "I'm leaving," the decision is usually already made. But before that, there are usually quiet signs — less participation in channels, one person dominating every conversation, late-night messages creeping up, someone slowly going silent.
What ExitSign does:
- Connects to Slack or Microsoft Teams
- Tracks message volume changes, channel dominance %, late-night activity % — never reads message content, never stores raw logs
- Shows managers an ExitSign Score + an industry percentile benchmark
- Suggests one concrete next action per alert (who to check in with, and why)
What it's NOT:
- Not a surveillance tool
- Not meant for performance reviews
- Not a pulse survey — employees don't fill out anything
Right now I have 2 companies using it (friends, free tier) while I build out an accuracy feedback loop — managers mark each alert as "accurate" or "off" so I can actually measure precision over time instead of just guessing.
If you run a team on Slack or Teams and want to try it: reply or DM me and I'll set you up with a free access code. Genuinely want more real usage and harsher feedback before I think about pricing seriously.
Especially curious to hear from anyone who's used pulse surveys (Culture Amp, Officevibe, etc.) and got frustrated by low response rates or "answers that don't mean anything." That's the exact gap I'm trying to fill.
🔗 https://exitsign.jp/
One thing I find interesting about products like this is that the difficult part often isn't noticing a pattern.
It's deciding what should happen because of it.
A declining participation score feels objective.
The decision to check in with someone is much less so.
Somewhere between those two moments, an observation becomes an action.
That's the part that stood out to me reading this.
Honestly, completely agree with you — that's the exact tension I'm still working through.
That said, ExitSign does try to bridge that gap a bit already. The logic right now is roughly: if it's an individual's participation dropping, it nudges toward a 1:1 check-in; if the pattern looks more structural (e.g. dominance/imbalance across a whole channel), it nudges toward addressing it at the team level instead of pointing at one person. So it's not just "here's a number, figure it out" — but I'd say the action layer is still the least mature part of the product.
Everything here is genuinely unfinished, so I'd love more opinions like this one. Thanks for actually taking the time to reply — it's the kind of pushback that's actually useful.
If you don't mind, I'd also be curious what you think of the framing/wording I'm using around this (the "observation vs action" distinction, how I describe the score, etc.) — not just the product logic. Happy to dig deeper if you're up for it.
That's actually the part I'd be most interested in digging into.
Not just the product logic, but what happens once an observation starts implying action.
The wording ends up mattering more than people expect there.
What's the best email to reach you on?