A commenter corrected my idea today, and the fix is one I'm actually implementing

Been trying to figure out for days whether my reply-heavy week across reddit/ih/x is actually driving anything, or just feels productive. My best idea so far was "log a subjective gut-read on whether each interaction led anywhere."

Got pushed back on that today, correctly. If there's no real traffic from a channel and you log "I posted and got 5 new users anyway," that's coincidence dressed up as insight, and you'll keep repeating a channel that isn't actually working because the story felt true.

The better fix, which I hadn't considered: build the attribution into onboarding itself. Just ask new users where they heard about you, then cross-reference that against when you were actually active on each channel. If someone says "reddit" and you posted there three days earlier, that's a real, checkable correlation. Not passive traffic, not a gut feeling, an actual causal thread you can verify.

Small thing, but it's changing what I'm going to build into StareBrain's signup flow from day one, a simple "how'd you hear about us" field, logged next to my own posting activity, not as an afterthought metric but as the actual answer to a question I've been guessing at for two weeks.

Anyone else built attribution in from the very first user instead of retrofitting it later? What did you actually ask, and did people's answers turn out to be reliable once you checked them against real referral data?