How do you decide when to kill a product vs give it more time?

[Rpust]

Been sitting on this question for a while and figured this crowd would have real answers.
I've got about 15 products running ads right now. Some are clearly working, some are clearly dead. But there's this middle group of maybe 5 or 6 products that get occasional sales, not enough to scale but enough that I feel weird cutting them. Like one of them sold 3 units last month which covered its ad spend but barely. Another one gets decent clicks but converts maybe 1 in 80 visitors.
The problem is I don't have a clear system for when something goes from "needs more time" to "cut it and move on." Right now I'm just going on gut feel which I know isn't great.
Do you guys have actual criteria you use? Like a minimum ROAS threshold, a specific number of sessions before you judge conversion rate, a time window? Or is it more situational for you depending on the niche and price point? Curious how people who've been doing this longer than me think about it.

 

[shears]

My rule is 300 sessions minimum before I judge conversion rate, otherwise the data is just noise. For ROAS I give anything 3 weeks at consistent spend before making a call - some products just need the algorithm to find its audience. That middle group you're describing is the hardest part honestly. What I do is cut ad spend to minimum on those, let them run passively, and if they can't even cover spend at the lowest budget they're done. If they occasionally convert organically they stay in the catalog, just not getting active budget.

 

[KatLopez]

That 1 in 80 conversion rate is actually the more interesting number to me than the ROAS. Decent clicks with weak conversion usually means one of three things - wrong audience reaching the page, product page not doing its job, or price point is off versus what people expected from the ad. Worth diagnosing which one before cutting it.
The product covering ad spend with 3 sales I'd just drop to minimum budget and leave alone. Not worth active attention but not worth removing from the catalog either.
My hard rule is if something can't hit 1.5x ROAS after 3 weeks of consistent spend and at least 200 sessions, it's done. Gut feel gets expensive fast.