Launched this week
marketfunkers

marketfunkers

One brain for ad research, insights and testing.

210 followers

Marketfunkers isn’t another AI that spits ideas. It’s a creative intelligence platform that tells you why ads work, why they fail, and exactly what to do next. Upload any ad and get real audience language from Reddit and reviews, pattern detection across ads, clear testing priorities, and one-click briefs. No prompts. No vibes. Just clarity.
marketfunkers gallery image
marketfunkers gallery image
marketfunkers gallery image
marketfunkers gallery image
marketfunkers gallery image
Free Options
Launch Team / Built With
Framer
Framer
Launch websites with enterprise needs at startup speeds.
Promoted

What do you think? …

Gligor Dinev
Hey there 👋 I built marketfunkers after ~10 years of making ads and leading creative teams - and slowly realizing something uncomfortable: Most teams don’t lack ideas. They lack clarity. We test more. Produce more. Prompt more. But we rarely know why something worked… or what to do next. So marketfunkers started as a selfish project: 👉 "What tool would I want before I spend another dollar testing ads?" The answer wasn’t "another AI copy generator." It was creative intelligence: understanding hooks, emotions, audiences, patterns, and turning that into clear next steps. You upload an ad. You get insight, direction, and opportunities - without prompts, fluff, or vibes. It doesn’t think for you. It thinks with you. I’m genuinely curious: - How do you currently decide what to test next? - Where does your creative process break down? Happy to answer anything, and would love honest feedback 🙏 Gligor
Cohen

Awesome idea! Will have to give it a go for an upcoming app I have in the making. Congrats on the launch though :)

Gligor Dinev

@cohenco thanks! Feel free to give it a go and lmk what you really think. Good luck on your app!

Emrah G.

@gligor_dinev Love the honesty in this — especially “teams don’t lack ideas, they lack clarity.” That hits hard.

Most AI ad tools generate more noise. Direction is rarer than content.

I’m building a SaaS product too and the hardest part is deciding what to test next, not generating variants.

How are you handling industry/context differences in analysis?

Upvoted — good luck with the launch 🚀

Gligor Dinev

@emrahg thanks and good luck! Great question. The system breaks the ad apart (so it takes all the context regarding industry/product/service/brand/market) and works based on that. Plus there's the ad dataset + performance insights + real audience insights etc.