JEV makes competitor research feel like a cheat code We fed it Resilia’s ad library and it classified 1,891 ads in 19 seconds for $0.12 Customer journey stage, ad style, and a full account deep dive Coming soon to the maxfusion MCP

@lukealittle you gotta get on jev

Try maxfusion MCP to scale your brand or agency

ha 🔥

Indeed :D

how is jev analysing images? it can’t do that

Annotation of all the videos as it isn't images from Gemini flash and it processes all the annotations, pretty cool huh?

👀👀

Insane efficiency 1,891 ads classified in 19 seconds for $0.12 is a literal cheat code for competitor research!

@andrewdsouza hey Andrew, take a look. Maxfusion’s competitor ad research via MCP for ad teams and agencies. I can help them get in front of the people buying creative intelligence.

insane speed with better results

that speed and cost are impressive, can’t wait to see it in action!

The speed is impressive, but label quality decides whether the output is useful. Ori, can JEV show confidence scores and let users correct categories before it creates the account summary?

JEV just made competitor research insanely fast and affordable.

It literally killed the competition

It didn’t just beat the competition. It completely changed the game.

100% truth right here

1,891 ads in 19 seconds is seriously wild.

Call me impressed.

Impressed :)

wow that's insane speed and cheap pricing, looks like a major shift for ad research

Destroyed everyone else tbh

I need to try this with creator search!

Can’t wait to use this in maxfusion, I currently have separate processes for this that my Claude Code has to manually manage alongside maxfusion. Doing this natively will be crazy useful 👀

Classifying thousands of competitor ads in seconds for twelve cents makes manual marketing research look like a complete waste of headcount.

The useful part is not just speed, it is turning a messy ad library into a comparable dataset with visible cost per decision. That makes the MCP feel like an analyst with a repeatable method, not a search wrapper. How are you validating classification quality across verticals?

jev doesn't understand image as input? am i missing something here?

Through gemini annotation and its videos

Thanks @OriSilver interesting. I had a few ideas also i think once it understand images and video this will be a must 100%

This is seriously impressive! 1,891 ads in just 19 seconds

tbh it was less

The $0.12 looks clean until you spot-audit 50 outputs. Journey-stage labels are where I kept finding the model applying its own stage definitions instead of the ones I had given it.