Adscalr
Ad intelligence for people who buy media. It turns competitor activity, audience language and performance data into the next ad to ship.
Status: web platform in private preview, in daily use for the media buying we do ourselves.
What it does
Six systems, and the output of each one is the input of the next.
Competitor ads, decoded daily
The Meta, TikTok and Google ad libraries are pulled each day. Vision AI breaks a static ad into about twenty structured fields with design, copy and strategy scores from 1 to 10. Video is read frame by frame for hook, pacing and CTA placement.
A market map on five axes
Competitors are clustered on price, quality, speed, trust and innovation, so you can see which message nobody in your category is making.
Audience language from five sources
Reddit, Amazon reviews, the App Store, Google Play and forums, captured as the exact phrasing people used. Four to six personas per project, each backed by at least three real quotes and mapped onto the five awareness stages.
Creative with an attention prediction
Static ads, UGC concepts and motion storyboards in four native formats. Statics get a saliency prediction from DeepGazeIIE, a model built on eye tracking data, and a copy critic scores them against six direct response principles.
Testing with real statistics
Composite scoring across six metrics with Bayesian shrinkage, Benjamini-Hochberg false discovery control at q = 0.10 in pattern mining, and Thompson sampling to pick what to test next.
Automation with four safety catches
Rules fire on eight metrics and can only pause or kill. A learning phase lockout, a ROAS floor, a minimum of three active ads under CBO and a frequency ceiling stop a rule from killing a winner. Every kill is reversible for 24 hours, and pacing is checked every five minutes.
Automatic up to the point where money moves
What runs unattended, and where a person signs off.
You make the last click
Nothing reaches Meta without a person approving it, and no creative is ever auto rejected or auto rewritten.
Budget plans stay proposals
Budget and language targeting come as recommendations you accept or ignore, and every estimated value is labelled as an estimate.
Video production is on the roadmap
UGC concepts and storyboards are real output today. Turning them into finished video is not built.
Where it stands
478 API routes and 47 database tables across three ad platforms, built end to end in about three months. The platform is in private preview, so access happens through a conversation for now.