Retargeting Without Annoying: The Frequency Rules I Follow
My best-performing retargeting audiences slowly rot when I let frequency drift. Here are the tiering, capping, and suppression rules I actually run in 2026.
Practical guides, ad breakdowns, product updates and original observations for ecommerce brands and performance marketing teams.
One undisclosed AI presenter cost me a rejected campaign. Now every AI video ad I ship passes the same pre-launch review — claims, disclosure, rights, platform rules.
Read the storyLatest stories
My best-performing retargeting audiences slowly rot when I let frequency drift. Here are the tiering, capping, and suppression rules I actually run in 2026.
My ad earned cheap clicks and still lost money. The leak wasn't the creative - it was the broken promise between the tap and the landing page.
A quarter of creative tests falls apart without a roadmap. Here is the planning system, tiered portfolio, and weekly cadence I use to actually stick to it.
Before budget goes out, every concept runs through a weighted scorecard of leading indicators. Here is how I build it, score it, and where it fails me.
I stopped reshooting every ad. One well-planned product shoot now feeds a batch workflow that turns a single master asset set into dozens of testable variants.
I design every feed ad for the muted majority first, then treat sound as a bonus layer. Here's the workflow my agency team runs before anything ships.
I stopped tweaking losing ads on instinct. Now I audit them frame by frame, find the retention breakpoint, and change exactly one thing per round.
I never brief an AI video ad until the first three seconds are solved. Here is the hook formula, hook wheel, and scorecard I run every time.
I do not wait for ROAS to collapse. Here is the creative-fatigue dashboard, thresholds, and rotation cadence I watch every single week.
A dashboard showing declining ROAS doesn't tell you whether to cut budget or change creative. Premature decisions based on incomplete signals waste more than inaction does.
A $20 AI-generated ad might waste more budget than a $200 human-made one if it passes quality checks but fails to persuade. Product accuracy is where AI video most often fails—and where brand damage becomes real.
Translation doesn't equal localization. A video that works in Los Angeles may fail in Jakarta if it ignores payment preferences, holiday calendars, and cultural humor.
Advantage+ audience expansion works like a double-edged sword: it finds customers you couldn't reach manually, but it also spends budget on people unlikely to convert.
You're not buying videos. You're buying rights, authenticity, and audience access—and the tradeoffs between them determine which approach makes sense for your brand.
TikTok's CPM is $8-12. Instagram's is $15-25. But lower cost doesn't mean better value if your conversion rate is half as good. Here's a merchant's framework for choosing between them.
A dashboard conversion count is not proof your ads caused it. An attributed order is not the same thing as an order that would disappear without the ad. Here's how to measure true incremental value at any budget level.
A $50 AI-generated ad might waste more budget than a $500 human-made one if it passes quality checks but fails to persuade. Here's how to calculate actual usable-output cost including retries, fixes, and QA.
A cleaning cloth sold $430k in one month on TikTok using before-after creative. Before you copy the format, understand what's observable versus inferred—and how to test systematically without blowing budget.
Meta's Advantage+ Shopping Campaigns automate targeting, creative, and budget pacing, but that automation hides what is actually working. Here is what merchants can control, what they cannot, and a practical framework for making informed ASC decisions in 2026.
A simple 20-second egg poacher demo shows how a strong performance ad can sell without feeling like an ad. We break down its hook, product proof, pacing and CTA.