Track ad creatives by giving every ad a name that records its concept, format and hook, then comparing cost per result, ROAS, click-through rate and video watch rates for each creative in Ads Manager. Judge each creative on the result you pay for, and use the earlier metrics to explain why it won or lost.
TL;DR
- Name every ad with the same pattern, such as concept, format, hook and date, so you can group results later.
- Judge creatives on cost per result and ROAS. Use hook rate, hold rate and CTR to explain why.
- Compare creatives only within the same ad set, audience and dates.
- Decide how much each creative may spend before you call a test.
- Watch for Creative limited and Creative fatigue in the Delivery column.
- Agencies: use one naming pattern, one column set and one report layout for every client.
Creative metrics, defined plainly
| Metric | What it tells you | How it's calculated |
|---|---|---|
| Hook rate | Whether the first seconds stop the scroll | 3-second video plays ÷ impressions |
| Hold rate | Whether people keep watching | ThruPlays ÷ 3-second video plays |
| Link CTR | Whether the ad earns a click | Link clicks ÷ impressions |
| Cost per result (CPA) | What each conversion costs | Spend ÷ results |
| ROAS | Revenue for each dollar spent | Purchase value ÷ spend |
| Frequency | How often the same people see the ad | Impressions ÷ reach |
Hook rate and hold rate are not standard Ads Manager columns. Add them as custom metrics, or work them out in a spreadsheet from the columns they use.
How do I set up Ads Manager to compare creatives?
- Open Ads Manager at the Ads level, and filter to one campaign and a fixed date range.
- Customize the columns: results, cost per result, purchase ROAS, link CTR, CPM, frequency, 3-second video plays and ThruPlays.
- Save the column set as a preset, so every review uses the same view.
- Use the Breakdown menu to split results by placement or age when a creative wins in one place only.
- For ads where Meta mixes several assets, such as dynamic creative, break results down by asset where Meta offers it.
- Keep the attribution setting the same in every comparison.
Let Madgicx's AI do the work
Its AI marketer finds what's working and prepares the next move. You approve.
Try for Free ($0 Trial)What naming convention should I use for ads?
Use one pattern for every ad, with each part separated by an underscore. For example: Concept_Format_Hook_Creator_Date, written as SocialProof_Video9x16_PriceHook_Dana_2026-10.
Good names let you export the Ads level and group results by concept or hook in a spreadsheet. That turns single ads into lessons, such as "price hooks beat problem hooks". Name ads before launch; renaming later breaks your history.
Which creative metrics matter most?
Two kinds of metric do different jobs.
- Outcome metrics decide. Cost per result and ROAS tell you whether a creative makes money. Judge winners on these.
- Diagnostic metrics explain. Hook rate, hold rate, CTR and CPM tell you where a creative breaks.
Read them together. A high hook rate with a low hold rate means the opening works but the rest doesn't. A high CTR with poor ROAS points to the page or the offer, not the ad. A low CTR with a good CPA can still be a winner, because it attracts fewer but better clicks. Our answer on improving Facebook ad creatives covers what to change in each case.
How do I know when a creative is tired?
Meta flags tired creative in the Delivery column for ad sets with a single creative. It shows Creative limited when cost per result is above your past ads but under twice as much, and Creative fatigue at twice or more.
How to check: chart frequency, CTR and cost per result for each ad over the last 14 days. Rising frequency with falling CTR and rising cost is the usual sign of ad fatigue.
The fix: launch a materially different creative, not a small edit, and keep the original running while the new one starts. Our guide to detecting creative fatigue goes deeper.
How do I test creatives fairly?
- Test one change at a time: the hook, the format or the offer.
- Put the creatives in the same ad set, so they share an audience, budget and dates.
- Set the spend each creative gets before you judge it, and write it down.
- Run the test for at least 7 days, so results cover every weekday and late conversions.
- Keep the winner, log the lesson by concept, and test the next idea against it.
For a cleaner split, run an A/B test in Ads Manager. Our guides cover ad testing basics, creative testing tools, automated testing with AI and how to choose an ad testing platform.
How can agencies track creative performance across clients?
Agencies need the same view for every account, so lessons carry from one client to the next.
- Use one naming pattern for every client, with a client code at the start.
- Save one column preset and use it in every account.
- Send each client the same report layout, with a creative section showing top ads by spend, cost per result and ROAS.
- Hold a weekly creative review: winners, tired ads and the next tests.
- Keep a shared library of winning concepts by industry, without client-specific details.
If you're choosing software for this, our comparison of creative performance platforms covers the options. Our Facebook ads reporting guide shows how to build the client report.
Where Madgicx fits
Madgicx analyzes creative performance across your Meta ads, so you can see which concepts and formats win. A fatigue chart plots each ad's frequency against its cost per click, which shows tiring ads before they waste much spend. When you need new tests, it generates image ads. For agencies, its cross-channel reports and dashboards can be shared with each client by link or PDF. Its AI chat explains what changed in an account and finds wasted spend.
FAQ
What is a good hook rate for Facebook video ads?
A good hook rate is one that beats your own account's average for the same format and placement. Hook rates vary by industry and audience, so compare a new video with your past videos, not a general number.
What is the difference between hook rate and hold rate?
Hook rate measures the share of impressions that become 3-second video plays, and hold rate measures how many of those viewers keep watching. A high hook rate with a low hold rate means the opening works but the rest doesn't.
How long should I run a creative test on Facebook?
A creative test on Facebook should run for at least 7 days, so results cover every weekday and late conversions. Decide the spend each creative gets before you start, and judge after it's reached.
Should I judge creatives by CTR or ROAS?
Judge creatives by ROAS or cost per result, because those measure the outcome you pay for. Use CTR to explain the result: a high CTR with poor ROAS points to the page or the offer.
What does Creative fatigue mean in Ads Manager?
Creative fatigue in Ads Manager means an ad's cost per result has reached twice or more that of your past ads. Meta shows Creative limited when the cost is higher but under twice as much.
How can agencies track creative performance across clients?
Agencies can track creative performance across clients with one naming pattern, one column preset and one report layout for every account. That makes concepts comparable between clients and cuts reporting time.
Related
- Ad tech platforms for creative performance, compared in depth
- Answers: How to scale ad creative production with AI, How to optimize Facebook ads for conversions
- Glossary: Ad fatigue, Dynamic creative optimization
Sources
- Creative fatigue recommendations in Meta Ads Manager, Meta Business Help Center
- About attribution settings, Meta Business Help Center
