If you're asking how to know if your ad creative is working, the short answer is: look at three layers in order — attention (did people stop and watch), engagement (did they take a signal action like a click or add-to-cart), and outcome (did it drive profitable sales or sign-ups). A creative can win at one layer and lose at another, so judging it on a single number — like "it got a lot of likes" — is how good hooks get killed and weak ads get kept alive too long.
What "working" actually means for ad creative
Before you check any dashboard, decide what job the creative is supposed to do. A top-of-funnel awareness ad and a retargeting ad for an abandoned cart are judged differently, even if they run on the same platform with the same budget. Define success in one sentence before launch, for example: "This ad should get a 3-second view rate competitive with our current top creative and a cost per add-to-cart at or below our account average." Write that sentence down. Without it, you'll unconsciously move the goalposts after you see the results.
Once the goal is defined, how to know if your ad creative is working becomes a matter of checking the right metrics at the right stage of the funnel, in the right order, over the right time window. The rest of this guide walks through each of those four variables.
The three layers of creative performance
Most ad accounts hand you dozens of metrics. You only need to read them in this order, because a problem at an earlier layer makes everything after it unreliable.
- Attention layer — Does the creative stop the scroll? Measured by hook rate (percentage who watch past the first few seconds) and thumb-stop rate (percentage who stop scrolling at all). If this number is weak relative to your other creatives, nothing downstream can save the ad.
- Engagement layer — Does the creative make people act? Measured by click-through rate, video completion rate, and for video-heavy placements, average watch time. This tells you whether the message and offer landed, not just whether the first frame worked.
- Outcome layer — Does the creative make money or drive the result you actually care about? Measured by cost per acquisition, return on ad spend, or whichever conversion event matters for your business. This is the layer that decides whether you scale the ad, pause it, or iterate on it.
A creative that gets a strong hook rate but a weak completion rate usually has a good opening and a weak middle — the promise isn't paid off fast enough. A creative with strong engagement but weak outcome usually has a mismatch between the audience it attracts and the audience that actually buys, or an offer that isn't compelling enough once people land on the page.
A step-by-step process to evaluate a new creative
- Set a minimum spend and time threshold before you look. Decide in advance, for example, that you won't judge a creative until it has had a fixed amount of spend or a fixed number of days in an active ad set. Looking too early almost guarantees you're reading noise.
- Compare against your current best, not against zero. A new ad with a decent cost per result can still be a loss if your existing top creative does noticeably better. Keep a running "champion" creative and test everything new against it.
- Check attention first. Pull hook rate or 3-second view rate. If it's meaningfully below your account average, the problem is the first few seconds — the opening shot, the first line of copy, or the pacing — not the offer or the landing page.
- Check engagement second. Look at click-through rate and completion rate together. High clicks with low completion can mean the ad over-promises early and under-delivers by the middle; low clicks with high completion can mean people watch out of curiosity but the call to action is unclear or buried.
- Check outcome last, and only after the first two layers look healthy. If attention and engagement are fine but cost per result is poor, the issue is more likely the offer, the price point, or the landing experience than the video itself.
- Segment by placement and device before you draw conclusions. A creative judged across feed, Stories and in-stream placements together can hide the fact that it's strong in one and weak in another. Break the data out before you kill or scale anything.
- Record the verdict in a shared log. Write down the creative name, the layer where it succeeded or failed, and the decision (scale, kill, iterate). This turns creative testing into a repeatable system instead of gut-feel guessing.
A simple scorecard for judging creative
A scorecard keeps decisions consistent, especially once more than one person is reviewing ads. The table below is a starting framework — treat the thresholds as something to calibrate against your own account history, not as fixed industry numbers.
| Layer | Metric to check | Healthy signal | Weak signal — what to fix |
|---|---|---|---|
| Attention | Hook / 3-second view rate | At or above your account's recent average for the same placement | Rework the first 3 seconds: faster payoff, clearer visual, louder opening line |
| Attention | Thumb-stop / scroll-past rate | Noticeably lower drop-off than your other active creatives | Test a different opening shot or pattern-interrupt before touching copy |
| Engagement | Click-through rate | In line with or above your champion creative | Clarify the offer and call to action; make sure the benefit is stated, not implied |
| Engagement | Video completion rate | Steady curve with no sharp drop-off before the offer reveal | Trim the middle section; move proof or offer earlier in the timeline |
| Outcome | Cost per acquisition / ROAS | At or better than account average over the test window | Check the landing page and offer before blaming the creative itself |
Signals that a creative has stopped working
A creative that worked well when it launched can quietly decline over time, usually because the same audience has seen it enough times that it stops feeling fresh. Watch for these patterns rather than reacting to any single day's dip:
- Frequency climbing while cost per result also climbs. This combination is the clearest sign the same people are seeing the ad repeatedly and responding less each time.
- Click-through rate trending down over several consecutive days, not just one off day, while spend and targeting stay the same.
- Comments or engagement shifting negative in tone, which can happen before the performance metrics fully catch up.
- A new creative in the same ad set consistently outperforming the old one on the same attention metric — a sign the old hook has simply been seen too many times by the available audience.
If you want a deeper framework for catching this early, see how to identify creative fatigue signals and how to track creative fatigue in paid social ads, which both go further into frequency and decay patterns than this overview can.
How long to wait before you judge a creative
There's no fixed number of hours that works for every account, budget and platform, but the underlying rule is consistent: wait until the ad has had a reasonable, representative sample of impressions and at least a few days of delivery across different days of the week. Weekday and weekend behavior can differ enough to distort a two-day read. If you're testing several creatives at once, keep budgets and start dates as close to equal as possible so you're comparing creative quality, not comparing a creative that launched on a high-traffic day against one that launched on a slow one.
Treat the first read as directional, not final. A creative that looks mediocre on day two but has a strong attention-layer score is often worth one more testing cycle before you pause it — the early numbers may simply reflect an unstable audience or a learning phase. A creative that fails at the attention layer from the very first hours rarely recovers and is usually safe to pause early.
Common mistakes when judging creative performance
- Judging on vanity metrics alone. Likes, shares and comment counts can feel like proof an ad is working, but they don't always correlate with the outcome you're actually paid to produce.
- Comparing creatives tested in different conditions. A creative tested during a sale period and one tested outside it are not a fair comparison, even if everything else is identical.
- Blaming the video for a landing page problem. If attention and engagement are strong but outcome is weak, check the destination page and offer before concluding the creative failed.
- Changing too many variables at once. If you swap the hook, the offer and the call to action in the same new version, you won't know which change moved the needle.
- Not keeping a record. Without a simple log of what was tested and what happened, teams relearn the same lessons every quarter.
A structured way to avoid most of these mistakes is to run tests inside a proper matrix rather than one-off launches. The guide on how to create a creative testing matrix covers how to isolate variables like hook, format and offer so each test actually tells you something. Pairing that with a clean creative testing calendar keeps your launch cadence consistent enough that comparisons stay fair.
Connecting creative performance to ROAS
Attention and engagement metrics tell you whether a creative is doing its job on the platform. Return on ad spend tells you whether that job is actually profitable once media cost, the offer and your margin are factored in. A creative with an excellent hook rate can still produce a poor ROAS if the product's price point doesn't support the acquisition cost, and a creative with mediocre engagement can still be worth keeping if its ROAS beats everything else in the account. For a full walkthrough of the calculation and how to attribute it correctly to a specific creative, see how to calculate ROAS on video ad creative.
Where FrameNotion fits into this process
Testing creative honestly requires having enough fresh variations to compare in the first place, which is often the real bottleneck — not the analysis. FrameNotion turns a product link into a finished 30-second vertical ad with its own hook, problem, benefit, proof, offer and call to action, so you can get a new variation into testing without waiting on a production cycle. Each ad also exports in 4:5, 1:1 and 16:9, which makes it easier to run the same creative across the placements you're trying to compare fairly.
Because FrameNotion can produce hook variants on Pro and Agency plans, you can isolate the attention layer specifically — same middle and offer, different opening few seconds — which is exactly the kind of controlled test this guide recommends. FrameNotion doesn't publish ads to ad platforms or report performance analytics; it's the creative engine, and the scorecard above is how you read the results once the ad is live. You can see sample output in the example ads or start with a free first ad on the FrameNotion pricing page.
FAQs
Frequently asked questions
How many days should I run an ad before deciding if it's working?+
There's no universal number, but aim for enough days to cover both weekday and weekend behavior, and enough spend to give each metric a representative sample rather than a handful of impressions. Treat an early read as directional and give borderline creatives one more cycle before pausing them.
What's the single most important metric for ad creative?+
There isn't one. Attention metrics like hook rate tell you if people stop and watch, engagement metrics like click-through rate tell you if the message landed, and outcome metrics like ROAS tell you if it's profitable. Reading only one layer hides where the real problem is.
My ad has a high click-through rate but a poor cost per acquisition — is the creative the problem?+
Probably not. A strong click-through rate usually means the creative did its job of getting attention and interest. A weak outcome after that points more toward the landing page, the offer, or the price point than the video itself.
How do I know if a creative is fatigued versus just underperforming from the start?+
A fatigued creative usually shows a clear decline over time — rising frequency paired with falling click-through rate or climbing cost per result across several consecutive days. An underperforming creative typically shows weak attention or engagement metrics from the very first hours of delivery.
Should I compare a new creative against my account average or against my best performer?+
Compare against your best performer, often called the champion. Beating the account average isn't the same as beating what's actually working right now, and that distinction matters when you're deciding what to scale.