All free tools

Meta Creative Fatigue Checker

Compare two reporting windows and see how CPM, click response and purchase conversion changed your CPA.

Loading tool…

Find what changed before replacing the ad.

Compare two equal reporting windows to see how auction cost, click response and purchase conversion changed your cost per purchase. Rising frequency belongs beside those numbers: it helps you decide what to investigate, while the funnel calculation shows the size of each change.

Use the same measurement basis

Choose the same ad or group of ads, audience scope, currency, attribution basis and number of days. Allow both windows enough time for purchases to be reported. Outbound CTR is outbound clicks divided by impressions. Click-to-purchase CVR is click-attributed purchases divided by those outbound clicks. Frequency is impressions divided by reach within each window.

For this calculation, use the same click denominator in CTR and CVR. Store session conversion rate and purchases attributed to views use different denominators. If your starting point is landing page views, use the Meta Ads Performance Doctor to review that funnel.

The three factors behind CPA

Modelled CPA = CPM ÷ (1,000 × outbound CTR × click-to-purchase CVR). Convert each percentage to a decimal before calculating. The ratio between the two CPAs is current CPM ÷ previous CPM, multiplied by previous CTR ÷ current CTR, multiplied by previous CVR ÷ current CVR.

Each factor above 1 pushes CPA up. A factor below 1 pushes it down. Multiply the factors together to reconcile the whole change. Adding their percentage changes gives the wrong result. Frequency stays outside this identity because impressions, clicks and purchases already capture the funnel.

A worked example

In this illustrative example, the first seven days have a $20 CPM, 2% outbound CTR and 3% click-to-purchase CVR. Modelled CPA is $33.33. The next seven days have a $24 CPM, 1.5% CTR and 2.5% CVR, producing a $64 CPA.

The CPM factor is 1.20, the click-response factor is 1.3333 and the purchase-conversion factor is 1.20. Together they make 1.92: CPA rose 92%. Frequency also rose from 2 to 3. The result identifies three separate pressures to review before making a creative decision.

Choose the next investigation

When CPM changes most, inspect placement, market and audience mix. When click response weakens, compare the hook, offer and creative against the same audience. When purchase conversion falls, review stock, traffic quality, product pages and checkout. Use the frequency guide for the reporting context and the creative testing pipeline to plan a replacement test.

Creative fatigue questions

What frequency is too high for Meta ads?

Judge frequency over a defined window alongside click response, purchase conversion and acquisition cost. Prospecting and remarketing can reach the same people at different rates. Compare your own consistent windows before setting a review point.

Does falling CTR prove that an ad is fatigued?

Falling CTR tells you click response weakened. Check whether audience, placement, season, offer or delivery mix also changed. A creative test can help separate those explanations.

Why does the tool use modelled CPA?

The calculation reconstructs CPA from the three entered funnel metrics. It will match spend divided by purchases when the metrics use exactly the same impressions, clicks, purchases and reporting basis. Rounded inputs can create small differences.

Can I compare windows with zero purchases?

Keep the zero conversion rate in the input. The tool leaves that window's CPA and the combined multiplier unavailable. Review raw counts and reporting maturity; the zero purchase calculator can assess a zero-count sample under an assumed CVR.

How much data makes the comparison reliable?

Read the counts behind the rates and allow for conversion lag. This tool compares point estimates. A small number of purchases can move a rate sharply, even when the apparent percentage change is large.

This descriptive comparison cannot establish causation or statistical significance. References checked 16 November 2026: Meta Ads Insights reporting and Meta bidding and optimisation.