Ad Creative Fatigue Crossover Calculator

How to read this number

Every figure on this page is arithmetic on the numbers you typed in. Nothing is averaged against a published benchmark, because there is not one that applies to your account. The headline number is your weekly frequency: impressions divided by reach, then divided by 4.33 so the monthly figure reads as a weekly one. If it says 1.1, the average person in your reach was shown one of your ads a little over once a week. If it says 4, they saw it roughly four times a week, which is a very different experience to being introduced to an ad once.

The threshold is the one input this tool cannot set for you. There is no universal benchmark for the number of impressions at which a given creative starts to decay. It changes with the category, the offer, the season, the audience size you are working within, and how the ad is framed. The only reliable threshold is the one you measure in your own ad account, and the section below shows how to find it. The slider here defaults to 5 purely so the box is not empty on load. Treat it as an assumption you are making out loud, and change it once you have your own data.

Weeks until your threshold assumes reach and impressions keep growing at the same rate you are running now. Reach plateaus long before spend does, so in practice frequency climbs faster than the straight line this tool draws. Read the number as a floor, not a schedule.

A worked example with these defaults

Take the starting numbers: $4,000 a month in spend, 420,000 impressions, 90,000 people reached, 6,300 clicks, 126 conversions, and a creative that launched five weeks ago. Frequency is 420,000 divided by 90,000, which is 4.67 impressions per person per month. Spread across a week that is 1.08 impressions per person per week. Click through rate is 6,300 divided by 420,000, or 1.5%. Conversion rate is 126 divided by 6,300, or 2%. CPM is $4,000 over 420,000 impressions times a thousand, or $9.52. Cost per conversion is $4,000 over 126, or $31.75. Set the threshold at 5 and you have roughly 15.8 weeks of headroom at the current rate of reach growth, though the creative is already five weeks old. Frequency is not the problem here. At five weeks old with under two impressions a week per person, if performance is soft the evidence points at targeting, the offer or the landing page, not at repetition.

Now change one input. Push reach to 40,000 people and leave impressions alone. Frequency jumps to 10.5 per month, which is 2.4 per week, and the weeks to your threshold collapse to about 4.6. Nothing about the creative changed. The audience got smaller and the same creative got busier. That is the whole mechanism this calculator exists to make visible.

Why frequency climbs on its own

The algorithm will keep showing your ad to the people most likely to act on it, and it will keep spending your budget until you stop it. What it will not do is manufacture new people. Reach hits a ceiling defined by how big and how well defined your addressable audience is, and it approaches that ceiling as a curve, fast at first and then flat. Impressions do not flatten. If you hold a dollar amount of spend steady against a plateaued reach, the only way to place the same number of impressions is to place them more often against the same smaller pool of people.

So frequency climbs on essentially every account that has been running long enough, and it climbs without anyone making a decision. The accounts that look immune to it are usually the ones where reach is still expanding fast, which in a small niche means reach is not yet near its ceiling. When a founder says performance fell off after six weeks and nothing changed, this arithmetic is usually the first thing worth checking.

Fatigue versus a broken offer or landing page

These two look identical in a summary report and have opposite fixes, so separating them matters more than it sounds. Creative fatigue is a frequency problem: the same asset keeps meeting the same people, so recognition builds faster than persuasion. A broken offer or landing page is a relevance problem: people see it once, or twice, and still do not act. Refreshing the creative in the second case costs money and changes nothing.

There are three ways to tell them apart using data you already have. First, watch click through rate against frequency. Fatigue shows up as a rate that was stable, then slid, then stabilised at a lower level while frequency kept rising. A landing page problem leaves click through rate alone and only moves what happens after the click. Second, watch when the decline started relative to the launch date. Fatigue tracks the calendar of the asset, and it resets when you swap the asset. An offer problem tracks the calendar of the market, a competitor's price move, a stock-out, a seasonal shift, and it does not care what creative you run. Third, look at whether other assets on the same account are declining at the same time. Account wide movement points away from any single creative and toward the offer, the audience or the landing page.

The cheapest real test is to run genuinely new creative against the same audience for one to two weeks and watch whether the same people convert at the old rate. If they do, fatigue was the cause. If they do not, the offer or the landing page is where the work is. The Store Funnel Health Scorecardruns the same kind of read on the rest of the funnel once you have ruled fatigue out.

How to find your own real threshold

You find it in your own ad account, week by week, and nowhere else. Pull frequency and conversion rate for the same ad set on the same weekly basis. Plot them against each other. Early on, conversion rate should be broadly flat while frequency climbs, because the creative is still pulling its weight and the extra exposures are cheap. The crossover is the point where conversion rate stops holding its level and starts to fall while frequency is still climbing. That frequency, on that account, at that spend, is your threshold. It is not a transferable number, and it will not be the same six months from now if your audience or offer has shifted.

Two practical notes. Keep creative age out of the way while you do this, because a new asset resets both the frequency clock and the response, and you want to isolate one thing at a time. And do it over at least six to eight weeks if the account is small enough to allow, because a single bad week will convince you of a threshold that is not real. Once you have a number, set the slider here to it and the rest of this page becomes a live read on how much room you have left.

What to do next

If the readout says fatigue is unlikely, do not refresh the creative yet. Spend the next diagnostic hour on the post click path instead: the offer, the landing page load, the shipping and returns policy, the checkout. High traffic with a click through rate that holds and a conversion rate that does not is a page problem wearing a creative costume. Run the Break-Even ROAS Calculator to confirm what a conversion needs to be worth before you go chasing a percentage point on the ad side, and theStore Funnel Health Scorecard to see whether the drop is concentrated in one step.

If the readout says fatigue is the likely cause, treat the creative refresh as a scheduled maintenance task rather than a rescue. The point of measuring a threshold in advance is that you swap the asset when frequency approaches it, while the replacement is still learning, rather than after a week of bad numbers. Decide now what the next asset will be, and if the Ad Hook Generator gets you through the first draft faster, the gap between the old asset burning out and the new one performing gets shorter.

The four things that actually reset frequency

There are only four levers, and it is worth being blunt that none of them is "bid lower". First, new creative. A genuinely different asset resets recognition, which is usually the largest single move available. Second, a broader or adjusted audience. Exclusions, a wider age or interest range, or dropping a lookalike that has been saturated all put the same money in front of different people. Third, a new placement mix. Adding a placement you have never run, or moving weight out of the one that has been churning, changes who is seeing the ad without changing the ad. Fourth, lower spend for a period. This is the one nobody wants to say out loud, and it works: spend less, reach grows slower, frequency stalls, and the audience gets a break. The account does not perform better during that window, it performs less badly, which is a different thing and worth knowing before you choose it.

What does not reset frequency is turning up the budget, duplicating an ad set, or editing the copy in place. All three leave the audience and the asset essentially unchanged, so the same people keep seeing the same ad at the same rate. If your account is climbing in frequency and the fix on the list you actually want is not on it, that is the honest read, and it usually means the audience is too small for the spend you are running rather than the creative needing another pass.