Open the CSV in Excel, or import it into Google Sheets and choose comma-separated values. Save your working copy in a controlled location. The CSV contains the rows below and space for your notes; it has no macros, formulas or live account connection. The separate Excel workbook contains editable example inputs and formulas that update the forecast.
How to use the template
Set the review date and thresholds
On the Settings tab, type the Monday you run the review, the days it takes you to make a new ad and to test one, and how many winner slots you want filled. Defaults: WATCH at 1.3x an ad's first-week cost per result, TIRED at 2x (Meta's Creative fatigue line), early signs at CTR down 30% with frequency 3.5+.
Add each winner to the Tracker
One row per live ad you care about: launch date, its cost per result and CTR in its first 7 days, its cost per result two weeks ago, and last 7 days' cost per result, CTR and frequency from Ads Manager. Formulas give a status, the weekly rise in cost, a projected tired date, a 'brief by' date and the action for this week.
Fill the refresh calendar
Every tracked ad appears with its dates. Pick a refresh type (new hook, format, angle or offer), write the replacement idea, an owner and the live date. Rows go amber when a brief is due within 7 days and red when it is overdue; the top line counts replacements to start this fortnight.
Log past ads once a month
On the Lifespan tab, log launch and stop dates and whether each ad became a winner. The tab works out your hit rate, how long your winners last, and how many new ads to test each week to keep your winner slots full.
Preview the worksheet
Every row shown here is included in the download. Scroll the table horizontally on a small screen.
| Rule | Condition | Default |
|---|---|---|
| Baseline | Cost per result and CTR in the ad's first 7 days | Its own numbers |
| FRESH | Cost per result below the WATCH multiple of its baseline | Below 1.3x |
| WATCH (cost) | Cost per result at or above the WATCH multiple | 1.3x |
| WATCH (early signs) | CTR down from the first week AND frequency high | 30% down and 3.5+ |
| TIRED | Cost per result at or above the TIRED multiple | 2x (Meta's Creative fatigue line) |
| Weekly rise | Square root of (last 7 days ÷ two weeks ago) minus 1 | Projected only above 2% a week |
| Projected tired date | Review date + 7 × ln(TIRED line ÷ current cost) ÷ ln(1 + weekly rise) | From the review date you type |
| Brief by | Projected tired date − days to make − days to test | 7 + 4 days |
| Refresh type | How different the replacement is | New hook, format, angle or offer |
| Swap | Test the replacement before retiring the original | Keep the tired ad at the same budget |
| Winners needed per week | Winner slots ÷ median winner lifespan in weeks | From your Lifespan log |
| New ads to test per week | Winners needed per week ÷ hit rate | From your Lifespan log |
Worked review example
Fictional coffee brand, review on 16 November 2026: of 8 live ads, 2 are TIRED (2.05x and 2.19x their first-week cost per result), 2 are WATCH (one at 1.39x, one on early signs: CTR down 35% at frequency 3.8) and 4 are FRESH. The UGC unboxing ad is rising 14.7% a week, so it is projected to be tired on 5 December and its replacement must be briefed by 24 November.
From 20 logged past ads, 5 became winners (25% hit rate) and the median winner lasted 38 days. To keep 6 winner slots full the brand needs 1.1 new winners a week, which means testing about 4.4 new ads a week.
What this template does and does not establish
Uses platform-reported results, which depend on Meta's attribution setting; a rising cost per result can also come from tracking, site or seasonal changes. The 2x TIRED line mirrors Meta's label; the other thresholds are starting points, not Meta rules or benchmarks. The projected date extrapolates two weeks of change and moves every week. Lifespan maths is shaky with fewer than about 20 logged ads. No account connection, macros or automatic changes.
Primary references
- Meta: About creative fatigue recommendations
- Meta: Significant edits and learning phase
- AdSpyder: ad fatigue benchmarks (May 2026 archive)
- Meta Ad Library: Grüns live ads
