Creative fatigue
The point where an ad stops performing because people have seen it too often, or have stopped caring about it. It isn't one number. It's a pattern across five signals, and the pattern tells you what to fix.
Listed roughly in the order they tend to move. Cost and return are usually the last to show it.
What is creative fatigue?
Every ad has a shelf life. At launch it reaches fresh people who react to it. Over time it either runs out of new people to show, or the people it reaches stop responding. Either way, cost per result climbs.
Those are two different problems, and "fatigue" covers both. Telling them apart is the whole point, because one needs a new audience and the other needs a new creative.
How it's measured
Each signal answers one question. Read together, they show where an ad is in its life.
| Signal | Question it answers | Full definition |
|---|---|---|
| New Reach % | Is the ad still finding people who haven't seen it? | New Reach |
| Lifetime frequency | How many times has the average person seen it? | Frequency |
| Hook rate | Does the opening still stop the scroll? | Hook rate and hold rate |
| CTR | Are people still curious enough to click? | Link clicks ÷ impressions |
| CPA and true ROAS | Is it still paying for itself? | True ROAS |
Tracking it daily
Fatigue is about change over time, so each ad needs its own history and its own baseline.
- 1Snapshot dailyLifetime reach and impressions each day, so New Reach and lifetime frequency can be computed.
- 2Set a baselineUse the ad's own first few stable days for hook rate, CTR and CPA. Compare it to itself, not to other ads.
- 3Smooth over 3 daysSingle days are noisy. Flag fatigue when the 3-day trend crosses your thresholds.
Worked examples
Example 1: One ad over 14 days
An illustrative prospecting video ad on a steady budget. Selected days shown.
| Day | New Reach % | Lifetime frequency | Hook rate | CTR | CPA | Stage |
|---|---|---|---|---|---|---|
| Day 1 | 100% | 1.15 | 31% | 1.6% | ₹420 | Learning |
| Day 3 | 82% | 1.38 | 32% | 1.7% | ₹390 | Healthy |
| Day 5 | 70% | 1.62 | 30% | 1.6% | ₹405 | Healthy |
| Day 7 | 58% | 1.90 | 27% | 1.4% | ₹460 | Saturating |
| Day 9 | 44% | 2.25 | 24% | 1.2% | ₹540 | Saturating |
| Day 11 | 31% | 2.70 | 20% | 1.0% | ₹650 | Fatigued |
| Day 14 | 22% | 3.30 | 16% | 0.8% | ₹820 | Fatigued |
- Day 7New Reach % drops below 60% while CPA is only ₹40 above its best. This is the early warning, and the cheapest moment to prepare a refresh.
- Day 9Hook rate and CTR start following. Repeat viewers are scrolling past.
- Day 14CPA has roughly doubled from its best day. Waiting for CPA to confirm fatigue cost about a week of inflated spend.
Example 2: Same CPA rise, different causes
Two illustrative ads whose CPA rose about 40% over two weeks. The signals point to opposite fixes.
What to do about it
Match the fix to the pattern. The thresholds here are starting points for prospecting ads; set your own from your best-performing creatives.
Common mistakes
- Waiting for CPAIt moves last. Act on New Reach and hook rate trends instead.
- One fix for every caseNew creatives won't fix a saturated audience, and new audiences won't fix a tired creative.
- A fixed shelf life"Refresh every 2 weeks" ignores budget and audience size. Let the signals decide.
- Comparing ads to each otherFatigue is decline from an ad's own baseline, not a gap to another ad.
- Trusting platform ROASOver-attribution can hide fatigue. Use true ROAS for the final call.
Frequently asked questions
How long does a Meta ad last before it fatigues?
There's no fixed number. A big budget on a narrow audience can fatigue in days; a modest budget on a broad audience can run for months. Watch New Reach % to see where yours is.
Can a fatigued ad come back?
Often, yes. After a break, or shown to a new audience, a creative that saturated one group can work again. A creative that wore out because the message stopped landing usually doesn't.
Does this apply to static image ads?
Yes, minus hook rate. For static ads, lean on New Reach %, frequency and CTR.
Should I pause or refresh?
If CTR held up and only New Reach fell, the creative still works: move it to a new audience. If hook rate and CTR fell, refresh. If both, pause.
Related metrics
Catch fatigue a week before CPA does
BooleanMaths tracks New Reach, frequency and creative signals for every Meta ad daily, next to true ROAS, so you know whether to refresh the creative or find a new audience.
Creative fatigue
The point where an ad stops performing because people have seen it too often, or have stopped caring about it. It isn't one number. It's a pattern across five signals, and the pattern tells you what to fix.
Listed roughly in the order they tend to move. Cost and return are usually the last to show it.
What is creative fatigue?
Every ad has a shelf life. At launch it reaches fresh people who react to it. Over time it either runs out of new people to show, or the people it reaches stop responding. Either way, cost per result climbs.
Those are two different problems, and "fatigue" covers both. Telling them apart is the whole point, because one needs a new audience and the other needs a new creative.
How it's measured
Each signal answers one question. Read together, they show where an ad is in its life.
| Signal | Question it answers | Full definition |
|---|---|---|
| New Reach % | Is the ad still finding people who haven't seen it? | New Reach |
| Lifetime frequency | How many times has the average person seen it? | Frequency |
| Hook rate | Does the opening still stop the scroll? | Hook rate and hold rate |
| CTR | Are people still curious enough to click? | Link clicks ÷ impressions |
| CPA and true ROAS | Is it still paying for itself? | True ROAS |
Tracking it daily
Fatigue is about change over time, so each ad needs its own history and its own baseline.
- 1Snapshot dailyLifetime reach and impressions each day, so New Reach and lifetime frequency can be computed.
- 2Set a baselineUse the ad's own first few stable days for hook rate, CTR and CPA. Compare it to itself, not to other ads.
- 3Smooth over 3 daysSingle days are noisy. Flag fatigue when the 3-day trend crosses your thresholds.
Worked examples
Example 1: One ad over 14 days
An illustrative prospecting video ad on a steady budget. Selected days shown.
| Day | New Reach % | Lifetime frequency | Hook rate | CTR | CPA | Stage |
|---|---|---|---|---|---|---|
| Day 1 | 100% | 1.15 | 31% | 1.6% | ₹420 | Learning |
| Day 3 | 82% | 1.38 | 32% | 1.7% | ₹390 | Healthy |
| Day 5 | 70% | 1.62 | 30% | 1.6% | ₹405 | Healthy |
| Day 7 | 58% | 1.90 | 27% | 1.4% | ₹460 | Saturating |
| Day 9 | 44% | 2.25 | 24% | 1.2% | ₹540 | Saturating |
| Day 11 | 31% | 2.70 | 20% | 1.0% | ₹650 | Fatigued |
| Day 14 | 22% | 3.30 | 16% | 0.8% | ₹820 | Fatigued |
- Day 7New Reach % drops below 60% while CPA is only ₹40 above its best. This is the early warning, and the cheapest moment to prepare a refresh.
- Day 9Hook rate and CTR start following. Repeat viewers are scrolling past.
- Day 14CPA has roughly doubled from its best day. Waiting for CPA to confirm fatigue cost about a week of inflated spend.
Example 2: Same CPA rise, different causes
Two illustrative ads whose CPA rose about 40% over two weeks. The signals point to opposite fixes.
What to do about it
Match the fix to the pattern. The thresholds here are starting points for prospecting ads; set your own from your best-performing creatives.
Common mistakes
- Waiting for CPAIt moves last. Act on New Reach and hook rate trends instead.
- One fix for every caseNew creatives won't fix a saturated audience, and new audiences won't fix a tired creative.
- A fixed shelf life"Refresh every 2 weeks" ignores budget and audience size. Let the signals decide.
- Comparing ads to each otherFatigue is decline from an ad's own baseline, not a gap to another ad.
- Trusting platform ROASOver-attribution can hide fatigue. Use true ROAS for the final call.
Frequently asked questions
How long does a Meta ad last before it fatigues?
There's no fixed number. A big budget on a narrow audience can fatigue in days; a modest budget on a broad audience can run for months. Watch New Reach % to see where yours is.
Can a fatigued ad come back?
Often, yes. After a break, or shown to a new audience, a creative that saturated one group can work again. A creative that wore out because the message stopped landing usually doesn't.
Does this apply to static image ads?
Yes, minus hook rate. For static ads, lean on New Reach %, frequency and CTR.
Should I pause or refresh?
If CTR held up and only New Reach fell, the creative still works: move it to a new audience. If hook rate and CTR fell, refresh. If both, pause.
Related metrics
Catch fatigue a week before CPA does
BooleanMaths tracks New Reach, frequency and creative signals for every Meta ad daily, next to true ROAS, so you know whether to refresh the creative or find a new audience.