On this page
A winning ad in iGaming has a shelf life. It finds its audience, the cost per registration looks great for a week or two, then the same people see it for the fifth time and stop clicking. If the replacement is still in a brief somewhere, you spend the next two weeks overpaying for a dying ad.
This is the routine we use to keep a pipeline of fresh creatives moving, plus the part most guides skip: how to tell real creative fatigue from the three other problems that look exactly like it.
What creative fatigue is, according to the platforms
Meta's definition is short: creative fatigue occurs when an audience has seen the same creative too many times (Meta Business Help Center). Ads Manager turns it into delivery statuses for ad sets with a single creative:
TikTok describes the symptoms in its auction best practices: click-through falling while frequency rises, reach shrinking, cost per action climbing. Its creative best practices recommend refreshing when results show a consistently declining trend, and running 3–5 creatives per ad group. Its Smart Creative tool even automates the swap: it watches new videos for 3–5 days on a small slice of budget, then pauses fatigued ones and brings in queued replacements (TikTok Smart Creative).
One useful counterweight: Kantar's analysis of tracked ads found that true wear-out of creative quality is rare, while high frequency across channels does cause irritation (Kantar). In performance terms: the ad usually isn't "worn out", your audience is. That matters, because sometimes the fix is a wider audience, not a new ad.
Fatigue or something else? A diagnostic
Before you brief a single new creative, check which problem you actually have:
| What you see | Likely cause | What to do |
|---|---|---|
| One ad: frequency up, CTR down, CPM flat, cost per registration up | Creative fatigue | Launch a materially different variant; widen the audience; lower the old ad's budget share |
| Every ad: CPM up at the same time | Auction (season, a big sports weekend, a competitor's launch) | Don't kill creatives; adjust bids or budgets and wait it out |
| CTR stable, conversion rate down | Page, offer or tracking | Check page speed, offer changes, the postback and the pixel |
| Everything drops overnight | Tracking or policy | Check events, disapprovals, limited delivery and account status |
| New ads never get out of learning | Structure | Too many ad sets for the budget; consolidate |
Compare each ad to its own first full week, not to the account average. Our rule of thumb: when click-through is down 25–30% from that baseline while frequency keeps climbing and CPM hasn't moved, the replacement should already be live, not in a brief.
The weekly refresh system
The point of a system is that refreshing stops being an emergency. Same days, same steps, every week:
| Day | Step | Output |
|---|---|---|
| Monday | Read last week: each ad against its own baseline, with the diagnostic above | Every ad tagged scale / hold / iterate / retire |
| Tuesday | Brief: iterations of winners plus new concepts | A one-page brief per concept |
| Wednesday | Produce: statics, cut-downs, sizes, languages | Variants, named and logged |
| Thursday | Compliance check per market, then launch | New ads live before the weekend peak |
| Fri–Sun | Learn: no edits while the new ads gather data | Clean data for Monday |
Launching on the same weekday matters more than which weekday: it keeps week-on-week comparisons fair. We like Thursday so new ads collect data through the weekend, when many gambling audiences are most active. Your data may say otherwise.
What goes into each week
We split new work roughly 70 / 20 / 10 (an opinion, not a platform rule):
- 70% iterations of winners. Same concept, new hook. The first two seconds of a video and the headline of a static carry most of the difference, so iterate there first.
- 20% new angles. A different reason to play: the feature, the character, the speed, the bonus round.
- 10% wild cards. A format you haven't tried: a playable ad, a different aspect ratio, a gameplay clip with no voice-over.
"Materially different" is Meta's phrase, and it's the right bar. A new background colour is not a new creative; the audience recognises the ad by its hook and its hero, so change those.
Name it so the data makes sense
Fatigue analysis dies in messy reporting. Every variant gets a name that encodes what changed:
2026w42_VMP_hookB_v3_9x16_EN
Week, concept code, hook, version, format, language. Keep a creative log with the same name, the launch date, the market, the compliance check and the result. After a quarter, the log tells you which hooks last longest per market, which is worth more than any single winning ad.
The iGaming layer: compliance before creativity
Gambling creatives have rules that generic UA playbooks ignore. Our pre-launch check:
- Platform permission first. Google requires gambling certification per country (Google), Meta requires account authorisation and never allows targeting under-18s (Meta), and TikTok allows gambling ads only in specified legal markets after certification (TikTok). A fresh creative doesn't help an account that isn't allowed to run it.
- No appeal to minors. In the UK, the ASA's May 2026 guidance says colourful, exaggerated, cartoon-style graphics are likely to appeal to children (ASA). That changes what "a fresh look" can mean for UK traffic.
- Honest gameplay. Use real footage from the actual game. Don't stitch together max wins as if they were typical, and don't show prize wheels or symbols that imply something nobody can win; the ASA ruled against social casino ads whose wheel graphics implied real money prizes.
- Bonus copy that matches the rules. In Great Britain, wagering requirements are capped at 10 times and incentives may not mix gambling products (LCCP 5.1.1). A creative refresh is the moment old offer copy sneaks back in.
- 18+ and terms on every variant, in every size. The 300×250 banner gets the same disclaimers as the hero video.
- No promises. No "guaranteed wins", no suggestion that gambling fixes money problems, no urgency that isn't real.
When to scale, iterate or retire
| Signal over the last 7 days | Decision |
|---|---|
| Cost per first deposit below target, frequency still low | Scale: raise budget, add sizes and languages |
| On target but CTR sliding against its own baseline | Iterate: new hooks on the same concept, launched alongside |
| Meta shows "Creative limited" | Iterate now; the replacement goes in this week |
| Meta shows "Creative fatigue", or cost per deposit 2× target | Retire or cut to a small budget share once the replacement is live |
| Any policy or rule change affecting the ad | Retire immediately |
One caution for automated placements: Google suggests waiting 2–3 weeks before replacing low-performing assets in Performance Max (Google Ads Help). Weekly is the rhythm for briefing and producing; it doesn't mean every channel gets new assets every week.
Creatives are only half the funnel. If the click lands on a slow or pushy page, the freshest ad won't save it. Read what to test first on pre-landers, see a full game launch page, or check what we make on the Growth page.
FAQ
What is creative fatigue?
Meta defines it as an audience having seen the same creative too many times. In practice you see frequency rising, click-through falling and cost per result climbing while the auction itself is stable.
How often should I refresh gambling ad creatives?
There is no universal number. Watch each ad against its own first-week baseline and the platform's signals (Meta flags 'creative fatigue' when cost per result reaches twice your past ads). A weekly rhythm with new variants ready before you need them works for most accounts.
Is it fatigue or something else?
If CPMs rise across every ad at once, it is usually the auction (seasonality, big sporting events, competitors). If click-through holds but conversions fall, look at the landing page, the offer or tracking. Fatigue is when one ad's click-through falls as its frequency climbs.
Should I turn off a fatigued ad?
Not always. Meta notes that keeping the original ad running alongside a new, materially different one may maximise results. Lower its budget share rather than killing it outright, unless it breaks a policy or a rule has changed.
What makes gambling creatives non-compliant?
Common problems are targeting or appealing to under-18s (in the UK that now includes cartoon-style visuals), missing 18+ and bonus terms, bonus offers that break local rules, and implying a chance of prizes that doesn't exist. Each platform also requires gambling certification or authorisation first.
18+Gambling content for business readers. Every demo we link to runs on play money. If gambling stops being fun, free confidential help is available: BeGambleAware (UK), 1-800-GAMBLER (US), Gambling Therapy (worldwide).
Sources
- Meta Business Help Center, Creative fatigue: www.facebook.com/business/help/1346816142327858
- Meta Transparency Center, Online gambling and games: transparency.meta.com/policies/ad-standards/restricted-goods-services/gambling-games
- TikTok Business Help Center, Smart Creative: ads.tiktok.com/help/article/smart-creative?lang=en
- TikTok Business Help Center, Creative best practices: ads.tiktok.com/help/article/creative-best-practices?lang=en
- TikTok, Web auction best practices (PDF): www.tiktok.com/business/library/TikTok_WebAuctionBestPractices.pdf
- TikTok Business Help Center, Gambling and games ad policy: ads.tiktok.com/help/article/tiktok-ads-policy-gambling-and-games?lang=en
- Google Ads Help, Performance Max asset guidance: support.google.com/google-ads/answer/14528220?hl=en
- Google Ads Help, Gambling and games policy: support.google.com/adspolicy/answer/6018017?hl=en
- Kantar, The wear-in debate is wearing us out (2024): www.kantar.com/inspiration/brands/the-wear-in-debate-is-wearing-us-out
- ASA / CAP, Betting and gaming: appeal to children (May 2026): www.asa.org.uk/advice-online/betting-and-gaming-appeal-to-children.html
- ASA ruling, SpinX Games Ltd (2024): www.asa.org.uk/rulings/spinx-games-ltd-a24-1241354-spinx-games-ltd.html
- UK Gambling Commission, LCCP SR code 5.1.1: www.gamblingcommission.gov.uk/licensees-and-businesses/lccp/condition/5-1-1-sr-code
