When a winning ad weakens, first check whether the comparison conditions changed. One day of lower return on ad spend does not prove creative fatigue. Compare the same asset, audience, placement, time zone, and period, then check whether attention, clicks, orders, comments, and post-click quality moved together. Investigate changes to price, stock, page, offer, tracking, or learning status before ordering a reshoot.
14 parts · 2 checkpoints · 3 free / 11 member parts →
In one line
Treat fatigue as a reviewable hypothesis. Record the baseline, current window, signals that moved together, counter-evidence, preserved proof, one planned change, and the condition that sends the decision back to watch.
Questions this lesson answers
Define the metrics before reading a decline
Name each metric’s denominator and job. Frequency is average exposures per reached person; read it with reach, impressions, audience size, placement, and budget. CPM is spend per thousand impressions, so it can flag traffic-price or sample changes. CTR is clicks divided by impressions; outbound CTR uses outbound clicks. Three-second views and watch time follow each platform’s reporting definition, so record the report field and denominator.
CVR is a conversion rate only after naming its event and denominator, such as attributed orders per session or checkout starts per click. CPA is spend divided by attributed orders within a stated attribution window and order-maturity cutoff. ROAS is ad-attributed revenue divided by ad spend. It is a return signal, not incremental revenue, and does not by itself deduct refunds, product costs, shipping subsidies, payment fees, or support costs. Break-even ROAS is the attributed return needed to cover the variable costs included in the chosen calculation. Contribution profit subtracts relevant variable costs; order quality also needs SKU (product or variant code) mix, refund reasons, and support issues.
No single association proves a creative caused an order or profit change. CTR can rise while conversion stays flat; ROAS can rise while refunds or margin worsen. Explain the metric first, then compare a clean window and check downstream evidence.
Read a clean window before choosing an action
The default review window is 5–7 days, not a fixed fatigue threshold. Extend it to 14 days for small budgets, few orders, high average order value, or long purchase cycles. A high-budget, high-frequency asset may enter a watchlist sooner so the team can prepare a replacement, but that is not a failure verdict. If the sample is thin, observe or collect evidence.
Pet travel bottle: inspect the nine-day window
A teaching example uses a pet travel bottle ad that ran for 9 days. ROAS looked acceptable on days 1–5. From day 6, first-three-second hold and CTR weakened for several days while comments still asked about capacity, leakage, and car cup-holder fit. Before calling that fatigue, compare price and stock in the product data feed, the product-page first screen, the checkout free-shipping threshold, and whether comments still question the same product evidence. If a condition changed, split the comparison periods and record why.
Check what the decline can tell you
Establish comparability before refreshing. A new price, audience, or attribution window changes the readout. Resolve known changes before deciding whether the opening, proof, or execution needs work.
Clicks hold, purchases fall: what comes first?
Self-check not answered
Pro · Parts 4-14 · 11 parts
Diagnose, protect, and reuse a weakening winner
Compare a clean window, preserve proof and rights, and choose the smallest action that the evidence supports.
This part includes
- Separate hook, proof, execution, angle, and page counter-evidence before choosing an action
- Protect post ID, comments, proof, raw files, and rights; measure each destination on its own
- Record parent and child assets, one change, profit guards, stop line, and next review
1 self-check question
Pro $20 / monthYearly $16 / month
Choose a plan, sign in with your email or Google account, and complete payment. You'll have instant access to all member benefits.
- Unlock every premium lesson
- Advanced tools and member resources
- Member-requested courses and ongoing updates
- Pro community and priority support
Checking your membership…
What's the difference between monthly and yearly?
Yearly plans save you 20% with annual payment. Monthly plans offer more flexibility for short-term needs.
No checkpoint finished in this lesson yet. Start with “Go to the exercise”.
Self-check not answered
Common questions
Three questions before the review
- Does one day of lower ROAS prove fatigue?
- No. Compare a consistent window, attention, clicks, and order quality, then inspect page, product, and account changes. ROAS is attributed revenue divided by spend; it does not identify the cause.
- Should high frequency trigger a pause?
- Frequency indicates repeated exposure. If profit and order quality hold, observe and prepare alternatives rather than stopping a useful asset on frequency alone.
- Where should I look when CTR holds but CVR falls?
- Check price, stock, shipping, offers, and checkout changes first. Match the event and denominator before deciding whether product proof is missing.