A long-running creative shows weaker results, and the impressions, frequency, placements and results by period are available to judge it. In this case, the immediate task is to investigate potential creative fatigue. The workflow should separate confirmed fatigue signals from other explanations, citing the evidence behind each.
First replace every [bracketed] placeholder with real store scope and constraints, then add any actual records or data you have, and paste it into your preferred AI assistant or Spark agent. Spark will follow this page's workflow to produce structured output and verification steps.
Act as a Shopify operations specialist.
Help me investigate potential creative fatigue.
My store context is [describe the business, products and target audience].
My scope is [specify the affected products, pages, customers or reporting period]. Use these inputs: [per-creative delivery history by period with impressions, reach and frequency, the placement breakdown, result events per period, the currency and attribution setting, and the merchant's comparison periods].
Compare each creative's recent period against its own earlier period on the same placement mix, treating fatigue as confirmed only when results fall while frequency rises and the audience, placement and bid settings hold steady, and marking a decline that appears while those inputs shifted as not explained by fatigue.
Return a fatigue assessment separating confirmed signals from other explanations and the evidence used for the specified task.
Show the relevant evidence or before-and-after examples and explain how I can verify the result.
If essential data is missing, ask for it and mark unsupported conclusions as unknown.
Do not invent metrics, product claims or customer facts.
For any publication, message, account or store change, use only connected tools within my explicitly authorized scope and report success only after verification; otherwise provide the draft or proposed steps.Split the work into input, judgment, and deliverable phases. Verify each step's output before moving to the next.
Pull the delivery history for each creative over the merchant-defined comparison periods, keeping impressions, reach and frequency, the placement breakdown and result events per period, with the currency and attribution setting attached and any period only partly reported marked as incomplete.
Compare each creative's recent period against its own earlier period on the same placement mix, treating fatigue as confirmed only when results fall while frequency rises and the audience, placement and bid settings hold steady, and marking a decline that appears while those inputs shifted as not explained by fatigue.
Deliver a fatigue assessment listing each creative as confirmed, suspected or not fatigued with the period-over-period figures behind the verdict, then reconcile the creative-level results with the campaign totals for the same window and note every creative whose placement mix changed enough to block the comparison.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
Wear-out shows results falling while frequency climbs with audience, placement and bid settings unchanged; if frequency is flat but results fall, the cause is more likely a delivery or audience shift, and the assessment should record that instead of naming fatigue.
On the creative, because several creatives inside one ad set age at different speeds and an averaged view hides the one that decayed; the ad set is used only to reconcile campaign totals at the end.
Mark it incomplete and exclude it from the verdict rather than annualising a partial period, since late-arriving conversions land disproportionately in the newest days and would make a healthy creative look fatigued.
Start with scenarios under the same topic, then expand into cross-topic matches suggested by similarity.
Ciwi Spark builds Shopify-native AI tools and agents around the workflows merchants actually run: research, ads, translation, and customer support. Stop rewriting the same prompts by hand.