Overview
Creative Fatigue Detection is an automated monitoring tool within the Creative Affinity application that identifies ads experiencing performance decline. By analyzing daily metrics and comparing current performance against each ad's launch baseline, the system flags fatigued creatives early—helping you optimize spend and maintain campaign effectiveness.
Table of Contents
How It Works
Creative Fatigue Detection uses a launch average comparison methodology to identify when ads start underperforming. Instead of applying static benchmarks, the system:
Establishes a baseline from each ad's first 3-7 days (the "honeymoon period")
Monitors ongoing performance using rolling 3-day averages
Compares current vs. launch metrics (CTR, ROAS, CPM, impressions)
Applies funnel-specific thresholds based on your ad's objective
Flags ads when performance drops below acceptable levels
Calculates days to fatigue to track when decline started
Why Launch Average Comparison?
Every ad has a unique performance trajectory. A top-funnel awareness ad naturally has different CTR expectations than a bottom-funnel conversion ad. By comparing each ad to its own launch performance, we eliminate false positives and catch meaningful declines earlier.
Eligibility Criteria
For an ad to be analyzed for fatigue, it must meet all three criteria:
Criterion | Threshold | Reason |
Lifetime Spend | ≥ $500 | Ensures sufficient budget has been invested to warrant optimization |
Lifetime Impressions | ≥ 10,000 | Provides statistically significant data for trend analysis |
Days Live | ≥ 7 days | Allows time to establish a meaningful launch baseline |
Note: Ads below these thresholds are excluded from fatigue analysis but remain visible in Creative Affinity.
Detection Methodology
Launch Average Calculation
The system calculates baseline performance from your ad's first 3-7 days:
Minimum: Uses at least 3 days of data
Maximum: Uses up to 7 days if needed to reach 1,000 impressions
Metrics tracked: CTR, ROAS, CPM, CPA, and impressions
Example: If an ad reaches 1,000 impressions in 4 days, the launch average uses those 4 days as the baseline.
Current Performance Calculation
Rolling 3-day average of the most recent data:
Updates daily as new data arrives
Smooths out day-to-day volatility
Provides consistent comparison point
Threshold Checks
The system performs multiple checks against funnel-specific thresholds:
CTR Decline % = (Launch Average CTR - Current CTR) / Launch Average CTR If CTR Decline % ≥ Funnel Threshold: → Fatigue signal triggered
Similar calculations apply for:
ROAS decline (15% threshold across all funnels)
CPM increase (10-15% depending on funnel)
Impression decline (reach collapsing or declining)
Revenue stability (zero or volatile ROAS)
Affinity score (platform match quality)
Fatigue Signals Explained
The system monitors 12+ distinct signals categorized by severity:
🔴 Critical Signals
Signal | Detection Logic | Impact | Recommended Action |
No revenue generated | ROAS = 0 for ≥80% of days with spend ≥$20 | Ad consuming budget without returns | Pause immediately or revise creative/targeting |
Reach collapsing | Impressions dropped from 1,000+ to <100 | Ad barely being shown | Check for disapprovals, audience saturation, or budget issues |
🟠 High Signals
Signal | Detection Logic | Impact | Recommended Action |
CTR down X% from launch | Current CTR declined ≥10-20% vs. launch (funnel-dependent) | Reduced engagement efficiency | Refresh creative, test new copy/visuals |
ROAS down X% from launch | Current ROAS declined ≥15% vs. launch | Declining profitability | Review conversion funnel, test new audiences |
CPM up X% from launch | Current CPM increased ≥10-15% vs. launch | Rising costs reduce ROI | Expand targeting or refresh creative |
Costs escalating | CPM increased 50%+ above average | Profitability eroding | Review competitive landscape, adjust bidding |
Unstable revenue | 40%+ days with ROAS = 0, but max ROAS > 10 | Unpredictable performance | Analyze timing patterns, adjust scheduling |
CTR critically low | Average CTR < 0.2% | Users not engaging | Test new headlines/images or pause |
Reach declining | Impressions down to <25% of initial volume | Limited exposure | Expand targeting or increase bids |
Poor platform match | Affinity score < 20 | Platform limiting delivery | Optimize for platform best practices |
🟡 Medium Signals
Signal | Detection Logic | Impact | Recommended Action |
CTR declining | Consistent decline with recent CTR 30% below average | Engagement dropping | Test creative variations |
CTR below threshold | CTR < 0.3% (early) or < 0.7% (mature) | Suboptimal performance | A/B test different variants |
Cost volatility | CPM variance > 50 with avg CPM > $15 | Unpredictable costs | Investigate timing patterns, consider dayparting |
Platform match declining | Affinity score < 40 or declining | Future delivery issues | Follow platform creative guidelines |
Funnel-Specific Thresholds
Different ad objectives have different performance expectations:
Top Funnel (Brand Awareness)
Goal: Reach and engagement
CTR Threshold: ≥10% decline from launch
CPM Threshold: ≥15% increase from launch
VTR Threshold: ≥10% decline (when available)
Frequency Threshold: >3 (future use)
Rationale: Top-funnel ads prioritize reach; CTR naturally lower but consistency matters.
Middle Funnel (Consideration)
Goal: Interest and interaction
CTR Threshold: ≥15% decline from launch
CPM Threshold: ≥15% increase from launch
Frequency Threshold: >4
Default thresholds when funnel unknown
Rationale: Balanced approach between reach and conversion metrics.
Bottom Funnel (Conversion)
Goal: Conversions and ROAS
CTR Threshold: ≥20% decline from launch
CPM Threshold: ≥10% increase from launch (most sensitive)
Frequency Threshold: >5
Rationale: Conversion ads are more cost-sensitive; even small CPM increases matter.
Universal Threshold
ROAS Decline: ≥15% across all funnels (reduced from 25% for earlier detection)
Using the Fatigue Monitor
Accessing the Tool
Navigate to: Intelligence → Creative Affinity → Fatigue Monitor
Dashboard Components
1. Overview Cards
Ads Showing Fatigue: Total count of flagged ads
% of Analyzed Spend: Percentage of your spend going to fatigued ads (in selected time period)
Average CTR: Mean click-through rate for fatigued ads
2. Fatigue Over Time Chart
Line chart showing daily counts of:
Low fatigue ads (1 signal)
Medium fatigue ads (2 signals)
High fatigue ads (3+ signals)
Use this to identify trends and seasonal patterns.
3. Fatigued Ads Table
Detailed table showing:
Ad creative (thumbnail preview)
Ad/Campaign/Client names
Platform
Funnel stage
Days live
Days to fatigue (when decline started)
Fatigue reasons (specific signals triggered)
Spend in period
Key metrics (CTR, ROAS, CPM)
Launch average (baseline for comparison)
Interactive Features:
Click any fatigue reason to see detailed explanation modal
Click ad thumbnail to view full creative
Sort by any column to prioritize actions
Filter by platform, client, or funnel stage (using main Creative Affinity filters)
4. Signal Explainer Modals
Click the info icon next to any fatigue reason to see:
What it means: Plain-language description
How we detect it: Technical logic and thresholds
Impact level: Critical/High/Medium with explanation
Recommended action: Specific next steps
Best Practices
1. Set Up Regular Review Cadence
Weekly: Review high-severity signals (no revenue, reach collapsing)
Bi-weekly: Analyze launch comparison signals (CTR/ROAS/CPM decline)
Monthly: Evaluate overall fatigue trends and creative refresh cycles
2. Prioritize by Spend
Sort by "Spend in Period" to focus on ads consuming the most budget first. A fatigued ad spending $10K/month has higher optimization priority than one spending $500.
3. Use Days to Fatigue for Planning
If your ads consistently fatigue after 21 days, build that into creative production timelines. Pre-schedule refresh campaigns before fatigue hits.
4. Combine with Creative Affinity Insights
Cross-reference fatigue signals with affinity scores. Low affinity + fatigue signals = strong refresh candidate.
5. Test Before Pausing
Unless signals are critical (no revenue, reach collapsed), test creative variations before completely pausing. Sometimes minor tweaks restore performance.
6. Respect the Funnel Context
A bottom-funnel ad with "CPM up 10% from launch" is more concerning than a top-funnel ad with the same signal. The thresholds reflect this.
7. Monitor % of Spend
If >20% of your spend is on fatigued ads, you have a systemic creative refresh problem. Accelerate production or rotate in proven performers.
FAQ
Why aren't all my ads showing in the Fatigue Monitor?
Only ads meeting the eligibility criteria (≥$500 spend, ≥10K impressions, ≥7 days live) are analyzed. This ensures statistically meaningful results.
My ad shows fatigue, but metrics look okay. Why?
The system compares performance to that specific ad's launch baseline. An ad that launched strong but declined 20% may still have decent absolute metrics, but the decline indicates growing fatigue.
Can I adjust the thresholds?
Currently, thresholds are fixed based on industry best practices and internal testing. Custom thresholds may be added in future releases.
How often does data refresh?
Daily metrics update once per day at 8:00 AM UTC. The dashboard shows the last update timestamp.
What if an ad has multiple fatigue signals?
Multiple signals indicate higher severity. An ad with "CTR declining" + "Costs escalating" + "Reach declining" needs immediate attention.
Does this replace Creative Affinity scoring?
No—they're complementary. Creative Affinity scores how well creative fits platform best practices. Fatigue Detection monitors performance decline over time. Use both for complete creative intelligence.
Why 3-7 days for launch average?
This captures the "honeymoon period" when ads perform best (fresh to audiences, platform learning phase complete). It balances statistical significance with recency.
What about seasonal campaigns?
For campaigns <30 days (early stage), thresholds are more sensitive to detect issues quickly. Seasonal patterns are better analyzed by comparing year-over-year rather than within-campaign launch periods.
Can I export fatigued ads?
Yes—use the main Creative Affinity export functionality after applying fatigue filters (coming in future release). Currently, you can manually note high-priority ads.
How does this handle ad edits/updates?
Ad edits that significantly change creative should be treated as new ads. Currently, the system tracks by ad ID, so major edits may show false fatigue signals. Best practice: create new ad campaigns for major creative changes.