PROPOSED PROJECT
Campaign Fatigue Monitor
Help advertising teams spot ads getting less response over time. Connect performance reports to the actual creative, group ads by message and visual style, and compare results alongside spending, repeat views, and audience changes. Suggest new versions for the team to test.
Response is falling. The explanation needs work.
One creative version
Authored campaign records
Click-through rate / clicks ÷ impressions
The project in practice.
Fictional ad A17 has 1,800 clicks from 100,000 impressions in week one, 1,680 from 120,000 in week two, and 1,540 from 140,000 in week three. Click-through rate falls from 1.8% to 1.1% while average exposure rises from 1.8 to 3.1. The returning-audience share also rises from 30% to 60%, so creative fatigue is only one possible explanation.
WHAT IS USUALLY DONE
An ad team notices fewer clicks and assumes people are tired of the creative. Weekly totals may mix changing spend, audiences, placements, and repeat exposure, making it hard to distinguish a creative issue from a delivery change.
WHAT SOVRUN CAN DO FOR YOU
Sovrun would connect performance exports to the actual ad versions and group them by message and visual approach. It would compare response rates alongside spending, repeat exposure, and audience mix, flag patterns worth investigating, and draft a focused creative test. The team reviews the evidence before deciding that fatigue is the cause.
More repeat exposure.
A different audience mix, too.
The creative stayed the same while average exposure and returning-audience share rose. Spend also increased. Those changes are part of the comparison and should remain beside the response trend.
1.8 average exposures
- Impressions
- 100,000
- Clicks
- 1,800
- Spend / example units
- 900
- Returning audience
- 30%
- Click-through rate
- 1.8%
2.3 average exposures
- Impressions
- 120,000
- Clicks
- 1,680
- Spend / example units
- 1080
- Returning audience
- 50%
- Click-through rate
- 1.4%
3.1 average exposures
- Impressions
- 140,000
- Clicks
- 1,540
- Spend / example units
- 1260
- Returning audience
- 60%
- Click-through rate
- 1.1%
Compare A17 with a refreshed opening under agreed audience, placement, budget, and measurement settings. Keep the product and offer constant. An improved result would still need a valid comparison before being attributed to the creative change.
A pattern to investigate.
A focused refresh to test.
The proposed workspace links actual creative to comparable records and helps the team decide what to make next. It does not label every falling rate as fatigue or promise a performance improvement.
Click-through rate / clicks ÷ impressions
The audience changed.
- Average exposure
- 1.8 → 3.1
- Returning share
- 30% → 60%
- Spend
- 900 → 1,260
- Fatigue cause
- Not confirmed
Test a new first scene.
- Keep
- Product + offer
- Change
- Opening scene
- Match
- Audience + placement
- Result
- Not measured
How we’d make it happen.
Join the ad and its delivery record
Connect each approved creative version to the correct export rows, dates, and metric definitions.
Compare the response with its context
Calculate rates and inspect spend, repeat views, and audience changes. Flag incomplete or non-comparable periods.
Test a specific refresh
Draft alternatives addressing the observed pattern. The team approves a controlled comparison before changing production or budget.
Where AI helps
Vision and language models classify supplied creative by message, scene, and format. Code calculates comparable rates from source counts and preserves attribution windows. Models summarize patterns and propose new versions, while missing metrics and changed delivery conditions remain explicit.
What people approve
Advertising teams approve the creative grouping, comparison periods, metric definitions, and test design. They distinguish an observed decline from a cause and control campaign changes and spending. The tool does not claim access to a platform’s recommendation algorithm.
What we’d need to start
- Creative files and IDs linked to original performance exports
- Spend, impressions, clicks, repeat-exposure measures, and audience definitions
- Comparable time windows, attribution rules, and campaign reviewers
How we’d test it
Recompute rates and review known examples of declining response with and without creative changes. Test missing metrics and changed audiences. Compare refreshed and existing ads under agreed conditions before attributing a result to fatigue or claiming an uplift.
Prepared for Marketing & advertising. We would agree on the work after reviewing your brief, materials, and requirements.
This is a proposed project, not completed client work. The visuals and numbers show example outputs, not measured results.
