The Quartile Drop-Off Audit: Diagnosing Where Viewers Bail
Objective: Given a real YouTube Ads quartile-report export, diagnose whether a weak hook or a weak middle is killing view-through, and recommend the specific fix.
You're a growth marketer at Freshworks reviewing a 30-day skippable in-stream campaign for a new helpdesk add-on. View rate has been sitting well below the account's historical average, and leadership wants a diagnosis before next month's budget renewal.
Read the view-rate and quartile drop-off numbers, decide whether the problem lives in the first 5 seconds or somewhere in the middle of the video, and recommend the exact fix, not a general 'refresh the creative' note.
Before you start
What you'll need
Free path (everything below is enough to finish)
No account setup needed, works directly on the exported CSV
The process
2 steps
Step 01 of 02
The lesson's measurement section gives a hard benchmark: industry average view rate is around 31.9%. Below 15% on a skippable ad usually signals a weak hook in the first 5 seconds, not a targeting problem.
This export shows 1,240,000 impressions and 168,000 views on the skippable in-stream ad group. Is this a hook problem, a targeting problem, or healthy performance?
Procedure
- Import quartile-export.csv and freeze row 1
- Divide total views by total impressions to get view rate
- Compare the result against the 31.9% industry benchmark and the 15% weak-hook floor
Freshworks Helpdesk Add-On, skippable in-stream Impressions 1,240,000 Views (30s/completed) 168,000 View rate 13.5% Industry benchmark 31.9% Weak-hook floor 15%
Healthy
View rate sits at or above roughly 20-25%, close enough to benchmark that the targeting is reaching an interested audience and the opening seconds are holding attention.
Unhealthy
View rate sits below the 15% floor, as it does here at 13.5%, which points to the first 5 seconds losing viewers before the message even lands.
What this means
13.5% is below the weak-hook floor, so the fix is a new opening 5 seconds, not a new audience. Reworking targeting here would waste a sprint on the wrong layer.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| View rate is under 15% while targeting matches the lesson's ICP guidance | Storyboard 3 new hook variants using the ABCD framework's Attract step before touching audience settings | half day |
Step 02 of 02
The lesson explains that Google Ads reports what percentage of viewers watched 25%, 50%, 75%, and 100% of a video. A big drop between two adjacent quartiles pinpoints exactly where the video loses its audience.
The same export shows 25% retention at 61%, 50% retention at 58%, and 75% retention at 22%. Where is this video actually losing people?
Procedure
- Chart the four quartile retention percentages as a line
- Calculate the point-to-point drop between each adjacent quartile
- Flag the single largest drop as the diagnosis
Quartile retention, Freshworks skippable ad 25% mark 61% 50% mark 58% (-3 pts) 75% mark 22% (-36 pts) 100% mark 19% (-3 pts)
Healthy
Retention declines gradually across all four quartiles with no single drop exceeding roughly 10-15 points.
Unhealthy
One quartile-to-quartile gap is dramatically larger than the others, as the 36-point collapse between the 50% and 75% marks is here.
What this means
The hook is actually fine, 61% still watching at the 25% mark is respectable. The video loses two-thirds of its remaining audience in the middle third, which points at a pacing or offer problem around the midpoint, not the opening.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| A single quartile shows a drop far larger than the others | Re-cut the video so the product demo or offer reveal moves earlier, before the 50% mark | half day |
Final deliverable
A one-page diagnosis memo stating whether the hook or the mid-video pacing is the problem, backed by the view-rate and quartile numbers, with one concrete next action.
See a reference example
Peloton, Q2 skippable in-stream diagnosis (excerpt) VIEW RATE: 24.1% — inside normal range, hook is not the issue. QUARTILE DROP: -31 pts between 25% and 50% marks. DIAGNOSIS: Video front-loads brand story before the workout demo; move the demo to the first 10 seconds. ACTION: Re-cut opening, retest within 2 weeks.
Success criteria
You're done when you can:
- Correctly calculates view rate and compares it against the 15%/31.9% benchmarks
- Identifies the single largest quartile-to-quartile drop rather than treating all four numbers as equally important
- Recommends a fix that matches the diagnosis (hook fix for low view rate, mid-video re-cut for a quartile collapse), not a generic 'refresh creative' note