A News Publisher Raised Ad Revenue 58% Without Adding a Single Ad Slot
More inventory was the obvious answer. It was also the wrong one — page speed was the real constraint.
The client
A Vietnamese digital news publisher, 11 million monthly sessions, primarily mobile web. Advertising was 89% of revenue. Yield had been flat for six quarters while traffic grew.
The proposal on the table was to add two more display slots per article.
Why we argued against more inventory
Their mobile pages took 6.8 seconds to reach largest contentful paint. Bounce rate before first ad impression was 34%. A third of the audience never saw a single ad.
Adding inventory to a slow page raises theoretical impressions and lowers real ones. You cannot monetise a visitor who left.
What we changed instead
1. Page speed (weeks 1–6)
- Deferred non-critical scripts and consolidated three analytics tags into one
- Lazy-loaded below-fold ad slots with a proper intersection margin
- Compressed and correctly sized hero images
LCP dropped from 6.8s to 2.1s. Pre-impression bounce fell to 11%.
2. Yield mechanics (weeks 7–12)
- Replaced a waterfall setup with header bidding across 8 demand partners
- Set price floors per placement and per geography, refreshed weekly
- Removed two chronically underperforming partners adding latency without fill
3. Audience data (weeks 13–18)
MS CDP built logged-in and inferred audience segments, sold as first-party targeting at a premium over run-of-network.
Results after two quarters
Metric | Baseline | Post-launch | Change |
|---|---|---|---|
Viewable impressions | index 100 | index 141 | +41% |
Average CPM | $0.94 | $1.09 | +16% |
Total ad revenue | index 100 | index 158 | +58% |
LCP (mobile) | 6.8s | 2.1s | −69% |
Ad slots per article | 4 | 4 | unchanged |
Revenue rose 58% on identical inventory. The gain came entirely from impressions that were previously being lost and from selling audience rather than pageviews.
The transferable lesson
Publishers under yield pressure reliably reach for more slots first, because it is the change they fully control. It is also the change most likely to make the underlying problem worse.
Measure what share of your audience sees your first ad before you add a fifth.
Learn about publisher solutions.