Study
Where an ad sits decides whether it gets seen
Active View viewability by placement, measured across the content sites we run. Trailing 30 days.
The net-positive test says viewability is the lever, so we pulled the real viewability of every ad placement on our own network. The gap between the best placement and the worst is 60 points, and it has nothing to do with how many ads you run. It is entirely about where they sit.
The benchmark
Each bar is one placement's share of impressions that were actually viewable. The dashed line is the 60% floor from the net-positive test.
dashed line = 60% viewability floor · green clears it, amber is on notice, red is a defect
The same kind of unit, two places
Look at the two ends. A dismissible sticky anchor lands at 88% viewable. A standard page-bottom unit, often the very same display creative, lands at 28%. The difference is not the ad. It is that the anchor sits in the part of the screen the reader is already looking at, while the page-bottom unit waits for a scroll that most readers never make. Same inventory, 60 points of viewability apart, decided by position.
The most-seen placement is also the most-clicked
Viewability is not the only thing that tracks position. At the two ends of that benchmark, click-through runs the same direction: the anchor draws a 4.8% CTR against 0.5% at the page bottom. A unit in a dead spot loses twice, it is not seen and it is not clicked, which is why moving it beats adding another one. (Click-through also depends on format, so we are pointing at the extremes here, not claiming a clean line across every placement.)
What it means for your layout
This is the net-positive test in real numbers. A unit earns its place by being viewable, not by existing. Three of these seven placements fail the 60% floor, and the fix for them is never "add more ads." It is to move the inventory to a position that earns, an in-content slot or a dismissible anchor, or to cut it. The publishers who stuff a page are usually adding more of the placements at the bottom of this chart, not the top.
How we measured it
AdSense Active View viewability, broken out by ad unit, trailing 30 days, across the content sites we run. We report viewability and click-through per placement because those are clean. We leave out per-unit RPM on purpose: it is distorted by which pages each unit happens to appear on, so it is not a fair cross-placement comparison. The placements are our own, named the way we name them.