How a view-through conversion is counted
An advertising platform records that your ad was served to a browser or an account. No click follows. Later, the same browser or account reaches your site by some other route and converts. If that happens inside the view-through window the platform applies, and no click on any of that platform’s ads intervened, the conversion is filed as view-through.
Every part of that definition is decided by the platform, not by you. Each one sets its own window length, its own rule for what counts as the ad having been seen, and its own order of precedence between a view and a click. That is why the same sale can appear as a view-through conversion in one account and go entirely unrecorded in another.
Why view-through conversions matter
Some advertising genuinely works without a click. Video, display placements and broad social formats are seen far more often than they are clicked, and judging them on clicks alone makes them look like waste beside a branded search campaign that mostly harvests demand created elsewhere. View-through counting is the platforms’ attempt to show that contribution rather than ignore it.
Used carefully, the number is a directional signal. An awareness or retargeting campaign with plenty of view-through activity and rising branded search alongside it is behaving the way upper-funnel spend should. Read next to an attribution window you actually understand, it adds context that click data cannot supply on its own.
Where view-through conversions go wrong
The core problem is that a served impression is a weak claim. Someone who was going to buy anyway, and who happened to be shown your ad while scrolling past, produces a view-through conversion the campaign did nothing to cause. Retargeting is where this bites hardest, because those audiences are made of people already on their way back, so the ad collects credit for a return visit it did not create.
Reporting habits make it worse. Adding view-through conversions into the same total as click conversions produces a cost per acquisition that looks strong on the screen and cannot be reproduced in the bank account. Comparing platforms is equally misleading when their windows and their definitions of a view differ. And because the same person may be exposed on several platforms in the same week, each of them can claim the same sale without anyone noticing.
How to act on it
Keep them out of the headline. Report click-based conversions as the primary number and view-through separately, so anyone reading the report can see how much of the argument rests on impressions. If a campaign only works once view-through is included, that is a finding worth investigating, not a result worth celebrating.
Then test rather than argue. A holdout — pausing the campaign in one region, or withholding it from a matched audience — shows whether sales actually fall, which is the only honest answer available. That is the logic behind incrementality testing, and it is worth running before meaningful budget moves onto view-based reporting. Where a holdout is impractical, shorten the view window and watch what happens to the reported total.