YouTube ad spy: research video ads by format and viewer intent

Name the format before you judge the hook

Google lists several YouTube formats. Skippable in-stream ads play before, during, or after other videos and can be skipped after five seconds. Bumper ads are non-skippable and last five to six seconds. In-feed ads appear as a thumbnail and text in places such as search results. Shorts ads appear between Shorts and can be swiped away.

Each format changes the research question. For skippable ads, ask what the first five seconds promise. For in-feed ads, ask what the thumbnail, title, and any preview that plays communicate before a viewer chooses to watch. For Shorts, ask whether the ad feels native to full-screen vertical viewing.

Official sources: Google Ads Help: About video ad formats

Separate video intent from search intent

YouTube research is not Google Search research with video added. A viewer usually came to watch something else. Even an in-feed ad that appears in YouTube search results competes with organic videos, not with text answers. Keep YouTube notes in their own set.

1. Open the Ads Transparency Center and set the platform to YouTube.
2. Search the advertiser or the destination domain.
3. For each video, record the format if you can confirm it, and mark it unknown if you cannot.
4. Write timestamps for the hook, the product reveal, the proof, and the call to action.
5. Compare where the product first appears across the advertiser’s videos.
6. Paste the timestamp map into the free brief tool, and use the cost calculator to price a CPV test from your own inputs.

Official sources: Google Ads Transparency Center · Google Ads Help: Ads transparency

Illustrative example: a timestamp map

Suppose a language app runs two videos. In one, a speaker orders coffee in Spanish at 0:02 and the app appears at 0:09. In the other, the app interface appears at 0:01. A timestamp map shows the choice clearly: situation first or product first.

That is a test idea, not a verdict. Build your own version with your own footage and claims you can support, and keep the offer constant so the sequence is the only change.

Follow the video to its page

Long YouTube ads often explain more than the landing page does. Check whether the page repeats the central demonstration or assumes the viewer remembers it. Record the page you actually reached and the date you reached it.

If you need placement research for channels and videos, that is a different task from creative research. Our ad spy tools comparison explains the difference.