The ad data we track
For every ad a competitor runs, we capture as much as Meta's public Ad Library exposes:
| Field | What it tells you |
|---|---|
| Library ID | The ad's permanent archive id, linked to its public Ad Library page |
| Start date | "Started running on" — when the ad went live |
| Creative type | Video (with duration), carousel (with every card), or single image |
| Creatives | The actual images / video posters, archived durably to our storage |
| Exact copy | The advertiser's word-for-word primary text |
| Headlines | Link titles, per carousel card where applicable |
| CTA & display domain | The button label (Shop Now, Learn More…) and the domain shown |
| Landing URLs | The real destination links with UTM parameters intact — see their tracking setup |
| Variants | "N ads use this creative and text" — how hard they're scaling it |
| Low-impression flag | Meta's badge for ads with minimal delivery — their unproven tests |
| Platform count | How many Meta surfaces (Facebook, Instagram…) the ad runs on |
Started, stopped, continued
The board diffs any window (day / week / month) using the source-reported dates:
- Started— the ad's start date falls inside the window. An ad that started andstopped inside the window counts as started (it's net new).
- Stopped — the ad disappeared from the active listing during the window.
- Continued— running since before the window and still going. These are the competitor's proven winners.
Media durability
Meta's CDN links expire within hours. During every sync we download each new creative and store our own copy, so your archive keeps working forever — even after the competitor kills the ad. When a creative can't be downloaded, we capture a screenshot of the ad card instead, so you never lose the visual.