Deep Crawl
Read 50 to 200 pages of one site and get a merged system and a cleanup plan.
Deep Crawl is a background job that reads many pages of one site, sampled by page template.
Start a crawl
- Open Deep Crawl and enter the site.
- Choose 50 to 200 pages. The credit cost is shown before you start.
- Start. The job runs in the background and you can leave the page.
What you get
- Merged tokens across the sampled pages.
- An inventory of buttons, cards, inputs and other components, with how many pages use each and which are one-offs.
- A list of places where pages disagree.
- Screenshots, and a cleanup plan you can download as markdown.
Cost and caps
| Setting | Value |
|---|---|
| Pages | 50 to 200 |
| Credits | 1 per 10 pages, rounded up (50 pages is 5) |
| Pages sampled per template | 3 |
| Pages read at once | 3 |
| Whole job | 40 minutes |
A job stops cleanly at the first cap it reaches and delivers what it has. Deep Crawl reads public pages only. For pages behind a login, see Agency.