Crawl Foundry documentation

Choose the guide that matches the task

You do not need to read the documentation from beginning to end. Start with the job in front of you. The Quick Start covers the first useful session. Workspace setup explains ownership, markets, access, and billing boundaries. The feature guides then go deeper into the data and controls on each surface.

If a label in the app has a help link, follow it. These links open the section that owns the definition, including the active route and anchor. That is usually more reliable than searching for a similar term on another feature page.

Know what the product record proves

Crawl Foundry combines several kinds of evidence, but they do not all answer the same question. Imported keywords prove what was added to a workspace. A current metric records the value and source available for that row. A requested SERP snapshot records a search environment at a particular market and time. A crawl records what Site Audit could reach under its settings. Activity records the state of supported jobs and changes.

Those records support a decision; they do not make it automatically. A high-volume keyword is not automatically a page brief. A ranking change does not establish its cause. A Site Audit issue still needs reach, template, business importance, and implementation context before the team prioritizes it.

• Check the workspace, list, domain, market, language, device, and date before comparing values.
• Treat missing data as unknown until the relevant source, request, and job status have been checked.
• Keep provider-specific metrics within their documented definition; similarly named scores are not necessarily interchangeable.
• Use exports as a handoff of the selected evidence, not as proof that an editorial or technical decision was approved.

For technical website problems, begin with the Site Audit workflow and follow each finding to its affected URLs.

A sensible first session

Paid data actions are deliberate operations

Searching, filtering, changing visible columns, organizing lists, and reading data already in the workspace are different from asking Crawl Foundry to fetch or process fresh external evidence. The second category can use balance. The confirmation surface should show the selected scope and estimated customer cost before execution.

Start from the smallest useful set. Existing fresh data, cache reuse, selected packages, market, and row count can affect the final operation. If a run cannot reserve enough balance, it should not be treated as successfully completed simply because a request was submitted.

Use the guides alongside the app

• Follow the menu and button labels to the relevant product view.
• Use the worked examples to apply the workflow to your own site and market.
• If data is missing, check whether it has been requested, is still processing, or is unavailable.
• Check the selected workspace and filters when your view differs from an example.

For dated product changes, read how to use the changelog alongside the current instructions.

When the result does not look right

First check scope and state: active workspace, list, filters, market, language, domain, source date, and job status. Many apparent data problems are a different selection, a still-running action, or a row that has never received the requested dataset.

Then open Activity and the relevant troubleshooting guide. Record the affected action, time, scope, and visible error before retrying. A focused report is easier to diagnose than a screenshot of an empty table without its filters or workspace context.

Example: A first week with Crawl Foundry

Frequently asked questions