My Pages Are Not Being Indexed
Most often indexing was never started, the source reads only a limited batch of pages, or a page could not be read. Check the source card, then the Indexed Content list, then your indexed-pages limit.
Ask AI answers only from content Flidu has indexed, and Flidu indexes a source only when you ask it to. If a page you expected is missing, open AI Chatbots in the dashboard, choose your assistant and open the Data Sources tab. Every check below starts there: the source cards at the top, and the Indexed Content list underneath. Work through them in order.
Check That Indexing Was Started
Adding a source does not index it. A new source is saved as a draft, and nothing is read until you select Start indexing on its card. The workspace Owner and Admins can do this. Read the status beside the name of each source:
| Status | What It Means | What To Do |
|---|---|---|
| Draft | The source was saved but never indexed. | Select Start indexing. |
| Queued or Preparing | A crawl is waiting or running. | Wait for it to finish. If it does not move, see My Crawl Is Stuck Or Paused. |
| Needs attention | The source could not be indexed. | Check that its address is public and reachable, then select Retry indexing. |
| Partly indexed | Some pages were added and some were not. | Go to the third check to see the reason for each page. |
| Indexed | The crawl finished. | Go to the next check. |
Check How Much The Source Covers
A finished card shows three numbers from the last crawl: Found, Indexed and Not added. A source reads a limited batch of pages per crawl, not your whole site, and how far it reaches depends on its type.
- Website URL with Follow links on this site switched off reads that one page. The card says “Indexes this page only.”
- Website URL with the switch on reads the start page and the pages it links to on the same site, one level deep. Links found on those pages are not followed. Neither are links with a query string, links marked nofollow, or links to files such as PDFs and images.
- Sitemap takes a limited batch of pages from the sitemap. A sitemap that points to other sitemaps is followed one level only.
- Manual URL List reads exactly the pages you list, up to 25 per list, all on the same website.
To make sure a particular page is covered, name it in a Manual URL List. A recrawl has the same reach as the first crawl, so recrawling does not go deeper into your site. Pages must be on the same site as the source address, and a different subdomain counts as a different site.
Check Why A Page Was Not Added
Scroll to Indexed Content and use the filters across the top. Failed lists pages the last crawl could not add, each with a reason. Ignored & Skipped lists pages you ignored and pages that match one of your patterns. When items need attention, a notice above the source cards offers Review Failed.
| Reason Shown | Likely Cause | Fix |
|---|---|---|
| The page could not be reached. | The page is behind a login, blocked, or returned an error. | Make the page public, then index its source again. |
| No readable text was found on this page. | The text is added by scripts after the page loads, or sits only in the header, menu, footer or sidebar. | Put the content in the body of the page as your server sends it. |
| This type of page can’t be read. | The address is not an ordinary web page, for example a PDF. | Add the text as a Manual Text source instead. |
| The page was too large to read. | The page is bigger than Flidu reads in one go. | Split the content across shorter pages. |
| This address can’t be crawled. | The address, or the place it redirects to, is not a public web address. | Use the public address of the page. |
| The page couldn’t be saved. | A temporary problem while saving the page. | Index its source again. |
| Matches an exclude pattern. | One of your patterns covers this page. | Open Included & Ignored Pages, then remove the pattern or use Review Skipped Pages to include the page again. |
Flidu reads each page as your server sends it and does not run scripts. It also follows your robots.txt file, so pages that file disallows are skipped. A row marked Needs reindexing is included but has no indexed copy yet: index its source again.
Check Your Indexed-Pages Limit
Each AI Chatbot plan sets an indexed-pages limit per assistant: 50 on Free, 250 on Starter, 500 on Pro and 1,000 on Business. Agency follows its preset and Enterprise its agreement; see Pricing. Every website page the assistant indexes counts, including pages you later ignore. The assistant’s Overview tab shows Indexed pages used against the limit.
At the limit, a new crawl is refused with “Indexed page limit reached.” and a notice offers Upgrade AI Chatbot. Removing a source you no longer need takes its indexed pages off the assistant. Ignoring pages does not make room. See AI Chatbot Allowances.
Check Whether Your Site Changed Since The Last Crawl
Flidu does not recrawl by itself. A page you published after the last crawl is not indexed until you select Recrawl on the source card. After a crawl, the card shows when Recrawl is available again, and each plan has a monthly recrawl allowance that the assistant’s Usage tab counts.
What A Correct Result Looks Like
The source card reads Indexed, and the page appears in Indexed Content with the status Indexed. Ask about that page in the assistant’s Preview tab: with Source Cards switched on, the answer can list the page under Sources. Preview questions count toward your monthly AI messages.
If a public page still fails after these checks, contact us with the page address and the name of the assistant.