What a Sitemap Page Count Can and Cannot Tell You
AGNURA’s Pages metric counts unique URLs from /sitemap.xml only: nested maps, safety caps, alternate-root limits, and why the number is not site size or indexed pages.
A sitemap page count is a count of unique URLs listed in a public XML sitemap tree. It is a discovery hint, not a census of the whole website, and not Google’s index size.
In AGNURA AI Assistant, the Pages metric is intentionally narrow: it starts at {origin}/sitemap.xml only, follows nested sitemaps from that tree when present, and stops at safety caps. If /sitemap.xml is missing, AGNURA shows an error instead of guessing another path.
Hub context: What Is Website Intelligence for Sales and Outreach?. Related SEO distinction: Canonical, Robots Meta, Robots.txt, and Sitemaps: Different Jobs.
Direct answer
| Question | Honest answer for AGNURA’s Pages metric |
|---|---|
| What is counted? | Unique <loc> values inside <url> entries in fetched urlsets |
| Where does it start? | Only /sitemap.xml on the site origin |
What if the site uses /wp-sitemap.xml only? | No count from this tool (by design) |
| Is it “how big is the business?” | No |
| Is it “how many pages rank?” | No |
| Can the number be partial? | Yes, when fetch or unique-URL caps hit |
Use the number as a rough scale signal for outreach research. Verify anything important on the live site.
Why people over-read sitemap counts
Common mistakes:
- Treating the count as every URL that exists
- Treating it as indexed pages in Google
- Comparing two brands as if both expose complete, comparable sitemaps
- Ignoring a capped flag and assuming the walk finished
Sitemaps are optional publishing choices. Teams omit sections, split files, or put the real index somewhere else.
What AGNURA does
Root rule
- Build
{origin}/sitemap.xml. - Fetch that URL.
- If it is a sitemap index, enqueue child
<loc>sitemap files. - Fetch queued sitemaps in parallel waves until urlsets are done or caps hit.
- Count unique page
<loc>strings from<url>entries.
AGNURA does not:
- Read
robots.txtfor Sitemap lines (for this Pages metric) - Try
/wp-sitemap.xml,/sitemap_index.xml, or other alternate roots - Invent a count when
/sitemap.xmlfails
Safety caps (current product)
| Cap | Value | Meaning |
|---|---|---|
| Max sitemap fetches | 80 | Stops further sitemap file fetches |
| Max unique URL locs | 50,000 | Stops adding new unique page URLs |
| Per-request timeout | 8 seconds | Slow sitemap requests abort |
When a cap stops the walk, the UI can show a capped flag and cap details. Treat that count as partial.
Caching
Per-origin sitemap count snapshots can live in chrome.storage.local for about one hour (same idea as analysis cache TTL). A popup reload that clears caches forces a live walk again.
What the number can tell you
| Useful reading | Example |
|---|---|
| Rough catalog scale | A store sitemap tree listing thousands of product URLs |
| “This property publishes a large public URL list” | Useful before a long research session |
| Cap warning | You may be looking at a truncated walk |
Error on /sitemap.xml | This tool’s entry path is missing or not a sitemap |
What it cannot tell you
| Not this metric | Why |
|---|---|
| True sitewide page inventory | Sitemaps omit URLs; no HTML crawl |
| Indexed URL count | Search Console / crawlers are different jobs |
| Content quality | Locs are addresses, not quality scores |
| Exact product count for every stack | Commerce product rollups are separate probes with their own limits |
| Alternate sitemap roots | Not fetched by Pages |
| Business launch date | Separate Archive hint, also not founding date |
Product/catalog estimates, when shown elsewhere in AGNURA, follow platform-specific probes and can still be “Not detected.” Missing product count ≠ zero products.
Options and trade-offs
| Approach | Upside | Trade-off |
|---|---|---|
AGNURA /sitemap.xml count | Fast, local, predictable entry path | Blind to alternate roots |
Manual check of /wp-sitemap.xml etc. | Catches WordPress-style indexes | Extra steps |
| Full-site crawler | Broader inventory | Different tool and time cost |
| Search Console coverage | Index reality for properties you own | Not available for arbitrary prospects |
Decision framework
| If… | Then… | Because… |
|---|---|---|
| Pages shows a healthy count | Use it as a scale hint only | Sitemap completeness varies |
Pages errors on /sitemap.xml | Do not assume the site is tiny; check other paths manually if needed | Entry path is fixed |
| Result is capped | Treat as lower bound / partial | Caps stopped the walk |
| You need “all URLs on the domain” | Use a crawler or owner tools | Outside AGNURA Pages scope |
| You are writing outreach | Pair with human page skim | Counts do not replace reading the offer |
Worked example
| Site behavior | AGNURA Pages outcome |
|---|---|
Publishes a complete tree at /sitemap.xml | Unique loc count (possibly large) |
Only /wp-sitemap.xml, no /sitemap.xml | Error / no count from this metric |
| Huge index with 200 child files | May stop at 80 fetches; capped |
| Sitemap lists 60,000 product URLs | May stop at 50,000 unique locs; capped |
| Sitemap omits blog but includes products | Count reflects the list, not “whole site” |
Risks and limits
- Sitemaps can include parameters, duplicates after normalization differences, or URLs you would not call “pages.”
- Bot protection or slow hosts can fail individual fetches.
- Comparing competitors on sitemap size alone is weak research.
- Exporting or storing URL lists may have compliance implications; you own lawful use.
How AGNURA helps this job
| Need | With AGNURA |
|---|---|
| Fast sitemap-based scale hint | Pages metric on active-tab analysis |
| Know when the walk was incomplete | Capped flag / details when limits hit |
| Avoid false guesses on missing roots | Error instead of inventing /wp-sitemap.xml |
| Refresh | Header reload clears related caches and re-runs live analysis |
Step-by-step analysis: Run Your First Website Analysis.
Compared with common alternatives
| Approach | Fit |
|---|---|
| Browser address-bar checks of sitemap URLs | Manual; flexible roots |
| SEO crawlers | Broader discovery; heavier |
| AGNURA Pages | Strict /sitemap.xml tree count for prospect research speed |
Related guides
- What Is Website Intelligence for Sales and Outreach?
- How to Research a Prospect’s Website Before Writing an Email
- Canonical, Robots Meta, Robots.txt, and Sitemaps: Different Jobs
- Run Your First Website Analysis
- What Is AGNURA?
Soft next step
On your next prospect, note whether Pages succeeded, errored, or capped, then open /sitemap.xml yourself once so the metric’s limits feel concrete. Install AGNURA if you want that count in the popup:
Try AGNURA in Chrome
Draft in Gmail, LinkedIn, WhatsApp, and Reddit. You review every message before send.
Add to Chrome