AI search, SEO & online discovery
Practical coverage from Dubai

Saved
Technical SEO

What HTTP Status Codes Tell You About Missing Content

Summarize with AI
Website development and code review; illustrative photograph.
Photo by Markus Spiske / PexelsIllustrative photography
In this article 5 sections

A customer follows a saved link to a price list and says the page is missing. The link might lead to a new address, return a genuine error or display an empty page with a success status. The browser’s address bar and the visible message give you clues, but the sequence of HTTP responses tells you how the server handled the request.

Start with the exact link, capture each response and compare the final body with the content you expected. Decide whether the page should exist, should move or should remain unavailable. The AI Visibility site access checklist covers general crawl access. This walkthrough explains how to distinguish the HTTP outcomes and hand each one to the right owner. AI Visibility site access checklist

Start with the intended result for the exact URL

Ask where the report came from and save the address as given, including the path, trailing slash and any meaningful query string. A copied link from an old brochure can behave differently from the link in the current menu. Record the intended outcome before diagnosing it. A retired offer may reasonably return a missing response, while the same result on an active service page calls for a repair.

Pick a working control on the same host and, where practical, the same template. Identify an expected content marker, such as the title of the intended price list or a section unique to that service. A branded header or generic title can appear on both the right page and an error template. The comparison needs a marker that tells you which content actually arrived.

Observed response What to verify next Likely handoff
200 with expected content Check the reported client, URL and time Reproduce the original conditions
200 with missing or unrelated content Compare response body and rendered page Route, template or data owner
Redirect Follow Location and check final body Redirect map or destination owner
404 or 410 Confirm whether the page should exist Publishing or routing owner
403, 429 or 5xx Capture response evidence and check permitted records Access, rate limit or service owner

Capture both the response chain and the final body

Use a public GET and save the header blocks and final body. A HEAD request is useful when you need metadata, but it does not deliver the body you need to inspect. Preserve the intermediate responses if the address redirects. This lets you tell which component sent a Location and where a later error arose. Set a timeout and a redirect limit so a broken route does not make the diagnostic run indefinitely. HTTP semantics, RFC 9110

url='https://example.com/replace-with-the-exact-link'
curl --silent --show-error --location --max-redirs 8 --max-time 20 \
  --dump-header response.headers --output response.html \
  --write-out 'status=%{http_code} final=%{url_effective} redirects=%{num_redirects}\n' \
  "$url"
grep -iE '^(HTTP/|location:|content-type:|retry-after:)' response.headers

If curl reports a DNS, connection, certificate or timeout error before an HTTP response arrives, record the transport error and the environment where it occurred. A missing response and an HTTP 404 are different observations. Check the working control from the same environment and compare the hostname or path involved before sending the issue to the application team. Google's crawler documentation discusses how its systems handle HTTP and network failures. Google HTTP status and network guidance

Read the header blocks in order. A 301 or 308 generally represents a permanent redirect, while 302 and 307 are temporary statuses under HTTP semantics. Capture the Location rather than assuming the redirect landed on the destination intended by the content owner. If the chain returns to a previously requested address, note the loop. If a request fails after crossing to a new hostname, record the first successful hop and the point of failure separately. The team responsible for the old address may need to verify the mapping, while the owner of the new host investigates delivery. HTTP semantics, RFC 9110

If there is a final response, open the intended route in a clean browser and compare the visible content with the recorded response. The site may return an HTML shell and obtain details from a later request. If the document succeeds but the relevant dependency fails, preserve both statuses. Look for a page-level error or empty content area after rendering. Google discusses missing critical resources and error messages as possible reasons a page is classified as a soft 404. Google crawling troubleshooting

Compare a working page, a redirect and a missing page

On 6 November 2026, a public GET of the AI Visibility access checklist returned HTTP 200. The version without the trailing slash returned HTTP 301 to the trailing-slash address, which then returned HTTP 200. A deliberately nonexistent article path returned HTTP 404 without a redirect. A fresh Chrome render of that missing path showed the Not Found document title. These are three observations from the same host. They demonstrate how the status chain and the body distinguish a working address, an intentional route variant and an absent article. Tested AI Visibility checklist page

An isolated local test supplied a useful contrast. One fixture returned HTTP 200 with the intended guide heading. A second returned HTTP 200 with a Page not found heading. A third used HTTP 410 for a withdrawn resource. The fixture also returned 503 for temporary maintenance with a Retry-After value of 120 seconds and 429 for a rate-limit case with a Retry-After value of 60 seconds. The values were deliberately set in the test. They do not describe either website's live policy. The two 200 cases demonstrate why a success code alone cannot show whether the requested content arrived. HTTP semantics, RFC 9110 · Additional HTTP status codes, RFC 6585

A search platform’s soft 404 classification should be recorded separately. Google uses that label when a response indicates that the content is missing despite a success response, and can also apply it when content is empty or error-like. The local fixture demonstrated the response pattern. Google did not inspect or classify it. With access to the relevant property, compare the platform report with a live test of the exact URL. Google crawling troubleshooting · Google Search Console URL Inspection

Match the repair to what happened

If a page moved, send the old link to a relevant replacement and verify what the user receives at the end. A redirect can still leave content missing if its final address is a generic homepage or error page. Check the title or other content marker on the final page. Luxe Nomad’s rebrand announcement names its new website and embeds an official Instagram post next to that announcement. Public communications can tell customers where to start. Check the old URLs independently to see what saved links actually do. Luxe Nomad’s website announcement with its official Instagram post · Google guide to site moves

If the page should still exist, confirm its publishing state, slug and routing. An incorrect link and an unpublished page can both produce 404. If the page was intentionally removed, decide whether a relevant replacement exists. A helpful missing-page template can offer navigation while retaining the appropriate HTTP status. HTTP 410 can express that the resource is gone and that the condition is likely permanent. Use the status that matches what the owner knows and the outcome intended for the URL. HTTP semantics, RFC 9110 · Google crawling troubleshooting

If the response is 403, 429 or 5xx, preserve the relevant identifier, Retry-After header if present and time. Check permitted edge or origin records to locate the source and owner. A 403 can reflect an intentional access decision. A 429 can indicate a limit was reached, and a 503 can signal temporary unavailability. Use the documented limits and wait conditions. Repeated requests during an outage or rate-limit window add traffic without necessarily improving the diagnosis. If a body contains an error while the document status says 200, trace the component that chose the page template and the dependency that supplied its content. HTTP semantics, RFC 9110 · Additional HTTP status codes, RFC 6585

For a practical case, imagine a business with a downloadable brochure and an HTML page explaining it. The HTML page can return 200 while the link sends the reader to a PDF that no longer exists. Test the link from the HTML and request the actual download directly. Record the expected type as well as the status. If the download is redirected to a current edition, check that the edition corresponds to the same service and that the response is the intended document. A successful landing page cannot establish whether its download works.

Check for timing or client differences if you cannot reproduce the original failure. A browser may have cached a previous redirect. An edge may serve different responses at different times or while a release is propagating. Repeat a bounded ordinary request and record its time. When a content owner has permission to inspect logs, match the route and any request ID and check the edge and origin where appropriate. Keep sensitive request data out of the shared handoff.

Record the change and test the same route again

Use a Missing URL diagnostic record with the reported URL, expected result, observed time, HTTP chain, final content marker, browser result and layer evidence where available. Record the intended owner. A status code can identify a symptom, while a matched request or a comparison with a working page can help locate the component responsible. Keep the original observation so the owner can compare the result after the repair.

Repeat the original address and the working control under comparable conditions. If a page was restored, confirm the expected status and visible content. If it moved, verify the redirect and check the destination directly. If the old offer was retired, confirm the missing page returns the intended status and that current internal links stop sending users there. Sample another page using the same rule if the repair changes shared routing or a common template.

If the page has several public entry points, include one from the original report and one from a current internal link. This can expose a repaired destination that still receives an old path from a menu, brochure or campaign.

If the report involved a time-limited offer, confirm the intended result with the business owner before restoring an expired page or changing the redirect.

Record search platform evidence in its own field. A current GET and browser result can confirm the delivered fix now. The platform report may reflect another requester or an earlier observation. Keep the reported crawl date, exact tested URL and live result together so the business can tell whether it is waiting for another observation or still has a reproducible delivery issue. Google Search Console URL Inspection

AI Visibility is published by Lunasol, which offers GEO and AEO services. The worked observations use public responses, one clean Chrome render and controlled local fixtures. About AI Visibility

You’ve reached the end of this article. Read as plain text

Search the blog

Summarize with AI

Choose your assistant. Copy the article and summary request, then paste it into a new chat.

The full article is included, so your assistant can read it without opening this website. Nothing is submitted automatically.

View or copy the summary requestOpen assistant

Summaries are generated by your chosen service and may miss details. Its account requirements and terms apply.