Server Response Code Patterns: The Complete 2026 Guide
Server response codes tell you exactly what happened when a bot requested a URL. In log file analysis, the distribution of response codes reveals site...
- 2xx Success.
- Redirect patterns in logs reveal important information: 301 permanent.
- 404 patterns reveal content management issues: Soft 404s.
- 5xx responses are the most damaging to crawl performance: 503 spikes.
- Healthy site benchmarks from a 2026 analysis of 500 sites: Code Category Healthy Range Warning Level Critical Level 2xx 95-99% Below 90% Below 80%...
- Export 30 days of log data Calculate response code distribution for Googlebot only Identify all URLs returning 4xx and 5xx Trace redirect chains...
Server response codes tell you exactly what happened when a bot requested a URL. In log file analysis, the distribution of response codes reveals site health, configuration errors, and crawl budget waste. Each status code pattern tells a different story.
Status Code Categories in Logs
2xx Success. 200 (OK) is the desired response for indexable pages. 206 (Partial Content) appears for byte-range requests from resume-capable downloads. High 2xx ratio (above 95%) indicates a healthy site.
3xx Redirection. 301 (Moved Permanently) and 302 (Found) are the most common redirect codes. 307 and 308 are HTTP 1.1 equivalents. A reasonable site has 3xx responses on 2-5% of crawl requests. Higher ratios indicate redirect chain issues.
4xx Client Error. 404 (Not Found) and 410 (Gone) indicate missing pages. 403 (Forbidden) indicates access denied. 429 (Too Many Requests) means rate limiting. More than 1% 4xx on crawl traffic indicates structural problems.
5xx Server Error. 500 (Internal Server Error), 502 (Bad Gateway), 503 (Service Unavailable), and 504 (Gateway Timeout) indicate server problems. Any sustained 5xx rate above 0.5% triggers Google to reduce crawl frequency.
Reading 3xx Patterns
Redirect patterns in logs reveal important information:
301 permanent. Googlebot transfers PageRank through permanent redirects. The redirected URL is typically removed from the index after recrawl. A high volume of 301s to new URLs suggests a site migration that may not be fully resolved.
302 temporary. Googlebot follows 302 redirects but does not transfer PageRank as strongly. The original URL stays in the index. A 302 that persists for months should be converted to a 301.
Redirect chains. Logs show multiple consecutive 3xx responses for a single request. Example: URL A returns 301 to URL B, which returns 301 to URL C. Each hop wastes crawl budget. Chains longer than 2 hops should be flattened.
Meta refresh redirects. These appear in logs as a 200 response followed by a delayed second request. They are slower and less efficient than server-side redirects.
Reading 4xx Patterns
404 patterns reveal content management issues:
Soft 404s. The server returns 200 with a "page not found" or "no results" message. These are invisible to log analysis tools that read only status codes. You must check response content to detect soft 404s.
404 from deleted pages. When a page is deleted without a redirect, Googlebot finds it through external links, bookmarks, or its index. The log shows a 404. Each request is wasted crawl budget.
404 from broken internal links. Internal links pointing to non-existent URLs generate 404s from both users and bots. These are the easiest to fix. A site-wide crawl finds them.
410 Gone. 410 is stronger than 404. It tells Googlebot the page was intentionally removed. Googlebot drops 410 URLs from the index faster than 404 URLs.
Reading 5xx Patterns
5xx responses are the most damaging to crawl performance:
503 spikes. A temporary 503 from server overload or maintenance causes Googlebot to back off. If the 503 persists for hours, Googlebot reduces its crawl rate for days afterward.
500 errors. Persistent 500 errors on specific URLs tell Googlebot those pages are unreliable. Crawl frequency drops. In severe cases, Google may deindex the affected pages.
502/504 from upstream services. These indicate problems with application servers, databases, or CDN origins. The pattern shows in logs as clusters of 502/504 responses from the same IP ranges.
Response Code Distribution Benchmarks
Healthy site benchmarks from a 2026 analysis of 500 sites:
| Code Category | Healthy Range | Warning Level | Critical Level |
|---|---|---|---|
| 2xx | 95-99% | Below 90% | Below 80% |
| 3xx | 1-4% | Above 10% | Above 20% |
| 4xx | Below 1% | Above 5% | Above 10% |
| 5xx | Below 0.5% | Above 2% | Above 5% |
These are aggregate numbers. Isolated 404s on deleted pages are less concerning than 20% of crawl traffic going to 404s.
Diagnostic Workflow
- Export 30 days of log data
- Calculate response code distribution for Googlebot only
- Identify all URLs returning 4xx and 5xx
- Trace redirect chains longer than 2 hops
- Check for soft 404s on high-crawl 200 URLs
- Fix or redirect all error pages
- Verify 5xx root cause with server logs
Note the gap between your current response code distribution and the healthy benchmarks above. Every error page that bots repeatedly hit is wasted crawl budget. Audit response code patterns monthly.
References
- HTTP Status Code Registry by IANA (https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml)
- Google's Guide to HTTP Status Codes (https://developers.google.com/search/docs/crawling-indexing/http-status-codes)
- Screaming Frog Status Code Analysis (https://www.screamingfrog.co.uk/log-file-analyser/)
- Moz Status Code Guide (https://moz.com/learn/seo/http-status-codes)