Loading tutorials…
Loading tutorials…
Your Health Score is 68 and the issue list has 4,000 items. This walks through the 10 errors that matter most — the ones causing real ranking damage — and how to fix each.
Who this is forMarketers staring at a Site Audit report with hundreds of issues, paralyzed by where to start. If you've been 'going to fix the audit' for three months, this is the priority list.
What you'll need
Step 1
Open Site Audit → Errors → 4XX page. Cross-reference with Broken backlinks. 301 each 4xx with inbound links to a relevant successor.
Open Site Audit → All issues → filter to '4XX page' under errors.
For each 4xx URL, check Ahrefs Broken backlinks (Site Explorer) to see if external links point at it. URLs with inbound links are the priority.
301-redirect each 4xx with backlinks to the closest relevant existing page (NOT homepage — soft 404 risk).
4xx URLs with no backlinks and no traffic can be left as 404s — Google de-indexes naturally.
Re-crawl after fixes. Expect Health Score lift of 3-8 points.
Step 2
Site Audit → Errors → Redirect chain. Collapse chains (A → B → C) into direct redirects (A → C).
Open Errors → 'Redirect chain' issue. Ahrefs lists every chain with >1 hop.
For each chain, identify the final destination. Update the first redirect to point directly there.
Common cause: years of incremental redirects (A → B in 2022, then B → C in 2024). The chain accumulates silently.
Redirect loops (A → B → A) cause crawl failures and traffic loss. Fix immediately — these are emergencies, not optimization.
After fixing chains, re-crawl. Health Score lift of 2-5 points typical.
Step 3
Site Audit → Errors → Broken link (internal). Update or remove each broken link on the source page.
Open Errors → 'Broken link' (filter to internal).
For each broken link, decide: (a) update the link to a working URL if a successor exists, (b) remove the link entirely if the target is sunset.
Internal broken links hurt crawl efficiency — Googlebot wastes budget on dead-end paths. They also hurt user experience metrics.
Common source: removed-product pages still linked from category nav, old blog posts referencing deleted content.
Use bulk-edit tools (WordPress Better Search Replace, Next.js codemod, or sed-based scripts for static sites) to fix patterns at scale.
Step 4
Site Audit → Errors → Canonical to 4XX or 5XX. Point each canonical at a working URL or remove the canonical tag.
Open Errors → 'Canonical points to 4XX' (or 5XX).
For each affected URL, check the rel='canonical' tag in the HTML head. If it points to a dead URL, Google won't index either page properly.
Fix: update the canonical to a live URL, or remove the canonical tag if it's not needed.
Common source: bulk-deleted pages whose canonicals still reference each other (set up before deletion).
Critical impact: pages with broken canonicals can drop out of the index entirely — this is a ranking emergency.
Step 5
Site Audit → Warnings → Noindex page. Audit which pages should genuinely be noindexed; remove the tag from any that shouldn't.
Open Warnings → 'Noindex page.' Review the list.
Legitimate noindex: thank-you pages, internal search results, staging URLs, /tag/ archives (often).
Mistakes: money pages, blog posts, and category pages with noindex usually got there from a misconfigured plugin or staging-import.
For each mistake, remove the noindex meta robots tag (CMS settings or template fix). Re-submit to GSC for re-indexing.
Critical impact: a noindex on a money page silently removes it from search results. Easy to miss for months.
Step 6
Site Audit → Issues → Title (missing, duplicate, or too long). Write unique 50-60 char titles for each affected page.
Open Issues → filter for title-related: 'Missing title,' 'Duplicate title,' 'Title too long,' 'Title too short.'
Missing titles: write a 50-60 character title that includes the primary keyword for the page.
Duplicate titles: rewrite each to be unique. Most common cause: category pages with paginated URLs sharing the same title — fix by templating page number into the title.
Too long: Google truncates at ~60 characters. Rewrite to fit.
Bulk fix: most CMSs support templated titles. Set the template at the type level (post, category, product) before editing individual pages.
Step 7
Site Audit → Issues → Meta description issues. Write unique 140-158 char descriptions for indexable pages.
Open Issues → filter for meta description: 'Missing meta description,' 'Duplicate meta description,' 'Meta description too long.'
Missing: write 140-158 character descriptions that include the keyword + a value prop. Google may rewrite, but having one is still important.
Duplicate: rewrite each. Like titles, most duplicate metas come from paginated category pages or templated thin content.
Don't waste time on noindex pages — only fix metas on pages that should be in the index.
Step 8
Site Audit → Performance → flag pages with LCP > 2.5s, CLS > 0.1, or INP > 200ms. Fix the top 10 by traffic.
Open Performance (or 'Speed' depending on Ahrefs UI version). Filter to pages with LCP > 2.5s.
Cross-reference with Ahrefs Site Explorer → Top pages to find which slow pages have traffic.
Fix the top 10 by traffic first. Common fixes: optimize hero images (WebP, lazy-load), defer non-critical JS, use a CDN, remove render-blocking CSS.
Run each fixed page through PageSpeed Insights or Lighthouse to confirm CWV move.
Critical impact: slow pages lose 5-20% of rankings on competitive queries. Top 10 fixes typically recover meaningful traffic in 30-60 days.
Step 9
Site Audit → Issues → Thin content (under 500 words). Either expand each thin page to 800+ words or noindex/delete it.
Open Issues → 'Thin content' or 'Low word count.'
For each thin page: decide (a) expand to 800+ words with substantive content, (b) merge into a parent page with 301, (c) noindex it (if it's necessary navigation but not search-worthy), or (d) delete it.
Don't 'pad' thin content to hit word count — Google penalizes filler. If you can't expand substantively, the page should be noindexed or removed.
Critical impact: many thin pages dilute topical authority. Removing 50 thin pages can lift the remaining 200 pages' rankings noticeably.
Step 10
Site Audit → Issues → Orphan pages with traffic. Add at least one internal link from a relevant parent page.
Orphan pages = indexed and getting traffic but not linked from any internal page.
Open Issues → 'Orphan page (in sitemap)' or 'Orphan page (has organic traffic).'
For each orphan with real traffic, identify the most relevant parent page and add an internal link to the orphan.
Internal links signal importance — orphan pages with traffic are 'accidentally ranking' and would rank higher with proper internal linking.
Critical impact: typically 5-15% ranking lift on orphan pages within 30-60 days of adding internal links.
Common mistakes
Fixing low-impact issues first
What goes wrong: You spend 3 weekends fixing missing alt-text on 800 images while broken canonicals on your top 10 pages remain. The Health Score goes up but rankings don't. Pure vanity work.
How to avoid: Always sort issues by 'URLs affected' and impact. 4xx with backlinks, redirect chains, noindex on money pages, and broken canonicals come first — every time.
Treating warnings like errors
What goes wrong: You fix every yellow warning (multiple H1s, short meta descriptions, no schema) while red errors sit untouched. The audit's prioritization gets inverted.
How to avoid: Errors before warnings, always. Most warnings are stylistic; most errors are ranking-impactful. Trust the severity sort.
Not re-crawling after fixes
What goes wrong: You ship 20 fixes but don't re-crawl. The Health Score still shows the old issues. You can't tell what's actually fixed vs what's broken in a new way.
How to avoid: After every batch of fixes, trigger a re-crawl. Confirm issues drop off the list before moving to the next batch.
Fixing one off but not patterns
What goes wrong: You fix 50 individual broken internal links pointing at /old-product, missing that there's a category template still generating these links daily. Whack-a-mole forever.
How to avoid: When you see a pattern in errors (same target URL, same source template, same regex), fix at the template/source level, not URL by URL.
Ignoring the Health Score trend
What goes wrong: Health Score drops 10 points after a deploy and you don't notice for two weeks. The deploy regression compounds. Fixing it later takes 3x the effort.
How to avoid: Set Site Audit alerts for Health Score drops > 5 points. Review weekly. Catch regressions inside the deploy cycle, not after.
Recap
Done — what's next
How to set up Ahrefs Site Audit the right way
Read the next tutorial
Hand it off
Site Audit fixes are a one-time sprint that should turn into a quarterly habit. A vetted technical SEO specialist on EverestX will close the current audit, set up regression alerts, and own the recurring review — typically $400-1,000/mo at $14-16/hr.
See specialist rates
4-8 hours of focused work for a 5K-URL site. 12-20 hours for a 50K-URL site. The first audit is the hardest; quarterly maintenance after that is 2-4 hours per pass.
Top 10 categories of issues, sorted by URLs affected, deliver 80%+ of the impact. The long tail of one-off issues rarely moves rankings. Don't chase a 100 Health Score — chase ranking lift.
Critical fixes (noindex removal, broken canonical fixes): 1-3 weeks after re-crawl. Performance fixes (CWV): 30-60 days. Content depth changes: 60-90 days. Internal linking changes: 30-45 days.
Only if you intend to index them later. Otherwise it's wasted effort. Confirm noindex status is intentional first, then ignore other issues on those pages.
Ahrefs flags everything; you decide what's worth fixing. For example, 'Multiple H1' is flagged but is fine in modern HTML5. Treat the audit as a checklist, not a mandate.
Ahrefs
Site Audit only earns its keep when the crawl actually mirrors how Googlebot sees you. This walks through the project + crawl settings that 80% of DIY setups misconfigure on the first pass.
Ahrefs
Every site has dead backlinks pointing at 404'd pages. Reclaiming them is the highest ROI link work in SEO — the links already exist; you just have to redirect or replace the destination.
Ahrefs
You're paying $249-449/mo for Ahrefs. The question isn't whether the tool is worth it — it's whether you're using more than 10% of it. This is the honest decision framework.