The SEO Link Audit Checklist That Actually Matches What Google Sees
A practical SEO link audit checklist for internal links, external links, and rel policy—without guessing from view-source alone.
In this guide7 sections
Most “link audits” start with a crawl export and a spreadsheet. That works until the page you care about is a guest post, a client site, or a SPA that only exposes real links after JavaScript runs. Then the spreadsheet and the rendered page disagree — and the disagreement is where money walks out of the room.
A useful SEO link audit answers three questions for every link on a URL you care about: is it internal or external, what rel policy does it carry (follow, nofollow, sponsored, ugc), and does the placement still match what you paid for or negotiated? LinkGlow’s analyzer is built for that page-level truth: paste a URL, get a server-side render, and classify the links captured from the rendered DOM — not a guess from raw HTML.
This is not a substitute for a whole-site crawler. If you need every URL on a 40,000-page property, use a crawler. Use this checklist when one page is the contract: a guest post, a digital PR hit, a money page’s internal graph, a client homepage after a template ship.
What “matches what Google sees” actually means
Google does not rank your view-source tab. It ranks (among other things) the links a browser ends up with after redirects, HTTPS upgrades, and client-side rendering. Affiliate modules, cookie banners, and “open external links in a new tab” plugins routinely rewrite rel after hydration. If your audit stopped at the first HTML response, you can certify a follow link that visitors never get.
A page-level audit that renders the URL is how you inspect the same graph a person — and most automated browsers — end up with. It still will not see links behind a login, a paywall, or a geo-block you cannot fetch. Say that out loud in the report.
What to capture on every page audit
- The final URL after redirects, not only the URL in the brief
- Full absolute hrefs, not leftover relative paths
- Internal vs external classification that respects your multi-domain properties
- rel tokens as Google documents them: nofollow, sponsored, ugc, and combinations
- Anchor text and enough surrounding context to prove it is this placement
- Whether the link sits in the article body or in nav, footer, or related-posts chrome
- Broken destinations (4xx/5xx after following redirects) when you run a health check
- Title, canonical, and robots from the real
<head>— they change how equity can flow - A shareable snapshot so a client can see the same page you audited
If you skip region (body vs chrome), you will eventually “verify” a related-posts widget and call it a placement. That is how false calm reports get written.
Internal links are not “set and forget”
Internal linking still moves crawl budget and equity between URLs you control. An audit that only counts external links will miss orphaned hubs, accidental nofollow on nav, and content blocks that never link to money pages. Treat internal links as first-class: same classification, same broken-link pass, same before/after when templates ship.
When a redesign lands, re-run the money page and the two hubs that are supposed to point at it. The failure mode is rarely “zero internal links.” It is “the links moved into a nofollow utility nav” or “the hub now points at a filtered listing that 404s.”
Domain groups: when “external” is really you
Agencies and multi-brand teams routinely own several hostnames. If your tool always treats other brands as external, your internal metrics lie. Domain groups in LinkGlow let you mark sister properties as internal for classification so dashboard stats match how you actually think about the graph.
Do not mark a partner or a paid publisher as internal. Internal means you control the site. Partners belong in a group that organizes them without flipping link type.
A checklist you can run in under ten minutes
- Open the live URL, not a staging host that never matches production.
- Paste it into LinkGlow (landing demo or a signed-in audit).
- Confirm the final URL and title match the page you meant.
- Scan the Glow view for unexpected nofollow, missing placements, or links only in chrome.
- Filter the links list to external + follow and external + nofollow.
- Open Page SEO for title, canonical, and robots surprises.
- If this is a sold placement, check rel on that URL and save it as a tracked link.
- If destination health matters, run the broken-backlink checker on the same page.
- Export CSV/PDF only after the Glow view matches the story you will tell the client.
What to write in the client note
Do not write “the page is fine.” Write: how many links were captured, how many external follows, whether the sold href is present in the article body, what rel it ships, and whether destination HTTP is clean after redirects. If you did not check a thing, omit it. Inventing completeness is worse than a short report.
Next step
Run a free analysis on a page you already know well. If the classified links match what you expect, scale the same workflow to client placements. If they do not, you just found the gap between “we bought a link” and “the page still ships the link.” Then read placement monitoring for linkbuilders before you promise a retainer that assumes the page never changes.
Next useful step
Run this workflow on a real page.
Paste a URL into the free analyzer, then save the audit when you need the result to stay useful.