What Is a Content Audit? A Step-by-Step Process (Keep, Merge, Prune)

A content audit is a page-by-page review of a website that ends in one decision for every URL: keep it, update it, merge it into another page, redirect it, or remove it. The decisions come from data (search impressions, clicks, links, conversions) combined with an honest read of each page’s quality.

Below is the process step by step, with the audit we ran on our own site in October 2026 as the worked example: 17 published pages and posts, every one of them reviewed and given a decision.

What a content audit is for

A content audit answers three questions:

1. Which pages earn their place? Traffic, impressions, links, conversions, or a clear job for visitors.
2. Which pages compete with each other? Two URLs targeting the same search split the signals between them.
3. What’s missing? Gaps the existing pages don’t cover, which become the content plan.

An SEO content audit leans on search data for the first two. But the final call is editorial. Google’s guidance on helpful, people-first content asks whether a page would satisfy someone who arrived at it, and no metric answers that for you.

When to run one

• Before a redesign or migration. You don’t want to rebuild pages you’d have deleted.
• When traffic is flat or falling and you don’t know which pages are responsible.
• When the site has grown for years without anyone deciding what to retire.
• Once a year, as routine.

Ours was triggered by a migration: we were moving digitalusmedia.com from WordPress to a static site, and every URL needed a decision anyway.

Step 1: Build the inventory

List every indexable URL: pages, posts, and also archives, tag pages, feeds and media pages the CMS created on its own. Sources:

• Your CMS export (for WordPress, the XML export under Tools → Export).
• Your XML sitemaps.
• A crawl of the site.
• Search Console’s Page indexing report, to catch URLs Google knows about that your own lists miss.

Our inventory had 11 published pages and 6 posts, plus 2 draft pages, 50 media attachments, category and tag archives, an author archive, feeds and several sitemap files.

Step 2: Collect the data for each URL

One row per URL, with:

Column	Where it comes from
Impressions, clicks, average position (12–16 months)	Search Console Performance report, Pages tab
Main queries	Search Console, filtered by page
Organic sessions and key events	GA4
Referring domains	Search Console Links report or a backlink tool
Internal links pointing to it	A crawler
Word count and last updated date	CMS or crawler
Target topic or keyword	You

For our site, the Search Console data covered 10 December 2025 to 25 October 2026. Five URLs had meaningful impressions:

Page	Impressions	Avg position
/building-digital-authority-and-backlinks/	6,490	73
/technical-seo/	2,515	79
/the-power-of-keywords-in-seo/	1,093	65
/ (home page)	1,006	27
/link-building/	371	67

Every other page had fewer than 250 impressions, and the whole site had 1 click in that period. That’s the honest picture for a small site with no authority: the data can rank pages against each other, but it can’t show which ones “work”, because none really did.

Step 3: Read every page

Numbers find candidates; reading decides. For each page, note:

• Is it accurate and current?
• Is it specific? Does it say something a reader couldn’t get from any other page on the topic?
• Does it overlap with another URL on the same site?
• Does it make claims you can’t back up?

Our pages failed most of these. They were 500 to 650 words each, generic, and written in a recognizably AI-generated style (“Unlocking”, “Elevate”, “Decoding”). Five of the six posts linked to no other page on the site. The home page had placeholder template elements still live: a “John Doe, CEO, Acme Inc.” testimonial and lorem ipsum counters.

Step 4: Decide keep, update, merge, redirect or remove

Every URL gets exactly one of these:

Decision	Use it when
Keep	The page does its job and needs no work now
Update	The topic is right but the content is thin, dated or generic
Merge	Two or more pages target the same topic; combine them into the strongest URL and 301 the others
Redirect	The page has no reason to exist on its own, but a closely related page does
Remove	No visitor value and no related page; return 404 or 410

A few rules of thumb we follow:

• Merge before you delete. If a weak page has links or impressions, fold its useful parts into a stronger page and redirect.
• Redirect to the closest match, not the home page. A redirect to a page on a different topic doesn’t help the visitor who clicked the old link.
• Don’t prune by age. An old page that still answers its question is an asset.

Our decisions

Decision	URLs	Why
Keep (same URL)	5 posts, contact, privacy policy, blog, home	Keep their URLs and impressions; rewrite later
Move (301)	5 service pages and the About page, to /services/<service>/ and /about/	A clear service structure under one hub
Merge (301)	/the-power-of-strategic-link-building/ into /services/link-building/	Same topic and keyword; the post competed with the service page
Redirect (301)	Category, tag and author archives, feeds, old sitemaps	Each to the closest real page, e.g. /tag/keywords/ to the keyword research post
Redirect (301)	13 old image URLs	To the new image files, so image links don’t break
Remove	Placeholder testimonials and counters	Unverifiable; we won’t publish claims we can’t prove
Not migrated	2 draft pages (the old home page and an empty “Clients” page)	Never published, no value

The merge is the decision that matters most for SEO. The 1,230-word post on strategic link building and the /link-building/ service page were chasing the same searches. Search engines had to pick between them, and neither ranked well. One URL now carries the topic.

We also kept the five old posts at their URLs even though the writing is weak, because they hold the site’s only impressions. The post with the most impressions, building digital authority and backlinks, was the first one we scheduled for a full rewrite around the searches it was already appearing for (“authority backlinks” alone had 1,961 impressions). Rewrites of the keyword research and on-page posts are planned for the following weeks.

Step 5: Implement without losing anything

• Redirect with 301s, one hop, straight to the final URL. Google’s redirect documentation treats permanent redirects as a strong signal that the target is the new canonical.
• Never redirect to a redirect. Our migration script refuses redirect chains, so a mistake stops the script instead of reaching visitors.
• Update internal links to point at the final URLs, not the old ones.
• Update the sitemap so it lists only live, canonical URLs.
• Follow Google’s site move guidance if many URLs change at once.

Step 6: Measure the result

Compare the same Search Console metrics for the kept and merged URLs over the following 3 to 6 months, and check the Page indexing report for redirected URLs dropping out and new ones being indexed. Pages Google crawls but doesn’t index after an audit usually need more work on the content itself; our guide to “Crawled - currently not indexed” covers that case.

Content audit vs technical audit

A content audit asks whether each page deserves to exist and what it should say. A technical audit asks whether search engines can crawl, render and index the pages you keep. They overlap at redirects, duplicates and indexing, and it makes sense to run them together: our technical SEO audit checklist is the other half.

What comes after the audit

The gaps you found become the content plan. In our case, the audit showed that every topic the site covered was generic and nothing was based on our own data, so the new plan is built around clusters of specific articles that each support one service page.

If you’d like an outside view on which of your pages to keep, merge or retire, content audits are part of our content marketing service.