The Complete Guide to Content Audit

Key takeaways

• A content audit inventories every page on a site and scores it against real performance data, not a gut feeling about what's good.
• The four outcomes for any page are keep, update, merge, or remove, and most audits find pages belong in the last two more often than expected.
• Thin, outdated, or duplicate content actively drags down a site's overall quality evaluation, not just the individual page.
• Traffic and rankings alone aren't enough to judge a page, conversion data and search intent match matter just as much.
• Run a full content audit at least once a year, and a lighter check quarterly on the highest-traffic pages.

What a content audit actually involves

A content audit starts with a complete list of every published page on a site, usually pulled from a crawler or the site’s own sitemap, cross-referenced against analytics and search console data for each one. The goal isn’t just counting pages. It’s scoring each one against traffic trend, rankings, conversion contribution, and how well it still matches what people searching that term actually expect to find.

A page that ranked well two years ago but has since been outpaced by a competitor’s more current, more thorough page needs a different fix than one that never ranked at all.

Building the inventory, the unglamorous first step

Pull every URL from the site, including pages that don’t show up in the main navigation, since forgotten pages from an old campaign or an abandoned blog category are often exactly where the biggest problems hide. For each page, record the publish date, last update date, current organic traffic, current rankings for its target keyword, and whether it still has internal links pointing to it from anywhere else on the site.

This step is tedious and there’s no way around that. A spreadsheet with one row per URL is the standard tool, not anything fancier. On a site over a couple hundred pages, budget a full day just for the crawl and export before any actual analysis starts.

The metrics that actually matter

Traffic trend over the past 12 months tells you whether a page is growing, flat, or declining, which matters more than its current absolute traffic number. Rankings for the target keyword show whether the page still competes for what it was built to rank for.

Conversion rate or goal completions show whether traffic is actually doing anything for the business, since a page pulling in visitors who never convert is a weaker asset than a lower-traffic page that consistently converts. Search intent match, checking whether the current top-ranking pages for that keyword still look like this page, catches cases where the market shifted format or angle since the page was written. Weigh all four together rather than picking a favorite metric. A page can look strong on traffic and still be quietly failing on conversions, and neither number alone tells the whole story.

The four outcomes: keep, update, merge, or remove

Keep applies to pages performing well against all four metrics above, don’t touch what’s already working. Update applies to pages with the right topic and decent traffic but content that’s gone stale, outdated statistics, an old screenshot, a pricing table nobody’s touched in two years.

Merge applies to two or more thin pages competing for overlapping keywords, better combined into one comprehensive page than left splitting traffic and confidence between them. Remove, with a redirect to the most relevant remaining page, applies to genuinely thin or duplicate content that isn’t worth the effort to fix and is actively pulling down the site’s overall quality signal.

Why old content drags down the whole site

Google evaluates quality signals across a site, not purely page by page, which means a large volume of thin, outdated, or abandoned content can suppress how well the genuinely good pages on the same site perform. This is the part of a content audit most site owners underestimate.

It’s tempting to leave an old page alone because deleting it feels like losing an asset, but a page nobody’s updated in four years, that never ranked and never converts, isn’t an asset. It’s weight the rest of the site is carrying.

Prioritizing which pages to fix first

Start with pages that were once strong and have declined, since those usually need the least work to recover and carry proof they can rank. Move next to pages targeting valuable keywords that never quite broke through, checking whether a content or intent mismatch explains why. Save the clear removal candidates for last, since deleting or merging them is quick once decided and doesn’t require new writing.

Trying to fix everything at once usually means nothing gets finished. Working through a prioritized list, ten or fifteen pages at a time, actually gets the audit’s findings implemented instead of filed away. A batch of ten completed updates beats a plan for two hundred that never gets past the spreadsheet.

Who should run the audit

A single person familiar with the site’s analytics and search console data can run a content audit alone on a smaller site, usually under 150 pages, in a few focused days. Larger sites benefit from splitting the work: one person builds the inventory and pulls the data, another makes the keep or cut decisions, since a second set of eyes catches emotional attachment to pages that should probably go.

An outside team auditing the site brings less bias toward pages someone on staff wrote and is proud of, which is often the exact reason a low-performing page survives three audits in a row without anyone flagging it.

How often to run a full audit

Once a year is the baseline for most sites, timed before an annual content planning cycle so the findings actually feed into next year’s calendar instead of sitting unread. Sites publishing frequently, more than a few posts a month, benefit from a lighter quarterly check on just the highest-traffic pages, catching decline early instead of a year into it.

A site that’s never run one before should do a full audit now, regardless of when the last content went up, since the backlog only gets larger and harder to work through the longer it waits. Waiting another year usually just means auditing twice as many pages with half as much memory of why each one was written the way it was.

Turning findings into an actual plan

An audit that produces a spreadsheet and nothing else wasted the time it took to build. Convert the findings into a scheduled plan: which pages get updated this month, which get merged next month, which redirects go live this week.

Pair the audit with a broader content marketing strategy so new content gets planned around the gaps the audit revealed, rather than repeating whatever pattern created the backlog in the first place. The audit only pays off once its findings turn into scheduled work, not once it’s presented in a meeting.

More content & copywriting guides