Run a content audit on a forty-page site or on one with nine thousand URLs and the shape of the problem barely changes. Somewhere in that library sits a small group of pages doing most of the work, a large and neglected middle that could do far better with an afternoon of attention, and a long tail that quietly drags on everything else. A content audit is how you find out which page is which, and then decide what to do about it.
This is not a theory piece. It walks through the process step by step, the columns that belong in your sheet, the calls people commonly get wrong, and what the data and Google's own team say about the parts everyone argues over. If you have ever opened a Search Console export and felt unsure where to even start, this was written for you.
What a content audit actually is
A content audit is a structured review of every piece of content you have published, scored against the goals you actually care about, so you can decide what to keep, improve, combine, or retire. You inventory the URLs, attach real numbers to each one, then read the patterns those numbers make.
People mix this up with a technical SEO audit, which looks at crawlability, page speed, redirect chains, and broken links. Those things matter, but they answer a different question. A content audit is editorial first and technical second. It asks something blunt of every page: is this earning its place, and if not, what is the shortest path to making it earn its place or letting it go?
The one-line version: take stock of what you have, measure it honestly, and act on the difference between what a page is doing and what it could do.
Why the work is worth it
Here is the finding that changes how people budget their time. When HubSpot began systematically updating and republishing older posts, a practice they named historical optimization, they reported lifting monthly organic search views of those posts by an average of 106 percent and roughly doubling the leads those posts generated. Pamela Vaughan, the HubSpot marketer who popularized the approach, has been clear that it works best on blogs with a few years of history and a real archive to mine.
+106% Average lift in monthly organic views of old posts HubSpot optimized (HubSpot) | 2× Roughly the increase in monthly leads from those optimized posts (HubSpot) | 85% Share of one agency's monthly blog traffic traced to old posts (IMPACT) |
The pattern worth internalizing is that the fastest gains are usually hiding in pages you already own. New content takes months to mature. A strong page sitting at position eight can often be nudged onto page one in a couple of weeks with a focused rewrite, which is a very different return on effort than starting from a blank page.
Every page lands in one of four piles
Once the numbers are in front of you, almost every page sorts into one of four decisions. This is the heart of the audit, and it pays to resist the urge to overcomplicate it.
| Decision | When it applies | What you do |
|---|---|---|
| Keep | Solid traffic and rankings, still accurate, and it converts or supports something that does. | Leave it alone and revisit on a schedule. Note which pages link into it so you protect that structure. |
| Update | Ranks around position 5 to 25, earns impressions but thin clicks, or the information has simply aged. | Rewrite for depth and intent, refresh the stats, sharpen the title and opening, then republish and log the date. |
| Merge | Several thin pages chase the same search intent and quietly compete with each other. | Combine them into one strong page, redirect the rest into it, and consolidate the internal links. |
| Remove | No traffic, no rankings, no backlinks, no conversions, and no strategic reason to exist. | Remove it and redirect to the closest relevant page. Only let it return a clean gone status if nothing fits. |
Two habits save people from regret here. Always check a page's backlinks before you touch it, because a quiet page with strong links is an asset in disguise. And when you remove or merge, redirect with intent rather than dumping everything to the homepage.
Every page lands in one of four piles
Once the numbers are in front of you, almost every page sorts into one of four decisions. This is the heart of the audit, and it pays to resist the urge to overcomplicate it.
| Decision | When it applies | What you do |
|---|---|---|
| Keep | Solid traffic and rankings, still accurate, and it converts or supports something that does. | Leave it alone and revisit on a schedule. Note which pages link into it so you protect that structure. |
| Update | Ranks around position 5 to 25, earns impressions but thin clicks, or the information has simply aged. | Rewrite for depth and intent, refresh the stats, sharpen the title and opening, then republish and log the date. |
| Merge | Several thin pages chase the same search intent and quietly compete with each other. | Combine them into one strong page, redirect the rest into it, and consolidate the internal links. |
| Remove | No traffic, no rankings, no backlinks, no conversions, and no strategic reason to exist. | Remove it and redirect to the closest relevant page. Only let it return a clean gone status if nothing fits. |
Two habits save people from regret here. Always check a page's backlinks before you touch it, because a quiet page with strong links is an asset in disguise. And when you remove or merge, redirect with intent rather than dumping everything to the homepage.
The tools worth reaching for
You do not need all of these. Search Console, Analytics, and a spreadsheet will carry a genuinely useful audit. Paid tools mostly buy you speed and backlink data once you are working at scale.
| Tool | What it does in an audit | Cost | Best for |
|---|---|---|---|
| Google Search Console | Clicks, impressions, average position, and indexing status, straight from Google. | Free | The non-negotiable starting point |
| Google Analytics 4 | Engagement and conversions, showing how content performs after the click. | Free | Tying pages to outcomes |
| Screaming Frog | Desktop crawler that inventories every URL with titles, meta, word count, and response codes. | Free to 500 URLs | Building the inventory fast |
| Semrush Content Audit | Reads your sitemap and returns metadata, backlinks, social shares, word count, and last-modified in one view. | From ~$140/mo | Auditing at scale in one place |
| Ahrefs | Top pages by traffic over time, the backlink profile, and spotting content that is decaying. | Paid | Finding decay and links worth saving |
What Google actually says about deleting pages
This is where a lot of advice online gets it wrong. Google's Helpful Content guidance did state that a site carrying a lot of unhelpful content can make its other content less likely to perform, and that removing unhelpful content could help the rest. That system has since been folded into Google's core ranking, so pruning does have a real basis.
The trouble is how people read it. Deleting pages by age is not what Google asks for. When publishers started mass-removing old articles, two people from Google's Search team pushed back plainly.
“Older content can still be helpful, too.”
Danny Sullivan, Google Search Liaison
“Removing content doesn’t make the rest rank higher.”
John Mueller, Google Search Advocate
Sullivan noted that trimming by age is not something Google's guidance encourages. Mueller has been just as direct: removing a genuinely dead page on a very large site can help Google crawl the good pages, but it is not a lever that lifts the whole domain. One example he gives makes the point. An About page gets almost no traffic and rarely changes, yet no one would delete it. Traffic is not the same as value. Hold that distinction and pruning becomes a scalpel instead of a wrecking ball.
Where AI helps, and where to keep it on a leash
AI earns a place during audits, but only for the parts where a second set of eyes is useful and easy to check. The judgment calls stay with the data.
Where it earns its place
• Clustering hundreds of URLs by topic so cannibalization is obvious
• Drafting an updated outline and flagging subtopics a page is missing
• Summarizing what currently ranks so intent gaps show up quickly
• A rough first pass at sorting thin pages from substantial ones
Where to keep it on a leash
• Do not let it invent stats or sources in a rewrite. Every number needs a real citation.
• It sounds confident about rankings it cannot see. Pair it with Search Console before trusting a claim.
• Bulk AI-written pages are the low-quality content pruning is meant to catch. Adding more works against you.
The mistakes worth avoiding
• Deleting by age or low traffic alone. The About page trap. Check rankings, links, and conversions before a page ever reaches the remove pile.
• Skipping the backlink check. A zero-traffic page with authoritative links is worth keeping. Improve it or redirect it, but do not bin it and lose the links.
• Auditing the whole site at once. On a large site this stalls fast. Work through one section or content type at a time and finish what you start.
• Cosmetic updates that change nothing. Swapping the date and a sentence rarely moves anything. Deeper rewrites consistently outperform light touch-ups, as Portent found reviewing blog updates.
• Never measuring afterward. If you do not record the change date, you cannot show what worked, and next quarter you repeat the same guesswork.
• Treating it as a deletion contest. The aim is not a smaller site. It is a site where every page has a reason to be there.
What it is actually worth
A content audit is the highest-return afternoon most sites are not spending. It is not glamorous. It means hours in a spreadsheet, and the wins are quiet ones: a page that climbs from position nine to three, a cluster of four weak posts becoming one that finally ranks, a redirect that stops five pages fighting each other for the same query.
The strongest argument for the work is simple. Updates to pages a site already owns routinely beat brand new content that took weeks to produce. One caution is worth stating plainly: do not turn the audit into a deletion spree. The goal is a site where every page has a reason to be there, not a shorter one.
Run one, act on the easy calls first, write down what you changed, and come back in a month. Do that twice a year and your archive stops being a pile of old posts and starts behaving like an asset.
