Skip to main content
AutoRankFlow

how to do a content audit

How to Do a Content Audit (And What to Do With the Results)

Learn how to do a content audit step by step: build the inventory, pull the right metrics, and decide what to keep, refresh, merge, or delete.

By Raúl Gómez··10 min read

AutoRankFlow research

Quality-scored · intent-matched · transparently published

Key takeaways

  • A content audit is a page-by-page review of your site that ends in one of four decisions per URL: keep, refresh, merge, or delete.
  • Build the inventory from your XML sitemap or a crawler export, then layer in Google Search Console clicks, impressions, positions, and conversion data.
  • The fastest wins are usually refreshes: pages ranking in positions 4–20 with slipping clicks often recover with updated substance and better internal links.
  • Merging overlapping posts consolidates authority and fixes keyword cannibalization; deleting thin, zero-traffic pages cleans up crawl waste.
  • Prioritize by expected impact divided by effort, and re-audit lightly every quarter — content decays faster than most teams assume.

What is a content audit?

A content audit is a systematic review of every indexable page on your site, scored against traffic, rankings, engagement, and quality, ending in a decision for each URL: keep it, refresh it, merge it, or delete it. The output is a prioritized action list, not a report that sits in a folder.

People use the term loosely, so two distinctions matter. First, an inventory is just the list of URLs; the audit is the analysis and decisions you apply to that list. You cannot have the second without the first. Second, a content audit is not a technical SEO audit. Technical audits look at infrastructure — crawlability, indexation, site speed, schema. A content audit looks at substance: is this page accurate, useful, aligned with search intent, and earning its place in the index?

The audit combines quantitative data (Search Console performance, analytics, backlinks) with a qualitative read (accuracy, depth, originality, tone). Neither half works alone. Data tells you a page gets no clicks; only reading it tells you whether the fix is a rewrite, a merge, or the fact that nobody searches for that topic anymore.

Why should you do a content audit?

Because publishing more content rarely fixes a weak library. Most pages never earn organic traffic at all, and pages that once performed decay as competitors update and search intent shifts. An audit tells you where effort will actually pay back.

Two numbers frame the problem. BrightEdge's channel research found that organic search drives 53% of trackable website traffic — for most small businesses it is the single largest acquisition channel. Yet when Ahrefs analyzed roughly 14 billion pages from its Content Explorer database, 96.55% received zero organic traffic from Google, and another 1.94% got between one and ten visits a month. Most of what gets published disappears without a trace.

The flip side is that maintenance works. Orbit Media's annual blogger survey has repeatedly found that bloggers who update old articles are about twice as likely to report strong results from their content. The pages closest to winning usually need upkeep, not replacement — and an audit is how you find them.

Audits also surface problems that quietly compound: five posts competing for the same keyword, outdated claims that create compliance risk in regulated industries, broken internal links, and thin pages that dilute your site's overall quality signals. Left alone, each is minor. Together, they drag down the pages that actually make you money.

How do you build a content inventory?

Export every indexable URL from your XML sitemap or a crawler such as Screaming Frog (free for up to 500 URLs), then organize the list in a spreadsheet with one row per page. This inventory is the backbone that every later decision hangs on.

  1. Export your URLs. Pull the XML sitemap first — it shows what you intend to be indexed. A crawler adds what is actually reachable, including orphan pages you forgot existed.
  2. Add page metadata. Title tag, H1, word count, publish date, last-updated date, and content type (post, page, product, category).
  3. Pull performance data. From Google Search Console, export 12–16 months of clicks, impressions, and average position per page. From GA4, add sessions, engagement, and conversions.
  4. Add link data. Referring domains per URL from Ahrefs, Moz, or a similar index. This matters most when you consider deleting anything.
  5. Assign a target keyword and intent. Note what each page is supposed to rank for and whether the page still matches what searchers want. Pages with no target keyword are audit red flags by default.
  6. Flag technical basics. Indexable or noindexed, canonical status, and HTTP status code. A page that redirects or errors out gets handled differently from a live one.

For a site under a few hundred pages, this is a day of work. Resist the urge to perfect the spreadsheet — you need enough structure to make decisions, not a museum exhibit.

Which metrics should you track for each page?

You need three layers: visibility (impressions, average position), traffic and engagement (clicks, sessions, time on page), and business value (conversions or leads). Add backlinks and the last-updated date, and you have everything required for keep, refresh, merge, and delete calls.

MetricWhat it tells youWhere to get it
Clicks and impressions (12–16 months)Whether the page gets seen and chosen in searchGoogle Search Console
Average positionHow close the page is to page oneGoogle Search Console
Sessions and engagementWhether visitors actually read the pageGA4
Conversions or leadsWhether the page earns its keep commerciallyGA4 or your CRM
Referring domainsWhether deleting the page would waste link equityAhrefs, Moz, or similar
Last-updated dateHow stale the substance likely isYour CMS
Word countA quick thin-content screen (not a quality score)Crawler export

Two cautions. First, do not judge pages on traffic alone — a low-traffic page that converts at 10% is a keeper, while a high-traffic page that never converts might just be answering a question your buyers never ask. Second, treat word count as a screening tool only. Short is not automatically thin, and long is not automatically good.

How do you decide what to keep, refresh, merge, or delete?

Score every page on performance and quality, then apply four rules. Keep pages that perform and are accurate. Refresh pages with real rankings but slipping traffic or dated substance. Merge pages competing for the same keyword. Delete or noindex pages with no traffic, no links, and no strategic role.

SignalDecision
Ranks well, converts, still accurateKeep — and protect it
Positions 4–20, clicks declining, content datedRefresh
Multiple weak pages targeting one keywordMerge into the strongest URL, 301 the rest
Zero traffic for 12+ months, no backlinks, no strategic valueDelete or noindex
Thin but genuinely needed for users (contact, policies)Keep for users; noindex if it should not rank

Keep

Leave winners alone, but note their last-updated dates. The most common audit mistake is tinkering with pages that already work and breaking what made them rank. Your only job for keepers is to schedule their eventual review.

Refresh

This is where most of the value sits. Update facts, statistics, screenshots, and dates; expand sections that are thinner than what now ranks; re-align the page with current search intent; and add internal links from newer, relevant pages. If a page is losing clicks month over month, that pattern has a name — content decay — and there is a full walkthrough on how to fix content decay before it costs you the ranking entirely. Run every refresh candidate through a consistent standard as well: our content quality checklist covers structure, evidence, and originality, so a refresh becomes a genuine improvement rather than a cosmetic date change.

Merge

When two or more pages target the same keyword, they split signals and usually both lose. Pick the strongest URL — the one with the most backlinks and best historical position — fold the best sections from the weaker pages into it, then 301 redirect the losers and update internal links to point at the survivor. Consolidation feels destructive. It is almost always additive.

Delete

Delete only pages with zero traffic for a year or more, no referring domains, and no strategic purpose. If a doomed page has any backlinks or residual impressions, 301 it to the closest equivalent instead. One honest caveat: on small sites, pruning rarely produces dramatic ranking jumps. The real benefit is a cleaner crawl and a clearer topical focus — worthwhile, but not magic.

How do you prioritize what to fix first?

Rank every action by expected impact divided by effort. High-impact, low-effort work — refreshing pages already in positions 4–20, reversing decay on former top performers, merging obvious cannibals — goes first. Full rewrites and deletion batches come later.

A simple scoring model works fine. For impact, estimate traffic potential (search volume of the target keyword times a realistic click share at the position you could reach) and multiply by business proximity (does this page sit near a conversion?). For effort, estimate hours honestly: a statistics-and-links refresh might take two hours; a full re-optimization might take six. Score each action, sort descending, and work the list.

Three categories reliably float to the top: pages stuck at positions 4–10 that need a push to the top three, pages that used to perform and have decayed, and high-converting pages with mediocre rankings. Categories that reliably sink to the bottom: full rewrites of pages targeting keywords you have no authority for, and deletion sprees motivated by tidiness rather than data.

Be realistic about the execution half. The audit for a 300-page site is a long weekend of spreadsheet work; acting on it is weeks or months of content production. The ongoing measurement is also easy to abandon — checking whether each refresh actually recovered requires returning to Search Console at 30, 60, and 90 days. That monitoring loop is where automation earns its keep: AutoRankFlow, for example, tracks GSC performance continuously, flags decaying pages as they slip, and shows whether refreshed URLs recover, which is the part a static spreadsheet cannot do on its own.

How often should you run a content audit?

Run a full audit once or twice a year, with a light quarterly check of your top 20–30 pages. Sites publishing weekly should lean quarterly for full audits; a small brochure site can stretch to annually without much risk.

Certain events should trigger an unscheduled audit regardless of the calendar: a sharp traffic drop after a core algorithm update, a site migration or redesign, a rebrand or repositioning, or acquiring a site with an unknown content history. In each case, the inventory-and-decide process is the same — you are just running it under time pressure.

Between audits, something needs to watch for decay, because decay does not wait for your schedule. Manually, that means a monthly Search Console review of pages with falling clicks. If you would rather not babysit spreadsheets, AutoRankFlow runs this monitoring in the background — decay detection plus weekly reports on what is gaining and losing ground — so the quarterly check becomes a confirmation instead of a discovery exercise.

Frequently asked questions

How long does a content audit take?

For a site under 100 pages, expect one to two focused days for the inventory, data pulls, and decision pass. Larger sites take a week or more, or you can audit section by section starting with your highest-traffic directory. Executing the resulting actions is a separate, longer phase.

Can I do a content audit without paid tools?

Yes. Google Search Console, GA4, and a sitemap export cover the essentials, and Screaming Frog is free up to 500 URLs. Paid link indexes add backlink data that matters for delete decisions, but you can run a solid first audit without them.

Should I delete old blog posts?

Not for being old. Delete only posts with zero traffic for 12-plus months, no backlinks, and no strategic or user-facing value. Age alone is a reason to refresh, not to remove — older URLs often carry authority that new ones lack.

Will deleting content hurt my SEO?

Usually not, provided you redirect anything with backlinks or residual impressions to the closest equivalent. Deleting pages that still earn traffic or links absolutely can hurt, which is why every delete decision should be checked against Search Console and backlink data first.

What is the difference between a content audit and a technical SEO audit?

A technical audit evaluates infrastructure: crawlability, indexation, speed, structured data. A content audit evaluates substance: quality, relevance, intent match, and performance. They are complementary, and it usually makes sense to fix critical technical issues before auditing content.

How is a content audit different from a content inventory?

The inventory is the raw list of URLs and their attributes. The audit is the analysis you layer on top — performance data, quality assessment, and the keep, refresh, merge, or delete decision for each row. The inventory is a means; the audit is the point.

What should I do right after the audit?

Turn the decisions into a backlog with owners and deadlines, ship the top quick wins within two weeks, and calendar measurement checks at 30, 60, and 90 days in Search Console. An audit without scheduled follow-through is just an expensive list.

Build your own search growth system

See the competitors, keywords and content opportunities available for your website.

Analyze my website →