You’re not wrong.
Most content audits are a waste of time.
Not because auditing is a bad idea — it is the right idea, and skipping it is one of the most expensive mistakes a content team can make. Because most content audits are designed around the wrong question. The honest assessment of why they fail is grimmer than most audit guides acknowledge.
The wrong question is: what do we have?
The right question is: what is working, what isn’t, and what is the highest-leverage thing we can do about it right now?
The first question produces a spreadsheet. The second produces a plan. This guide is about producing the plan.
Before You Pull a Single URL: Define What You’re Auditing For
An audit without a purpose is a census. It tells you what exists. It does not tell you what to do.
Before you touch a tool or pull a URL, decide what problem you are auditing for. The four main types are not mutually exclusive, but trying to do all of them at once produces the overwhelming spreadsheet that kills the audit before anyone acts on it.
Performance audit: What is driving traffic and what isn’t? What should be doing better than it is?
Quality audit: What meets our editorial standards? What is embarrassing to have on the site?
Technical audit: Where is the architecture broken — orphaned pages, keyword cannibalization, outdated evergreens, redirect issues?
Strategic audit: Does the content library reflect what we are actually trying to build? Are we covering the right topics for the right audience?
Pick one. Do it. Act on the findings. Then do the next one. The technical debt version of this problem is where most large, aging content libraries end up — and it compounds.
Step One:
Pull A URL Inventory
Export every indexed URL from Google Search Console. If the number is alarming, do not be alarmed — you are filtering, not reading every page.
Add any live pages not appearing in GSC. These may have indexing issues worth flagging separately.
This is your universe. Everything else is filtering.
Step Two:
Pull Performance Data
For each URL, pull from GSC: impressions, clicks, average position for the trailing ninety days. From GA4: sessions, engagement rate, average engagement time.
Sort the results into four buckets:
Working well: Good traffic, good engagement, ranking in positions one through ten. Leave these alone. Do not edit content that is performing well on the instinct that you can make it better. You will usually make it worse.
Underperforming despite visibility: Reasonable impressions, poor click-through rate. This is a title and meta description problem, not a content problem. Fix the title and meta first before touching the content itself.
Getting traffic but not keeping it: Sessions arriving, low engagement time, high bounce rate. The content is findable but failing the reader. Content quality or audience mismatch. Worth a rewrite.
Invisible: Low or zero impressions. Do not spend time here until you understand why — technical issue, wrong audience targeting, not indexed, or just not competitive. Each requires a different response.
Step Three:
Do the Human Read
This is the step most audits skip because it does not scale. It is also the step that produces the most valuable findings.
Select your top fifty pages by traffic. Read them — not skim. Read them the way a reader who found them through search would read them.
Ask of each one: does this actually answer the question that brought the reader here? Is the answer easy to find? Is the information accurate? Does it reflect the current brand voice? Is there something specific and valuable here, or could this have come from anywhere?
Assign each one to a category: keep as-is, update, rewrite, consolidate, or remove. This is your action list.
Step Four:
Structural Diagnosis
With performance data and the human read complete, look at the structural issues.
Keyword cannibalization: Multiple pages competing for the same or similar queries split the authority that should accrue to one strong piece. Find these and consolidate — one authoritative piece, the others redirected or rewritten to cover distinct adjacent angles.
Orphaned pages: Cross-reference your URL list against your internal links. Pages with no internal links pointing to them are structurally isolated. Either connect them to the architecture with appropriate links or decide they should not exist.
Outdated evergreens: Pages published more than two years ago still generating meaningful traffic. Pull these for a freshness review — updated information on an established page is one of the highest-leverage interventions available. The authority is already there. Use it.
Step Five:
The Output
That Gets Used
The deliverable of a useful content audit is a prioritized action list in three tiers:
This week: Three to five specific pages to update or fix. Chosen for low effort and high potential impact. Quick wins that produce momentum.
This quarter: The larger consolidations, rewrites, and structural fixes. No more than ten items, ranked by expected impact.
Next planning cycle: Strategic gaps — topics that should exist but don’t, audience segments underserved, structural investments that require significant resources.
Hand someone a spreadsheet of two hundred URLs and you will get a nodding meeting and no action. Hand someone five things to do this week and you will get five things done.
The audit’s job is not to be comprehensive. It is to produce the next right action. Produce that.
What an Audit Tells You
That the Data Doesn’t
Numbers tell you what happened. The human read tells you why.
The most useful findings from a content audit are almost always qualitative, not quantitative: the piece that is ranking for the wrong reason, the topic that is covered seven times without depth, the question the audience is asking constantly that the archive never answers, the piece that is genuinely good and structurally isolated and could perform much better with ten minutes of internal linking.
The content that never gets made — the gap between what the data shows and what it cannot show — is often where the real audit finding lives. Stay curious about what is not in the spreadsheet. For a real-world example of how this plays out, see the content audit case study — and for the specific situation that taught me where most audits go wrong, The Gym Cancel Audit.
How to Do
a Content Audit,
Step-by-Step
- Define what you are auditing for (performance, quality, technical, or strategic)
- Pull the full URL inventory from Google Search Console
- Pull performance data from GSC and GA4, sorted into four performance buckets
- Do the human read of your top fifty pages, categorizing each for action
- Run the structural diagnosis: cannibalization, orphans, outdated evergreens
- Produce a tiered action list (this week / this quarter / next cycle)
The audit tells you what needs work. Before anything goes back to publishing, the authority checklist that tells you whether a piece is ready to publish is the standard worth holding each updated piece to — covering the E-E-A-T signals, AI citation readiness, and editorial accountability that determine whether a revised piece will actually perform.
Running a content audit without editorial authority behind the recommendations is a common failure mode — which is often when bringing in a fractional managing editor is the right move. Before committing to that hire, the operational diagnostic that should precede any editorial hiring decision is worth running first. The SEO guide for content teams covers the performance signals the audit should surface.
A focused audit of fifty to one hundred pages takes three to five days for a single practitioner. A full audit of a large library (several hundred pages) takes two to four weeks. Trying to do it faster produces a spreadsheet instead of a plan.
A full performance audit annually. A targeted audit of your top pages (by traffic or by strategic importance) every quarter. A structural check for new orphaned pages and cannibalization issues after any major publishing push.
Google Search Console and GA4 are sufficient for most content audits. Screaming Frog helps with crawl-based structural audits. Ahrefs or Semrush add keyword and backlink data. Start with what you have — GSC and GA4 alone produce most of the actionable findings. For how AI changes the audit workflow specifically, see The Content Audit in the Age of AI.
A content inventory lists what exists. A content audit evaluates it — performance, quality, structural fitness — and produces recommendations. The inventory is a step in the audit, not the audit itself.
Depends on why. Low-traffic but high-quality content may need better internal linking or title optimization. Low-quality content that could be improved should be rewritten. Thin, redundant content should usually be consolidated with related pieces and redirected. Truly irredeemable content with no traffic and no strategic value can be removed.
Jacob Clifton is the principal of Clifton Creative, an editorial strategy consultancy based in Austin, Texas. He spent fourteen years as a flagship staff writer at Television Without Pity and has written for Tor.com, Vulture, BuzzFeed News, and the Austin Chronicle.