Most sites that have been publishing for three or more years are carrying dead weight — articles that were relevant once, near-duplicates written by people who did not know the earlier version existed, and pages that have never received a single visitor. A content audit is the exercise that finds them and decides what happens to each one. It is unglamorous, it takes a day or two, and it routinely produces larger traffic gains than the same effort spent publishing something new.
What It Is and What It Is Not
A content marketing audit is a systematic review of every content page on your site, ending in one of four decisions per page: keep, update, merge, or remove.
It is not a technical SEO audit — that covers crawling, indexing, speed, and site structure, and we cover it in the SEO audit checklist. The two are complementary and often run together.
It is also not a UX content inventory, which catalogues content for a redesign. Same word, different exercise. This one is about performance.
When to Run One
- Before planning a year of content, so you know what already exists
- After a traffic decline you cannot attribute to anything specific
- After a merger, rebrand, or site migration
- When publishing has outrun anyone’s ability to remember what is published
- Annually, as maintenance
If you have never run one and you have more than a hundred content pages, you almost certainly have merge and removal candidates.
Step 1: Build the Inventory
Crawl the site and export every indexable content URL. Screaming Frog’s free tier covers 500 URLs, which is enough for most sites.
You want a spreadsheet with one row per URL and these columns to start:
- URL
- Title
- Publication date and last updated date
- Word count
Exclude utility pages — contact, privacy, cart. You are auditing content, not the whole site.
Step 2: Add Performance Data
This is where the decisions come from. Pull in:
From Search Console (last 12 months): clicks, impressions, average position, and the top query per URL.
From your analytics: sessions, engagement or bounce rate, and conversions or goal completions if you track them.
From your SEO tool: referring domains per URL, and the number of keywords each page ranks for.
Twelve months is the right window. Shorter periods over-weight recent publishing; longer periods include content from a different era of the site.
Step 3: Add Internal Link Counts
Your crawler reports inlinks per URL. This matters more than people expect — pages with zero or one internal link are effectively orphaned, and poor performance may be a linking problem rather than a content problem.
Fixing that is often the cheapest win in the entire audit.
Step 4: Categorise Every Page
Now make a decision on each row. Four outcomes:
Keep As-Is
Performing well: ranking, receiving traffic, converting, accurate. Leave it alone. Note anything worth replicating.
Update
The largest category on most sites, and the highest-return one. Signs a page should be updated:
- Ranking positions 5–20 — close enough that improvement pays off quickly
- Good impressions, poor click-through — usually a title or snippet problem, covered in how to write title tags
- Outdated statistics, screenshots, or references
- Thinner than what currently ranks
- Has backlinks but has slipped in rankings
Updating an existing page with established authority almost always beats writing a new one. A page at position 12 with links already has what a new page would take a year to build.
Merge
Two or more pages covering the same intent. Pick the strongest URL — usually the one with more backlinks, not necessarily the better writing — merge the useful content in, and 301 redirect the others.
This is the fix for keyword cannibalization, and the audit is where you find it. Sort your inventory by top query and look for repeats.
Remove
Some content should not exist. Candidates:
- Zero traffic, zero links, no rankings, no strategic purpose
- Genuinely outdated and not worth rewriting — old event pages, discontinued products
- Off-topic content from an abandoned direction
- Thin pages that dilute topical focus
Be deliberate about the method. Return 410 for content you are deliberately removing, or 301 to a genuinely relevant page if one exists. Do not bulk-redirect everything to the homepage — Google treats that as a soft 404 and it achieves nothing. See 410 vs 404 vs 301 for deleted pages.
Step 5: Decide Using Rules, Not Feelings
Removing content people worked on is emotionally harder than it should be. Rules help. A workable starting set:
| Condition | Action |
|---|---|
| Position 5–20, decent impressions | Update — highest priority |
| Good impressions, CTR below 2% | Rewrite title and description |
| Multiple URLs, same top query | Merge |
| Zero traffic + zero links + 12 months old | Remove |
| Has backlinks but poor rankings | Update, never remove |
| Accurate, ranking, converting | Keep |
Adjust the thresholds to your site. The point is having thresholds at all, decided before you start looking at individual pages.
One rule that should be absolute: never remove a page with backlinks. Update it or merge it, but do not throw away earned authority.
Step 6: Prioritise the Work
You will finish with more actions than time. Order them by expected return:
- Merges resolving cannibalisation on commercial queries — usually the fastest measurable gain
- Title and description rewrites on high-impression, low-CTR pages — hours of work, no new content
- Updates to pages at positions 5–20 on valuable terms
- Internal linking fixes for orphaned pages worth keeping
- Removals — cleanup, real but slower to show
- New content for the gaps the audit exposed
Note that the first four require no new writing at all. That is generally where the return is concentrated, and it is why an audit before a content plan is worth the delay.
Step 7: Record What You Decided
Keep the audit spreadsheet with the decision and date for each URL. Twelve months on, it tells you what was changed and whether it worked — and stops someone re-proposing an article you deliberately removed.
If you maintain a keyword map, update it as you go. The audit is the natural moment to reconcile the two.
What to Expect
Realistic outcomes on a site that has never been audited:
- 10–30% of pages are removal or merge candidates
- A third or more are update candidates
- Traffic often rises after pruning, which surprises people — removing thin content concentrates topical focus and internal link equity on pages that deserve it
- Changes take weeks to show. Google has to recrawl, reprocess redirects, and reassess. Judge it at eight to twelve weeks.
Do not expect a straight line. Consolidation often dips before it improves.
Audit Before You Publish
The instinct when traffic is flat is to publish more. On a site with several years of accumulated content, that is usually the wrong move — you are adding to a pile that already contains competing, outdated, and orphaned pages, and the new article inherits all of that as competition.
An audit reverses the order. Find out what you have, fix what is close to working, merge what is competing, and remove what is dragging. Then publish into a site that is organised enough to support the new content.
The work is tedious and the return is high, which is roughly the opposite of most SEO advice — and the reason it gets skipped.
If you want a full content audit run against your site, including the merge and removal recommendations and the priority order, our SEO services cover exactly that. Get in touch with the team at blogthememachine.com, and subscribe to our newsletter below for more practical guides.
