Content audit template

A content audit answers one question per page: keep, update, consolidate, redirect or delete. The template works that out from your own figures instead of your instinct.

The method

1. Export every URL. Search Console for the last 12 months, or your sitemap if you also want pages with no impressions at all.
2. Exclude anything published in the last three months. It has not had a fair run, and judging it early is how good pages get deleted.
3. Add clicks, impressions and conversions per page. Conversions change the verdict more than traffic does.
4. Add backlinks per page. A page with links is never deleted — it gets redirected.
5. Mark each page still accurate, or not.
6. Read the rows marked Update. That list is your next quarter of content work.

The verdict formula

Verdict	When it applies	What to do
Keep	Earning clicks or conversions, and still accurate	Nothing. Set a refresh date and move on.
Update	Impressions but few clicks, or good traffic and out of date	Rewrite for the query it actually ranks for. Usually the highest-return work in the whole audit.
Consolidate	Two or more pages competing for the same intent	Merge into the strongest URL, redirect the others. Keeps the links and stops the cannibalisation.
Redirect	No traffic, no longer relevant, but it has backlinks	One hop to the closest genuinely equivalent page. Never to the homepage.
Delete	No traffic, no links, no longer accurate, nothing equivalent	Remove and let it 404 honestly. A 404 is not a penalty; a misleading redirect is worse.

Two rules are built into the formula deliberately, because both are commonly got wrong:

Accuracy overrides traffic. A page marked “No” for still-accurate returns Update if it has traffic and Redirect if it does not — never Keep. An inaccurate page is a liability regardless of what it earns, and a popular wrong page does more damage than an unpopular one.

A page with backlinks is never deleted. It returns Redirect instead. Deleting a page with links throws away the one thing about it that was genuinely hard to acquire.

The columns

Column	What to put in it
URL	One row per page. Export from Search Console or your sitemap.
Title	What it is, so you can scan the sheet without opening tabs.
Type	Comparison, template, how-to, definition, news, case study. Dropdown.
Published	Original date. A page under three months old has not had a fair run yet.
Last updated	Blank here is itself a finding.
Clicks (12mo)	From Search Console. The primary input to the verdict.
Impressions (12mo)	High impressions with low clicks means it ranks for the wrong thing.
Conversions	If you have them per page. Changes the verdict more than traffic does.
Backlinks	A page with links is never deleted — redirect it instead.
Still accurate?	Yes / No / Unsure. Dropdown. An inaccurate page is a liability, whatever it earns.
Verdict	Calculated: Keep, Update, Consolidate, Redirect or Delete.
Action and owner	What happens and who does it. Without this the audit changes nothing.

The workbook handles 150 rows and totals how many pages fall into each verdict, so the output is a plan rather than a spreadsheet: “14 to update, 6 to consolidate, 3 to redirect, and 41 that are fine.”

If you have thousands of pages

The sheet holds 150 rows, which is deliberate. A site with 3,000 posts does not have a 3,000-row problem, it has a prioritisation problem — and the audit that never finishes is the one that tried to cover everything.

Work it down in this order:

1. Filter to pages with impressions in the last 12 months. Pages Google has never shown to anyone do not need a verdict; they need to be found later in a crawl. This alone usually removes half the list.
2. Exclude anything published in the last three months. It has not had a fair run.
3. Sort by impressions descending and take the top 150. Highest impressions with lowest clicks is where the recoverable traffic is, so this ordering puts the money first.
4. Audit that pass, act on it, then take the next 150. Two passes a quarter clears 1,200 pages a year without anybody burning out on row 400.
5. Handle the long tail as a group, not as rows. Pages with no impressions, no links and no conversions can be judged as a batch — one decision for the batch, applied once.

Consolidation is the underused verdict

Two pages competing for the same intent is extremely common and rarely noticed, because neither page looks broken — they just both underperform. Merging the weaker into the stronger URL and redirecting it in one hop keeps the links, stops the competition, and usually produces a page better than either.

Common mistakes