A blog that's been running for three or four years usually has the same shape. A handful of posts still bring in most of the search traffic. A middle band gets a trickle. Then there's the long tail: announcements, thin how-tos, two posts on the same topic written eighteen months apart by different people, and a "trends for next year" piece nobody has opened since that year ended.
The instinct is to do something big about it. Delete half the blog, or rewrite all of it. We'd push back on both. Most old posts need a small decision, made one URL at a time, with the data in front of you.
Here's the method we use. It fits in a spreadsheet.
First, get the data
You need three things for every indexable URL older than about 12 months:
- Impressions and clicks from Google Search Console, over the longest window you have. Search Console keeps 16 months. Don't judge a page on the last 28 days.
- The queries each page shows up for. This is how you spot two pages doing the same job.
- A note on what the page is for: which service, product or customer question it supports, if any.
Not sure which URLs you even have? Start with your sitemap. Our free sitemap decay scanner reads it and sorts every URL by its last-modified date, so you can see how much of the site hasn't been touched in a year or two. Treat that as a first look, not a verdict. An old date doesn't mean a page is failing.
The four questions
Ask them in this order. The order matters, because the first "no" often settles it.
- Does it get impressions? Is Google showing it to anyone, even if few people click?
- Does it overlap another page? Do two or more of your URLs show up for the same searches and answer the same question?
- Is it still true? Prices, screenshots, rules, product names, steps, links. Would you be comfortable if a customer read it this afternoon?
- Does it support something you sell? Does it link naturally to a service or product page, or answer a question your buyers actually ask?
The decision table
| Impressions? | Overlaps another page? | Still true? | Supports a service? | Verdict |
|---|---|---|---|---|
| Yes | No | Yes | Yes | Keep. Check the links, then leave it alone |
| Yes | No | No | Yes | Refresh. Fix what's wrong, add what's missing |
| Either | Yes | Either | Either | Merge into the stronger page, redirect the other |
| Yes | No | Yes | No | Keep. It can still earn readers and links |
| No | No | Yes | Yes | Refresh and link to it. It may just be buried |
| No | No | No | No | Retire. Redirect if there's a close match, otherwise let it go |
Two notes on reading it.
The overlap row beats everything else. If two posts compete for the same searches, refreshing both just gives you two better posts that still compete. Merge first, then refresh the one that survives.
And "no impressions" on its own isn't a death sentence. An accurate page that supports a service might have zero impressions because nothing on your site links to it. Try internal links before you try the delete key.
A hypothetical worksheet
Here's the table applied to five made-up URLs from a made-up plumbing company's blog. None of this is client data.
| URL | What we'd find | Verdict |
|---|---|---|
| /blog/tankless-water-heater-cost/ | Steady impressions, prices from three years ago | Refresh: update the cost ranges, recheck the rebate links |
| /blog/tankless-vs-tank/ and /blog/is-tankless-worth-it/ | Both appear for the same comparison searches | Merge into one guide, 301 the weaker URL |
| /blog/we-sponsored-the-5k/ | No impressions, no service tie, still accurate | Keep. It isn't hurting anything. Just don't spend refresh money on it |
| /blog/winter-pipe-tips-2021/ | Impressions every November, tips still valid | Refresh, drop the year from the title, update the date honestly |
| /blog/new-website-launch/ | No impressions, out of date, nothing close to redirect to | Retire with a 410 |
Look at the sponsorship post. Not every old page needs a verdict that costs money. "Keep, do nothing" is a legitimate answer, and on most blogs it's the most common one.
Redirect basics
This is where cleanups go wrong, so slow down here.
When you merge
- Pick the surviving URL. Usually it's the one with more impressions, more links from other sites or the cleaner address.
- Move anything useful from the other page into it. A merge that only deletes isn't a merge.
- Set a 301 (permanent) redirect from the old URL to the survivor.
- Update your own internal links so they point straight at the survivor, not through the redirect.
- Avoid chains. If A already redirects to B and you now merge B into C, point A straight at C.
When you retire
- If there's a genuinely close page, redirect to it. If there isn't, don't send everything to the homepage. Someone who wanted a specific answer and lands on your homepage hasn't been helped.
- With no close match, a 404 or 410 status is fine. Both tell search engines the page is gone.
- Take the URL out of your sitemap and out of your internal links.
Google documents redirects and status codes in detail on Search Central, and those pages get revised. Read the current versions before you run a large batch.
What we wouldn't do
- Delete in bulk to make the site look "fresh." Google's helpful-content guidance asks whether you're removing lots of older content mainly because you think it'll make the site seem fresh, and answers its own question: it won't. Remove pages because they're wrong, redundant or useless to readers. Not to hit a page count.
- Rewrite everything at once. You lose any way of telling what worked.
- Change the date without changing the page. That one gets its own article: changing the date isn't a refresh.
- Skip the record. Every refresh, merge and redirect goes in a change log with the date and the old URL. Six months later, someone will ask.
How long this takes
For a 60-post blog, one person with Search Console access can work through the table in a day or two. At 400 URLs, pulling and sorting the data becomes the job.
That's what our content decay audit is for. It's $1,450 for up to 300 indexable URLs, plus $350 for each further 300, up to 1,500. You get every URL with 16 months of Search Console data, a keep / refresh / merge / redirect / retire verdict for each one, the overlapping pairs, the top 20 refresh candidates ranked by lost clicks, a 90-day maintenance plan and a 45-minute walkthrough. It's delivered 10 business days after we get Search Console access.
Already have the list and just need the work done? A refresh pack covers five pages for $1,350. Ongoing upkeep is what our content refresh service is built around.
None of this guarantees a traffic number. Search engines change how they rank pages, and results depend on your market, your competitors and your own site. What a cleanup does is make sure the pages you keep are worth finding.
