Most content refresh advice boils down to one instruction: update your old posts. Change the year in the title, add a paragraph, swap a screenshot, republish. It’s satisfying work. It produces visible activity. And for a large share of aging pages, it’s the wrong move.
Some pages should be left alone because they’re fine. Some should be merged into a stronger page covering the same ground. Some should send visitors somewhere more useful. Some should be retired entirely. A refresh program that only rewrites is like a gardener who only waters: busy, well-intentioned and blind to everything that actually needs pruning.
This guide lays out a decision path for every aging URL, so each page gets the move it actually needs.
What are the five moves?
Every page in a content inventory can be assigned one of five statuses:
- Keep: accurate enough. No rewrite this cycle.
- Refresh: the page has a job worth doing and needs updating to keep doing it.
- Merge: the page overlaps with another, and the content should be folded into the stronger URL.
- Redirect: the page no longer serves its purpose, but a living page does, so visitors should go there.
- Retire: the page has no useful successor and should be removed or noindexed according to your process.
Four of those five aren’t rewrites. That’s the point. Redirects, merges and retirements are first-class moves, not admissions of failure. A healthy content library loses dead weight as deliberately as it gains new pages.
What can libraries teach us about pruning?
Librarians have thought about this problem for a long time, and three practices are especially useful.
The CREW method. Developed for public libraries and published by the Texas State Library, CREW (Continuous Review, Evaluation, and Weeding) gives librarians a structured way to decide what stays on the shelves. Its MUSTIE criteria flag items that are Misleading, Ugly, Superseded, Trivial, Irrelevant, or available Elsewhere. Swap “shelf” for “site” and the list works remarkably well for web pages.
The Oxford English Dictionary’s revision. The OED has been undertaking a comprehensive revision of its entries for years, updating definitions, etymologies and citations as evidence improves. That’s a refresh: same entries, same job, better information.
Britannica’s print edition. In 2012, Encyclopaedia Britannica announced that it would stop publishing its printed edition and focus on digital. The content lived on. The format was retired, and readers were, in effect, redirected.
Weeding, revising and redirecting are all normal parts of keeping a collection useful.
How do you decide which move each page needs?
Walk each URL through five questions, in order. Stop at the first “yes.”
1. Is it accurate and still doing its job?
If yes, Keep. Note the date you reviewed it and move on. Not every page needs attention every cycle, and leaving good pages alone is a decision, not a skipped task.
2. Does another page on your site cover the same topic better?
If yes, Merge. Move any unique, still-accurate material into the stronger page, then redirect the weaker URL to it. Two half-useful pages competing for the same question rarely beat one complete one.
3. Is the page’s purpose over, but a related living page exists?
If yes, Redirect. Expired promotions, discontinued products and past events usually fall here. A visitor who arrives at last year’s sale page should land somewhere useful, such as the current collection, rather than a dead end.
4. Is the purpose over, with no sensible successor?
If yes, Retire. Remove it or noindex it according to your process. Not every page needs an afterlife.
5. Does the page still have a job, but the content has drifted?
If yes, Refresh. This is where rewriting belongs: pages whose job matters and whose content no longer does it well.
Running the questions in this order matters. Most teams ask the refresh question first, which is why most refresh programs rewrite pages that should have been merged or redirected.
How do merges and redirects work in practice?
A merge is two jobs done in order. First, editorial: open both pages, list what each one says that the other doesn’t, and move the still-accurate, still-useful material into the stronger URL. Second, technical: set a permanent redirect from the weaker URL to the stronger one, then update your own internal links so they point directly at the survivor rather than bouncing through the redirect.
Redirects need one rule above all: send visitors somewhere that actually answers what they came for. A discontinued product page should go to its replacement or its category, not to the homepage. Redirecting everything to the homepage is the web equivalent of a shop answering every question with “it’s around here somewhere.”
Log both kinds of change with the old and new URLs, so the next person who wonders where a page went can find out in seconds.
How do you prioritize the pages that need refreshing?
Once you know which pages need a refresh, you need an order. A useful formula multiplies three ratings, each from 1 to 3:
- Impact: how much the page matters to readers and the business.
- Decay: how outdated it is, from mostly accurate (1) to materially wrong or risky (3).
- Effort: how much work the refresh takes, inverted so easier work scores higher.
Priority = Impact × Decay × (4 − Effort), giving a range of 1 to 27 when every input is filled in. An important page that’s badly out of date and easy to fix scores high. A minor page that’s mostly fine scores low.
If any of the three ratings is missing, the priority should read Incomplete, not zero. A page nobody has assessed isn’t low priority. It’s unknown, and treating it as zero quietly buries it.
What goes into a refresh brief?
A refresh brief is shorter than a new-content brief, because the page already exists. It should answer:
- What stays: sections, facts and examples that are still accurate and useful.
- What changes: specific outdated claims, screenshots, prices, steps or links.
- What goes: material that’s wrong, redundant or now covered elsewhere.
- What’s new: anything the page needs that it never had, such as a direct answer to a question readers now ask.
Without a brief, refreshes tend to sprawl into rewrites, and rewrites tend to lose the parts that were working.
How do you record what you changed?
Every change gets a row in a change log: the date, URL, change type (refresh, merge, redirect or retire), owner, a short summary, a link to the brief or ticket and a status such as Shipped, Scheduled or Rolled back.
Keep outcomes out of the change log. The log records what you did. A separate results tracker records what you observed afterward, with dated baselines. Mixing the two makes it too easy to credit every later improvement to whatever you happened to change last.
What mistakes make refresh cycles stall?
Six common ones:
- Inventorying everything at once. Bound the scope. A section of the site is a fine first cycle.
- Refreshing without merging or redirecting. The library grows and overlaps.
- Updating dates without updating content. Readers and machines notice.
- Guessing scores to fill blanks. Leave them Incomplete until someone actually assesses the page.
- Skipping the brief. Refreshes turn into rewrites, and good parts get lost.
- No change log. Next cycle, nobody remembers what was done or why.
How long should a refresh cycle be?
Quarterly suits most small and medium sites. Each cycle: inventory a bounded section, assign statuses, prioritize the refresh candidates, brief and ship the top few, and log everything. The next cycle picks up a new section or revisits the highest-priority pages.
The goal isn’t a perfect site. It’s a library that gets a little more honest every quarter.
How does the Website Content-Refresh Kit help?
The Website Content-Refresh Kit turns this decision path into a working cycle. Its seven-tab workbook includes a Page Inventory with status tracking, a Prioritization sheet with formulas that show Incomplete instead of coercing blanks to zero, a Refresh Brief template, a Change Log, a Results Tracker that waits until both before-and-after values exist before showing a change, and a fully worked fictional example for an outdoor retailer called Harbor Thread. A 12-page guide explains the statuses, the prioritization math, the brief method and how the kit fits alongside an AEO audit or a new-content brief.
It’s a practical cycle for updating what you’ve already published, built for the full set of moves, not just the rewrite.
Quick answers
Should old blog posts ever be deleted?
Some should. If a post has no useful purpose and no sensible successor, retiring it is a legitimate choice. If a related living page exists, redirecting is usually better than deleting.
Is updating the publish date enough?
No. Change the date only when the content meaningfully changed, and log what changed.
How many pages should one cycle cover?
Enough to finish. A bounded section you complete beats a full-site inventory you abandon.
Give every page the move it needs
Rewriting is the most visible refresh move and often the least needed. A decision path that considers keep, merge, redirect and retire first saves time, shrinks overlap and puts your rewriting effort where it actually helps.
Get the Website Content-Refresh Kit and run your first bounded cycle this quarter. Inventory one section, walk every URL through the five questions, and let the library get smaller where it should and better where it matters.
The Website Content-Refresh Kit is on Content & SEO Workflows, with the other site-maintenance tools in this line.
Leave a Reply