The SaaS Content Refresh Playbook: When To Update, Rewrite, Or Kill A Post

Most old blog posts should be updated, not republished and not deleted. Update a post that still ranks or holds backlinks, rewrite one whose topic fits but whose intent has drifted, and kill one with no demand, no links and no fit with your product. Check the topic, then the data, then the intent.

This rule applies to a SaaS blog with enough history to read a year of Search Console and analytics data per URL.

The numeric thresholds below are rules of thumb from published guides, so adjust them to your own traffic base, and send genuine judgement cases to a person rather than a formula.

Republishing, Refreshing, Rewriting, And Killing: What Each Term Means

Teams use these four words as if they meant the same thing. They don't, and mixing them up is how a quick edit gets billed as a rewrite and a deletion gets called a cleanup. Here is the vocabulary the rest of this guide uses.

• Republishing: Putting an existing post back in front of readers, usually with a new date. On its own it changes the date and nothing else, so it only counts when the body changes too.
• Refreshing: Editing the post in place at the same URL. Facts, examples, screenshots, links and the opening get updated, the URL stays, and the original date history is respected.
• Rewriting: Replacing most of the body and often the angle or format, while keeping the URL because the address still has links or rankings worth keeping.
• Killing: Taking the post out of play. That means a 301 redirect to the best surviving page, a merge into a stronger page that shares its intent, or removal where there is nothing to preserve.

Syndicating a post to another site and cross-posting it to a social platform are a different job with licensing and attribution questions of their own. This guide does not cover them. It covers what to do with the old posts on your own blog.

The Three-Way Rule: Update, Rewrite, Or Kill

Every old post gets one of three outcomes. The skill is in the order you check things, because the order decides which signals can overrule which. Check the topic first, a year of data second, and rankings and intent third.

The table shows how the signals line up. Read it from the top, because a failure on the first row beats a pass on every row below it.

The three-way rule runs in a fixed order. First, check whether the post's topic still fits the product and the buyer. Second, check a year of performance data for the URL. Third, check the post's rankings and whether search intent has drifted away from the page. A post that fails the topic check is a kill or consolidate candidate whatever its traffic. A post that passes all three checks is an update. A post that passes the first two and fails the intent check is a rewrite.

Underneath the table sits one default. When in doubt, update before you delete. Default to small updates, because deleting a post can erase the SEO gains it earned. That asymmetry is the whole reason to start with the gentlest outcome. A bad update costs you an afternoon and can be reversed. A bad deletion costs you a URL, and getting it back means rebuilding whatever it earned.

Removal is not all or nothing either. Content Marketing Institute's guidance is to 301 redirect an old post that draws little organic traffic or is no longer accurate, and to keep the original live when it still gets substantial organic traffic and is current. So even the kill outcome splits in two. Some posts redirect, some merge, and a few stay put.

Step One: Pull A Year Of Data Before You Decide

Open with the window. Pull one full year of data per URL, because a single month hides seasonality and a single quarter hides a slow slide. Content Marketing Institute's candidate selection runs on about a year of data covering traffic and conversions, and that is a sensible floor for a SaaS blog too.

Export these five fields for every post, one row per URL.

1. Organic clicks: From Search Console, for the last 12 months, plus a second column for the last 28 days.
2. Impressions: From Search Console, over the same two windows. Impressions tell you whether a post is still being shown even when clicks have dried up.
3. Average position: From Search Console, per URL. Sort by it, because the posts sitting between position 8 and 20 are your cheapest wins.
4. Referring domains: From your backlink tool. A post with links is a post with authority you can lose.
5. Last updated date: From your CMS. Age is not a verdict, but it tells you where to look first.

Then add the comparison window. lots.blog's prioritisation guide compares the last 28 days with the prior 28 days, which is short enough to catch a decline while it is still recent and long enough to smooth out a single bad week.

Pair Search Console with Semrush to see each old URL's organic position and traffic history alongside its referring domains in one export, which makes the keep, rewrite or kill call faster than checking each page by hand.

Put everything in one sheet and add two blank columns, one for topic fit and one for the outcome. You will fill them in over the next two steps. If you want help choosing software to watch for decay on an ongoing basis, our list of tools that catch content decay early is a good starting point.

One warning. Resist the urge to sort by traffic and start at the bottom. Low traffic is not the same as low value, and a post with three clicks a month and twelve referring domains is not a kill. The sheet is there so you can see clicks, links and age side by side before you decide anything.

Step Two: Check The Topic Still Fits Your Product

Do the fit test before you look at traffic. A post on a topic you no longer sell into can still rank, and ranking for the wrong audience is worse than not ranking, because it fills your reports with visits that never turn into pipeline.

Four things should put a post on the retire list. The topic no longer fits the company, the product or customer focus has pivoted, two or more pages share the same intent, or the post needs a full rewrite rather than patching. One way to apply them is to let each trigger point to an outcome. A post that shares intent with a stronger page is a consolidate, which means a merge and a 301 to the survivor. A post that needs a full rewrite goes to the rewrite bucket, but only if the topic itself passes the fit test. A post whose topic no longer fits the company, or that was written for a customer you have moved away from, is a kill candidate.

The test itself takes a minute per post. Open the post beside your current positioning page and ask whether a buyer you want to attract would still land on it and find it useful. If the honest answer is no, write "no fit" in the sheet and move on. Do not let a healthy traffic number talk you out of it.

Notice what this step does to the workload. On some blogs part of the archive fails here, and for those posts you never need to analyse rankings at all. Fit first, then data. It saves hours.

Step Three: Read Rankings, Clicks, And Intent Drift

Posts that pass the fit test move on to the numbers. These are the three triggers, in the order to apply them. They are rules of thumb. The source says so itself, and you should adjust every number to your own data.

1. Striking distance: lots.blog's refresh thresholds put first in line the pages with an average position of 8 to 20 and impressions at or above the site median.
2. Click decline: The same guide puts next the pages whose clicks fell 30% or more against the prior period, from a base of at least 5 clicks, after you have checked seasonality.
3. Snippet rewrite: Where average position is 5 or better, impressions are above the median and click-through rate is under 2%, source suggests rewriting only the title and meta description.

Intent drift is the check that no number gives you, so do it by eye. Search the post's main keyword today and compare the top results with your page. If they are all comparison tables and yours is an essay, or all step-by-step guides and yours is a definition, the page has drifted from what searchers want, and it is a rewrite candidate whatever its position.

A few notes on using the triggers sensibly. The site median matters more than it looks. On a small blog the median post may have a few dozen impressions a month, so "at or above the median" is a low bar, and you will want to cap the list by hand. The seasonality check matters too. A 30% click drop on a post about year-end planning in February is the calendar, not decay. Do not refresh it because of that.

If you want a sense of whether your own click-through and ranking figures are healthy, compare them against our B2B SaaS content benchmarks before you decide the numbers are bad.

When a post trips none of the three triggers and the intent still matches, leave it alone. Not every old post needs work. The goal is a short, ranked list of posts where an hour of effort is most likely to move a number.

When To Update A Post In Place

Update in place when the post still ranks or holds links. The evidence for the approach is good. Historical optimisation of older HubSpot blog posts raised monthly organic views by 106% on average, and 76% of HubSpot blog views come from posts published before the current month. Old posts are where a mature blog's traffic lives, so protecting and improving them is the main job, not a side task.

HubSpot is a very large publisher, and your numbers will not match its numbers. What carries over is the logic. If most of your traffic comes from posts older than a month, most of your upside comes from them too.

What to change in an update is a short list.

• The opening: Rewrite it so the answer comes first, in the first two sentences.
• Facts and figures: Replace anything that has aged, and link the new source.
• Examples and screenshots: Swap out anything that shows an old interface or a retired feature.
• Internal links: Add links to newer posts, and fix any that now point at redirects.
• Gaps: Add the sections the top results now cover and your post does not.

What to leave alone matters more. Keep the URL unchanged so the links and rankings the page has accumulated stay with it, and update the existing post rather than publishing a duplicate on the same topic. Keep the headings that already rank unless you have a reason to change them, because every heading you rename is a small bet that the new wording is better. Keep the parts that were already good.

Before you start, confirm that the post is underperforming and has link authority. Updating a post that has neither is rewriting by another name, and that is a different outcome with a different cost.

If you want refreshed posts to be picked up by AI search as well as ranked, our generative engine optimization work covers that side of the update.

When To Rewrite A Post From Scratch

Rewrite when the topic fits, the URL has value, and the angle, format or intent no longer match what searchers want. All three conditions have to hold. If the topic doesn't fit, you are in kill territory. If the URL has no value, a new post at a new address does the same job with less baggage. If the intent still matches, you only need an update.

The test for the third condition is the intent-drift check from step three. Study the top-ranked pages and match your post's format and angle to what searchers currently want. A post that explains a concept at length, up against results that are all templates and tools, will not recover with a fresh paragraph at the top. It needs to become a different kind of page.

A rewrite means the old body is gone. You keep the URL, the best of the research, and any section that still earns its place, and you rebuild everything else around the new angle. Treat it as a new article with an inherited address.

Our Buddy Punch work is the clearest first-hand example. Buddy Punch had a blog of 800+ posts and had plateaued at about 28,000 monthly organic visits. We short-listed 85 posts for full rewrites, and the account reported a 240% organic traffic lift. The full story is in our Buddy Punch rewrite case study. The point for this guide is the ratio. Out of more than 800 posts, 85 were chosen for a full rewrite. Rewriting is for a minority of the archive, the posts where the topic is right and the page is wrong, and not for everything that looks old.

If you do decide to rewrite, the right tooling can speed the work up. See our roundup of AI tools for rewriting old posts for what is worth using.

When To Kill, Consolidate, Or Redirect A Post

Delete a post only when it is irrelevant or outdated beyond repair and has no traffic, backlinks or value. If any of those three is still true, one of the gentler outcomes fits better. When the post does meet that bar, work through the removal in this order.

1. Confirm the post fails the fit test and carries no traffic, no backlinks and no value. If it has any of those, stop and go back to update or rewrite.
2. Check referring domains in your backlink tool. Even one good link is a reason to redirect rather than delete.
3. Choose the destination, which is the best surviving page on the closest topic. A relevant page is better than the homepage.
4. Merge anything worth keeping into that page first, if two pages share intent. Fold in the missing section, example or data point, then retire the weaker page.
5. Redirect the old URL to the destination with a 301. Content Marketing Institute advises a 301 redirect for an old post that draws little organic traffic or is no longer accurate.
6. Return a 410 only where there is nothing to preserve and no related page to send people to.7. Fix internal links so they point straight at the destination and not through the redirect, and drop the old URL from your sitemap.
7. Log the change with its date, so you can read the effect later at 30, 60 and 90 days.

A redirect sends visitors and crawlers from the old address to the new one, which means the links that pointed at the old post still land somewhere useful. Deleting with no redirect sends all of them to an error page. That is the main reason to prefer a redirect over a bare deletion whenever a relevant destination exists.

Deletion is the wrong call in a few specific cases. Do not delete a post that still earns traffic, even if it embarrasses you. Do not delete one with valuable backlinks. Do not delete one whose topic still fits your buyer and whose weakness is fixable, which is just a post waiting for an update. Deleting can erase SEO gains, which is the reason to keep this list close.

Should You Change The Date Or The URL When You Republish?

Keep the URL by default and update the post in place. Move to a new URL only when the original still earns substantial traffic and stays accurate.

Those two answers come from two published positions that look like they conflict. Updating in place is the simplest route, because the URL keeps the links and rankings it has built up. Content Marketing Institute describes a second route, which is keeping the original live with an editor's note while publishing an updated version at a new URL. It applies when the original still gets substantial organic traffic and is current.

The condition that picks between them is simple. If the old version is still right and still earning, leave it alone and publish the new angle beside it. If the old version is wrong, thin or sliding, update it where it stands. Most decayed posts fall into the second group, which is why in-place updating is the default.

On the date, the rule is short. Change it only when the content has changed in a way a reader would notice. Bumping the date over unchanged copy is a cosmetic edit, and it makes your freshness claim false. Readers see a recent date, open the post, and find the same stale advice. That costs trust and gains nothing.

How We Audit A Blog Before Touching A Post

Our audit method scores each URL from 0 to 5 on five categories, which are traffic, backlinks, freshness, topical fit and content quality. The total is out of 25. A score of 20 to 25 means keep or refresh. A score of 10 to 19 means consolidate. A score below 10 means prune.

We back the score with four working thresholds that flag a post for action.

• Traffic: Under 10 clicks in 6 months.
• Freshness: Over 24 months since the last update.
• Visibility: Under 100 impressions on an indexed page.
• Engagement: Under 30 seconds time on page in GA4.

These are our own working thresholds, not validated industry figures, so adjust them to your own blog before you trust them. A blog with very low overall traffic will flag almost everything under 10 clicks, and the flag is only useful if it separates the posts that need attention from the ones that don't.

The scoring is what stops the audit from becoming an argument. Without it, every review turns into a debate about whether a particular post "feels" dead. With it, you have a number and a band, and the debate shrinks to the handful of posts that land near a boundary. Those are the judgement cases, and they go to a person. Fit sits in the score on purpose. A post can score full marks on traffic and still land in the consolidate band because its topical fit scores zero. That is the fit-first rule from step two, expressed as arithmetic.

If you want the whole process written out as a checklist, see our content audit checklist for B2B SaaS. For running the audit across a very large archive, our guide to AI content auditing software covers the options.

How To Measure A Refresh At 30, 60, And 90 Days

We measure changes at 30, 60 and 90 days after they go live. Wait for the 30-day mark before judging a change, and check again at 60 and 90 so you can see whether it is still working. Here is the routine we suggest, using the same prior-period comparison you used to find the post in the first place.

1. Day 30: Compare impressions and average position against the prior 28 days for the refreshed URL. Confirm the page was recrawled and the change is live. A rise in impressions or a better position is an early result. No movement means wait, not panic. This checkpoint decides whether anything is broken.
2. Day 60: Compare clicks and click-through rate against the prior period. If position has improved but clicks haven't, the title and meta description are the next edit. If nothing has moved, check the intent match again. This checkpoint decides whether to adjust the snippet or the angle.
3. Day 90: Compare clicks, referring domains and conversions against the year of data you started with. A post that has moved is a result and moves to the monitored list. A post that hasn't moved is a miss, and it goes back through the three-way rule, because the likeliest reason is that it needed a rewrite and got an update. This checkpoint decides the final outcome.

Refresh in batches small enough to measure. If you change fifty posts in one week, you can't tell which change did what, and a miss teaches you nothing. Start with a batch you can finish, log each change and its date, and let the first results set the size of the next one.

For a realistic view of how long returns take on a blog, our piece on SaaS blog payback time sets expectations.

Keeping Refreshed Posts Visible In AI Answers

Start every refreshed post with the answer. The first two sentences should state the conclusion in plain words, with the subject named, so an AI system or a reader can lift them without needing anything around them. That is the answer-first rule, and refreshing is the cheapest moment to apply it, because you are already editing the opening.

A passage that stands alone has a few features. It names the entities it talks about instead of saying "it" or "this tool". It carries its qualifier in the same sentence as the claim. It puts units on every number and says what period a figure covers. Where a claim depends on the year, it says which year, so a reader or a model can tell whether it is still current.

Refreshing also means removing what has aged. A statement about a feature that no longer exists, or a figure from several years back with no date, makes the whole page look unreliable, and a model choosing between sources has no reason to pick it. Replace it or cut it.

We make no promises about how much any of this changes citations, and you should be wary of anyone who does. It is simply good editing that happens to suit how AI systems pick passages. For the deeper framework, see our guide to keeping evergreen posts in AI Overviews. If you would rather have the work done for you, our answer engine optimization service covers it.