How to Update a Page People Have Already Cited
Once somebody has cited your page, a piece of their work depends on it saying roughly what it said. You still have to keep it accurate — an out-of-date page is worse than a corrected one. The trick is updating in a way that keeps existing citations honest instead of silently making them wrong.
That’s a lower bar than it sounds. But the ways of failing are specific, and one is common enough to have become a tactic.
The obligation you took on
When a writer linked you, they wrote a sentence around the link — something like deposits are held for ten business days, followed by your URL. That sentence is now a claim about your page as much as about the world.
Change the page and one of three things happens. Their sentence still checks out, so nothing is wrong. Or the page no longer contains what they said it did, so a reader who follows the link concludes the writer made it up. Or, worst, the page now says the opposite, and the citation contradicts itself.
You never agreed to freeze the page. You did implicitly agree not to embarrass the people who vouched for you, and which outcome you get is almost entirely about how you handle the edit.
Two kinds of edit
Nearly every change to a cited page is one of two things, and conflating them is how pages end up misrepresenting their own citers.
A refresh keeps the claim and improves everything around it: better evidence, current details, clearer writing. The thing you were cited for still holds.
A reversal changes the claim itself. You were wrong, or the situation changed, or you thought harder and now believe something else. The thing you were cited for no longer holds.
Refreshes are routine and need light handling. Reversals have to be announced. Treating a reversal as a refresh — rewriting the conclusion in place and moving on — costs you future citations.
Handling a refresh
Keep everything a citer might have pointed at. Same address, same headings — or at least the same wording for the claim inside them — and a summary at the top that still says the same thing, better.
Update the last-changed date so a reader can see the page is maintained. It’s the whole reason to carry two dates, as making pages easy to cite covers.
Add one line saying what changed. “Updated July 2026: added the current filing thresholds.” Ten seconds, and it converts an ambiguous edit into a documented one. A returning reader who remembers the page differently can tell whether they misremembered or you revised.
That’s the whole procedure. Refreshes are cheap; do them often.
Handling a reversal
Say that the previous version said otherwise, and why you changed your mind. Not buried in a footnote — where somebody landing on the page will see it.
The instinct is to avoid this, because a visible correction feels like admitting you were unreliable. It’s exactly backwards. A correction note is what makes you safe to cite in future: it proves the page is watched by someone who will say when it moves, which is what a careful writer is looking for.
Consider the alternative from the citer’s side. They wrote a sentence, backed it with your link, published. Months later a reader follows that link to a page flatly contradicting the sentence, with no indication anything changed, and concludes the writer was careless. The writer, if they notice, concludes you can’t be relied on — permanently. Nobody cites a source twice after being made to look wrong by it. Two sentences of correction buy you the benefit of the doubt for as long as the page exists.
Keep a change log on the pages that matter
Date plus one line, newest first, at the bottom of the page. No formatting project, no separate system.
It’s absurdly cheap and does more for credibility than a redesign. Technical writers and reference editors, the most conservative linkers there are, read change logs to judge whether a page is maintained. Four entries over three years says “somebody owns this.” No log and no dates says nothing, and they fill the blank in pessimistically. Restrict it to load-bearing pages; a log on every post is busywork.
Never repurpose a cited URL
Here is the tactic, stated plainly so you recognise it when it’s recommended: take a page that has accumulated links, replace its contents with a different subject, keep the links.
Don’t. The links say the page is about one thing; pointing them at another is deliberate misuse of a signal you didn’t earn for the new subject, whatever traffic argument comes attached. Google’s spam policies are not ambiguous about the general shape of this.
Set search engines aside and it’s still wrong, because it makes every existing citation false. Everyone who linked the page now sends readers to something that doesn’t support their sentence. You’ve retroactively falsified other people’s work to recycle an asset, and the damage lands on the people most likely to cite you again. If you want a page about the new subject, write one.
Retiring a page honestly
Sometimes a subject genuinely stops existing: a rule repealed, a format discontinued, a product you documented that no longer ships.
Mark it obsolete and keep it. A note at the top saying what changed and when, plus a pointer to whatever replaced it. The page becomes a historical record, often more useful than the original: people arriving from old citations get an accurate answer about the past instead of a dead end.
Deleting is the one clearly bad option, because it destroys every reference pointing at it. (If you must move a page, at minimum leave a redirect at the old address; the configuration belongs elsewhere.) Wiping it makes every citation unverifiable — a small act of vandalism against other people’s writing.
Review by load, not on a schedule
Uniform review cycles waste effort on pages nobody depends on and under-serve the ones everyone does. Sort by load instead: how many people point at the page, how time-sensitive its claims are, and what it would cost somebody to be wrong because of it.
A page stating a rule that changes yearly needs checking yearly. A page explaining a mechanism that hasn’t changed since the web began needs checking when you notice. And a page with real citations behind it comes first, because other people’s reputations are attached to it.
What to do next
Before you rewrite anything substantial, find out what points at it. Pull a backlink export, see which pages carry citations and roughly what they’re cited for, and treat that as the constraint on your edit — it’s part of what a backlink audit is for. Then decide honestly whether the change is a refresh or a reversal, and label it. The citable part is usually one sentence: writing claims people can quote covers keeping it intact.