Content Update vs Rewrite: Life-Extension Rules for Old Articles After HCU

Whether content should be updated or rewritten — I struggled with this question for a long time, until a traffic crash made it completely clear. That old article used to steadily bring 200-300 visits a day, then suddenly dropped to double digits. At first I thought it was an algorithm issue, but after digging around I found the intent had changed — what users searching this keyword wanted was no longer the answer written in the article. Updating numbers couldn’t save it, it had to be rewritten.

First, distinguish updating from rewriting

Many people mix updating and rewriting together, but they’re actually very different. Updating is changing numbers, adding cases, adding sections — the article’s skeleton and viewpoint stay put. Rewriting is changing the angle, restructuring, re-arguing — equivalent to rebuilding from the bones.

There’s only one judgment standard: does this article’s core argument still hold? If yes, update; if no, rewrite. The most typical failure case I’ve seen: an article’s viewpoint was already outdated, but the team only changed the title and numbers, and traffic kept dropping — because readers could immediately see the content didn’t match the up-to-date viewpoint.

A decision table

Over the years I’ve总结出 a judgment process, roughly like this: stable rankings, stable CTR, only outdated data — updating is enough, done in half a day. Rankings dropping but traffic structure normal — update plus add content, fill in missing new sections. Traffic halved and suspect intent was misjudged — must rewrite. Rankings haven’t moved up in six months — most likely content depth insufficient, also needs rewrite. And one special case — cut off by AI overviews — no need to rewrite, just restructure for citations.

Before each judgment, I first search the keyword and see what structures are currently ranking at the top. The results page is the current standard answer — following it to judge is more accurate than guessing.

How to update effectively

Updating looks simple, but there’s nuance in doing it. My checklist goes like this: first verify all data and dates, replace outdated numbers with latest sources; then add a real case from the current year to boost credibility; next add a conclusion paragraph at the opening — this gives readers a quick-read entry and also makes it easier to excerpt; then check internal links, link in newly published pillar pages and strong articles; finally update the publish date so search engines know the content changed.

The most easily overlooked part is adding a conclusion at the opening. Many old articles open with long-winded buildup that hasn’t been changed in forever — readers take one look and leave. Change it to conclusion first then展开, and the reading experience is completely different.

How to rewrite without wasting effort

Rewriting is heavier than updating, but done right it’s worth it. My process: step one, go back to search intent — re-search the keyword, look at the mainstream structure on results pages; step two, pick a more specific angle — switch from “how to do” to “three things beginners most easily get wrong” type; step three, tear down the structure and rebuild — reorganize with tables, steps, cases, don’t patch over the original; step four, keep the URL and rewrite directly, don’t create a new page; finally, observe for sixty days — if no improvement, then consider downgrading or merging.

The biggest waste in rewriting is “reluctance to let go” — reluctant to delete the old draft, and as a result the new draft gets dragged down by the old structure, changed and changed but still the same old thing.

One-page decision table

The cost math of updating vs rewriting

Many people get stuck on “should I spend the time.” Updating can be done in half a day, rewriting requires going through the full new-content process and often takes a day or two — so prioritize judging whether updating alone is enough. But don’t keep patching just because you’re afraid of the effort — an old article with a dead argument, you can update it ten times and it won’t come back. The earlier you rewrite, the more you save. I set a rule for myself: if the same article shows no improvement after two consecutive updates, the third time directly rewrite, don’t耗 with it. Write this rule into your handling process, and decisions no longer纠结 — time gets spent where it matters, and you won’t sink into the “changed and changed but still the same” quagmire.

Observation period after rewriting

Rewriting isn’t checking a box — you have to give it time to prove itself. I习惯 leaving a sixty-day observation window after rewriting, watching whether rankings return to where they should be and whether traffic stabilizes. Don’t pronounce it dead in the first two weeks — search engines need time to re-understand the new structure; if still no improvement after sixty days, then consider downgrading, merging, or completely redoing. Writing the observation period into your process blocks two impulses: one, tearing it down again right after changing because no immediate effect; two, giving up because it’s slow to show results. The judgment standard is always one thing — is the core argument still right? If yes, give it a chance; if no, switch routes.

Frequency recommendations

I divide update frequency by content type: tutorial-type — review data quarterly; trend-type — check ranking changes every six months; evergreen-type — add a case once a year; time-sensitive-type — triggered by events. No need to hold out for big moves — make updating a routine action, spend half a day each month refreshing five old articles. Accumulate for half a year, and the whole site’s freshness is visible to the naked eye.

Traffic drops aren’t scary — what’s scary is not being able to tell whether to update or rewrite. First judge whether the viewpoint still holds, then decide the action — much better than blindly diving in.