Content Refresh vs New Post: When to Rewrite an Old Article

Most blogs I crawl look the same: fifteen to forty articles, three of them bringing in 80% of the organic clicks, the rest slowly sinking. The obvious reaction is to write something new. The cheaper and faster move is usually a content refresh. An existing post already has a URL, a few links, index history, and real data about which queries it shows up for. A new article has none of that and spends months earning it.

Why a refresh moves faster than a new post

A new article starts from zero. Google has to find it, classify it, test it at some position, and gradually decide where it belongs. That takes weeks to months, and longer on weaker domains.

A refreshed article starts ahead. It is already indexed, already appearing for some queries, and its relevance is already established. When you add what it was missing, you are not going from nothing to position 15 — you are going from 15 to somewhere higher. Shorter jump.

The second reason is practical. An article sitting at position 12 with 2,000 impressions a month has proven demand. With a new article you estimate demand. With a refresh you read it off the chart.

The third is time. Rewriting a text that already has a skeleton and a structure is faster than inventing one. I am not saying stop writing new posts. I am saying that when you have 30 articles and half of them are rotting, the order of work is different from what you assume.

How to pick the articles worth refreshing

Pick from data, not from which article you personally like least. Open Google Search Console, Performance report, last three months, and switch the view to Pages.

I look for four patterns:

1. High impressions, low clicks. The page gets shown, nobody clicks. Typically CTR under 1% on hundreds of impressions. The problem here is usually not the content but the title and meta description. Sometimes rewriting those two lines is the whole job.

2. Average position 8 to 20. This is the most profitable group. The article is near the first page or at the bottom of it. Moving five positions here means several times more clicks than moving from 50 to 45.

3. Declining trend. Compare the last 90 days with the same period a year ago. Articles that lost more than a third of their clicks have usually gone stale — prices changed, procedures changed, a tool's interface changed, the rules changed.

4. Articles ranking for queries they never answer. When Search Console shows a post collecting impressions for a question it does not cover, you have a brief for a new section written for you.

Articles with zero impressions and zero clicks over a long period: leave them alone. There is usually nothing to revive. Either the topic has no demand, or the page overlaps with another one on your site.

What to actually change in the article

A refresh is not changing the publish date. If you change the date and leave the content, nothing changes except the date.

Title. Check that it contains the query the article actually ranks for. Keep it around 60 characters so search engines do not truncate it. Also check you do not have the same title elsewhere on the site — duplicate title tags can push clicks to the wrong page.

Meta description. On articles with high impressions and low CTR, this is the first thing I change. The description should be specific and answer the question in the query, not describe how nice the website is.

Intro. Rewrite the first two paragraphs so they answer the question immediately. Long warm-ups are the most common reason people bounce back to the results page.

Missing sections. Add what the pages above you cover and you do not. Do not copy them — look at which sub-questions they answer, then answer those your own way.

Outdated facts. Years in the text, old screenshots, discontinued features, dead links. This is the most common reason an article drops.

Internal links. An old article links to nothing published after it. Add links to newer texts, and link back to it from those newer ones. Internal linking is the cheapest part of a refresh with the biggest effect.

Alt text and images. If images were added without descriptions, write them. During a refresh it costs a minute. If you are unsure what belongs in there, I cover it in the piece on image alt text.

When to merge articles instead

Common situation: three articles on almost the same topic, each 200 words, each collecting a fraction of the impressions. They are not competing with your competitors, they are competing with each other. That is keyword cannibalization.

The fix is merging them into one solid text:

1. Pick the URL with the most clicks and the longest history. That one stays.
2. Move the useful parts of the other articles into it.
3. Set 301 redirects from the retired URLs to the surviving page.
4. Fix the internal links that pointed at the retired URLs, so you do not create redirect chains.

What not to do: do not delete old URLs without a redirect, and do not change the slug of the surviving page unless there is a reason. Changing a slug is a separate operation with its own risks — how to do it safely is in the article on URL structure and slugs.

How to tell whether the refresh worked

Without measurement a refresh is just a feeling. The process I use is simple:

• Write down the date the change went live.
• Compare 28 days before and 28 days after for that specific URL: clicks, impressions, average CTR, average position.
• Watch seasonality. If your whole industry drops in August, one page dropping is not necessarily about the refresh.

First movements usually show up once the index refreshes — days to weeks, depending on how often the crawler visits the page. Bigger position changes get judged later. Anyone promising a jump within three days is not measuring, they are guessing.

You can tell two things apart by what moved. If CTR grew on roughly the same impressions, the title and description did the work. If impressions grew, the added content did — the article started appearing for more queries. I go deeper on that distinction in the piece on measuring SEO impact.

How many refreshes, how often

For a small business a simple rhythm works: once a quarter, go through Search Console, pick three to five articles to refresh, and do them. Alongside that, write new texts on topics the site is missing.

The ratio between new content and refreshes shifts with the age of the blog. With five articles there is nothing to refresh — write new ones. With thirty, half of them declining, give refreshes at least a third of your time. How many new texts are worth adding is covered in the article on how often to publish SEO articles.

What I do

During a regular check of a site I pull Search Console data and look for exactly those four patterns above — pages with high impressions and low CTR, pages at positions 8 to 20, declining articles, and queries the existing text does not cover. For each finding I do not write advice like "rewrite the title". I write the finished title, the meta description, and the text of the missing sections, ready to paste. You deploy them through the WordPress plugin, the API, or a CSV export, and all you do is approve them. After deployment I measure 28 days before and after, so it is visible whether the change did anything.

If you are still working out what to write next, start with topics from Google Search Console. Refreshing what already exists is usually the step that pays back sooner.

#seoobsah