Understanding average search position drift and when to update an old post requires monitoring your site analytics. Average search position drift is the gradual decline of a page in Google search results as competitors publish newer material. You should update an old post when its average position drops out of the top three spots and into positions 8 through 20. Catching this drift early prevents traffic loss.
Why do search rankings slowly decline over time?
A page loses its search ranking because the query intent changes or newer pages offer better technical answers.
Search engines evaluate freshness for many technical topics. An article about setting up a specific database version slowly becomes obsolete as new major versions are released. When a reader clicks your link and sees an outdated version number, they immediately click the back button. Search engines register this fast bounce rate as a signal that your page no longer solves the query.
Competitors also study the top-ranking pages. They look at what your article misses and write a version that covers those gaps. Over months or years, search engines test these new pages in the live results. If users stay on those new pages instead of returning to the search results, your page drops a spot. This happens gradually. A page at position two might take six months to drift to position five.
Search intent also fractures. A broad keyword might originally bring informational traffic to a tutorial. Later, the algorithm might determine that users typing that exact phrase actually want a software tool or a template. When intent fractures, informational posts drift downward while product pages rise.
How do you measure search position drift in Google Search Console?
I measure drift by comparing two consecutive three-month periods in the Google Search Console performance report.
Open the performance tab and select the date range filter. Change it to the compare tab and select the option to compare the last three months to the previous period. Apply the filter. Click the average position box above the chart to display position data in the table below.
Sort the table by the position difference column. You are looking for queries where the position has increased by a positive number. In Google Search Console, a higher position number means a lower rank. A difference of plus five means the page dropped five spots.
Switch the view from queries to pages. This shows you exactly which URLs are losing ground across all their ranked terms. Click into a specific URL that is losing traffic. Switch back to the queries tab to see which specific keywords are dragging the page down.
I always filter out branded search terms during this audit. People searching for your specific app name will almost always find your site. You only want to measure the drift on unbranded, generic keywords where you compete against other sites.
Sometimes the position holds steady but the clicks drop. You can read more about high-impression low-click search queries to figure out if the problem is your title tag rather than your ranking.
At what average position should you update an old post?
You should update an old post when its average position rests between 8 and 20.
Pages ranking in positions one through three are doing their job. Modifying them carries a high risk of breaking whatever search engines currently favor. Minor copy edits are safe, but structural changes or heavy rewrites can easily cause a top-ranking page to drop. I leave these alone.
Pages ranking worse than position 20 usually miss the search intent completely or lack the necessary topical authority. Tweaking a few paragraphs will not push a page from position 35 to position three. These topics require entirely new articles or massive, ground-up rewrites.
The sweet spot for an update is the bottom of page one or anywhere on page two. Positions 8 through 20 indicate that search engines understand your page is relevant to the query. The page just lacks the depth, freshness, or formatting of the pages ranking above it.
This is where an update provides the highest return on your time. The click-through rate curve drops steeply after position three. Pushing a page from position 12 to position 4 dramatically increases your daily organic traffic, often multiplying it by ten.
What specific elements of a technical post need updating?
I update deprecated code snippets, broken external links, and the structural formatting of the page.
Start with the code. If your tutorial uses an API that released a breaking change last year, readers will bounce the moment they see the outdated syntax. Update the code blocks to the current major version. Add a small note at the top of the post indicating which software versions the article covers.
Check your outbound links. Documentation moves, and linking to 404 pages signals that your article is abandoned. Swap broken links for current documentation URLs. If you included screenshots of software interfaces, take new screenshots to reflect the current UI.
Look at the search results for the keywords you are trying to reclaim. Open the top three pages. Look at their heading structures. They might answer a specific sub-question that your article ignores. If every page outranking you includes a section on troubleshooting common errors, you need to add a troubleshooting section to your post.
Update your internal links. When an old post slowly loses traffic, it often lacks internal links from your newer content. Find three or four articles you published recently that mention the topic of your old post. Edit those newer articles to include a direct link back to the post you just updated. This passes fresh link equity to the older URL and helps search engines discover the new changes faster.
Do not change the URL slug. Changing the URL forces search engines to evaluate the page as a brand new entity, discarding the historical signals it built up over time. If you absolutely must change a URL, implement a permanent redirect from the old slug to the new one.
How do you handle keyword cannibalization during an update?
I handle cannibalization by ensuring the updated post targets a distinct intent that does not overlap with my other published articles.
Cannibalization happens when two pages on your site compete for the same search query. Search engines get confused about which page is the canonical answer, and both pages end up ranking poorly. You can find these conflicts by typing a site search operator into Google, followed by your target keyword. If multiple articles from your blog show up, they might be cannibalizing each other.
When you update an old post, you might be tempted to broaden its scope to capture more keywords. If you already have a dedicated page for those new keywords, broadening the old post harms your site structure. Keep the updated post tightly focused on its original primary keyword.
If you find two old posts that answer the exact same intent, do not update both. Merge the weaker post into the stronger post. Redirect the weaker URL to the stronger URL. This consolidates your ranking signals into a single, highly relevant page.
How do you turn drifting keywords into new content?
You turn drifting keywords into new content by extracting queries where you rank between 8 and 20 and writing dedicated articles for them.
Sometimes an old post ranks at position 15 for a keyword it only mentions in passing. The post might be a broad guide to deploying a web app, but it catches a few impressions for a specific continuous integration tool. Updating the old post might not be enough to secure a top spot for that specific tool. The search intent demands a dedicated page.
I take that specific keyword and draft a new article centered entirely on it. I then link from the old post to the new, highly targeted article. This builds a topical cluster and passes authority directly to the new page.
If you manage a blog for a small site, writing new articles for every slipping keyword takes time away from writing code. Social distribution handles itself after publication, though you can use prefilled composer handoffs for networks where you want to keep manual control.
How do you force search engines to evaluate your updated post?
I force evaluation by submitting the specific URL to the URL Inspection tool in Google Search Console and requesting indexing.
Do not wait for search engines to naturally crawl your site. Depending on your crawl budget, a natural recrawl of an old URL might take weeks or months. You want credit for your updates immediately.
Paste the full URL of your updated post into the inspection bar at the top of Google Search Console. Wait for the live test to complete. Check the page resources tab to ensure all your CSS and JavaScript files load correctly, as broken rendering can negatively impact your rank. Once the report loads, click the button to request indexing. This puts the URL into a priority queue.
The initial re-evaluation usually happens within a few days. The rankings will fluctuate immediately after the recrawl. The page might even drop a few spots temporarily as search engines test the new content against user behavior.
Wait at least four weeks before judging the success of an update. If the average position stabilizes higher than it was before the update, the content refresh was successful. If it remains flat, you need to revisit the search results, check what the top three pages are doing differently, and plan another revision.
To pick keywords directly from your Google Search Console data where you rank at position 8 through 20 and turn them into scheduled drafts, use AmplifySignal.