Design

How to Rebuild or Retire an Auto-Generated Blog Post (Without Losing Rankings)

By Ken Key Apr 8, 2025 6 min read Updated Oct 16, 2026

How to Rebuild or Retire an Auto-Generated Blog Post (Without Losing Rankings)
The short answer

Check Search Console first: impressions, clicks, average position, and whether the URL is indexed. No impressions and not indexed means delete or redirect. Real clicks or backlinks mean keep the URL and rewrite the content. Never 404 a page with links.

Part of the complete SEO guide.

On this page

Most sites that have been online for a few years are carrying dead weight: posts that were spun up by a plugin, scraped from a feed, or written by an AI tool with no editing pass. They read like filler because they are filler. The good news is you don’t have to nuke them. You have three real options, and picking the right one per URL is the whole job.

First, figure out what the post is actually doing

Before you touch anything, pull the data. In Google Search Console, open the Performance report, filter by page, and look at the last 16 months for each suspect URL. You want three numbers: impressions, clicks, and average position. Then check the Pages report to see if it’s indexed at all.

That splits your auto-generated posts into four buckets:

  • No impressions, not indexed. Google already decided it’s junk. Delete it or redirect it. Nothing to preserve.
  • Impressions but almost no clicks. It ranks on page two or three for something. There’s a topic here worth keeping, but the content needs to be rewritten from scratch.
  • Real clicks. Rewrite it. Don’t delete a URL that earns traffic, even if the content embarrasses you.
  • Backlinks from real sites. Keep the URL alive no matter what. Rewrite or redirect, never 404.

One caveat that trips people up: a page can pull clicks from a query it doesn’t deserve to rank for, and rewriting it can cause it to lose that accidental ranking. Check the actual queries before you commit. If the traffic is coming from your brand name alone, that’s not a signal the content is good. It’s a signal people already know you.

Option 1: Rewrite it properly

This is the right call for anything with clicks or backlinks. The trick is to keep the URL and the intent, and replace the substance. Not a light edit. A rewrite.

What made the original bad is usually one of these:

  • It answers a question nobody asked, because the title was generated from a keyword list.
  • It says nothing specific. No numbers, no examples, no code, no names.
  • It repeats the same idea in five different phrasings to hit a word count.
  • It’s dated in a way that’s now wrong. Old pricing, old tool names, old best practices.

A rewrite fixes all four at once. Pick one specific reader (a small business owner, a junior dev, a marketing manager) and write only to them. Cut every sentence that doesn’t teach them something or move them forward. If a section could be deleted without anyone noticing, delete it.

When you republish, don’t change the URL. Update the dateModified in your structured data and bump the visible “last updated” date. If the title was the awkward part, you can change the <title> tag and the H1 without changing the slug, which is the safest move for rankings.

Option 2: Consolidate several weak posts into one strong one

If you have five thin posts all circling the same topic, that’s not five assets. It’s one asset spread across five URLs, each too weak to rank. Merge them.

Pick the URL with the most clicks or the best backlink profile as the survivor. Write one comprehensive piece on that URL. Then 301 the other four into it. Here’s the redirect block if you’re on Apache:

# .htaccess
Redirect 301 /blog/old-post-one/ /blog/the-real-guide/
Redirect 301 /blog/old-post-two/ /blog/the-real-guide/
Redirect 301 /blog/old-post-three/ /blog/the-real-guide/

On Nginx it looks like this:

# nginx.conf, inside your server block
location = /blog/old-post-one/ { return 301 /blog/the-real-guide/; }
location = /blog/old-post-two/ { return 301 /blog/the-real-guide/; }
location = /blog/old-post-three/ { return 301 /blog/the-real-guide/; }

Two things people get wrong here. First, redirect to the final destination directly, not through a chain of redirects. Second, update your internal links so they point at the survivor, not at the old URLs. Otherwise you’re bouncing every click through a 301 for no reason. A quick search-replace in your database handles it:

UPDATE wp_posts
SET post_content = REPLACE(
 post_content,
 'https://example.com/blog/old-post-one/',
 'https://example.com/blog/the-real-guide/'
)
WHERE post_type = 'post';

Back up the database before you run anything like that. I’ve watched people run a replace on the wrong table and spend a weekend restoring from a backup they weren’t sure existed.

Option 3: Delete it

If a post has no traffic, no backlinks, and no strategic value, delete it. Not redirect. Delete. A 410 Gone is the honest signal, and Google handles it fine. The reason people are scared to delete is a myth that clearing out old content hurts your site. It doesn’t. Thin content sitting around doesn’t help you, and it can dilute how Google evaluates the rest of your site.

The one thing to check: does anything link to it internally? If so, remove those links or point them somewhere useful first. Orphaned links to 404s are a small but real annoyance for both users and crawlers.

What “auto-generated” actually means in 2026

The category has gotten blurrier. In 2026 you’re dealing with at least four flavors of machine-generated content, and they don’t all deserve the same treatment:

  • Spun or scraped content from the 2010s. Almost always delete. There’s nothing to save.
  • AI-drafted posts with no editing. Usually rewrite. The bones might be fine, the flesh isn’t.
  • AI-assisted posts you actually edited. These are often fine. Audit them like any other post.
  • Programmatically generated pages (location pages, product variants). These can be legitimate if each one has real, unique value. If they’re just a template with a city name swapped in, they’re a liability.

Google’s guidance has been consistent for years: it’s about whether the content helps people, not how it was made. That’s genuinely the test. Read the post as a stranger and ask if it told you anything you didn’t already know. If not, it’s not an asset.

A practical order of operations

Don’t try to fix everything at once. Work in batches and measure as you go.

  1. Export all your URLs with their clicks, impressions, and backlink counts into a spreadsheet.
  2. Sort into the four buckets above.
  3. Delete the dead ones. Do this first. It’s fast and it clears the noise so you can see what’s left.
  4. Consolidate the clusters. Merge weak siblings into one strong page.
  5. Rewrite the ones with real traffic, one at a time, properly.
  6. Recheck Search Console 4 to 6 weeks later. Watch for the URLs that lost position and investigate those specifically.

If you’re doing this yourself and you’re not sure whether a URL is worth saving, the backlink check is the tiebreaker. A single link from a site that matters is worth more than a month of impressions from a query nobody converts on.

For a lot of small businesses, this cleanup is the single highest-return thing you can do to a site, more than any new post you could publish. If you’d rather hand the audit off than do it yourself, that’s the kind of work a Long Island web design shop deals with regularly, and it’s usually a few days of work, not a rebuild.

FAQ

Will deleting old blog posts hurt my SEO?

Not if the posts have no traffic and no backlinks. Deleting genuinely dead content helps by removing noise. The risk is deleting a URL that has links pointing at it, which is why you check backlinks before you delete anything.

Can I just run old posts through an AI rewriter?

You can, but you’ll usually get the same problem in different words. The issue with a bad post is almost never the phrasing, it’s that there’s no real information in it. A rewrite needs a human who knows the topic deciding what the post should actually say.

How long until I see results after a rewrite or consolidation?

Give it 4 to 6 weeks before you judge anything. Google needs to recrawl and reassess the page, and rankings often dip slightly before they settle. Track it in Search Console rather than checking rankings daily, which will just make you anxious for no reason.

Have a project like this?

I do Long Island web design for Long Island businesses. If you need this done right, let’s talk.