Removing Content and Ranking Better

Why it works at all

The result is counterintuitive enough that businesses resist it, so the mechanism is worth setting out.

Three things are happening, none of them mysterious.

Weak pages compete with strong ones for the same terms, so a search engine has to choose between four mediocre pages about a topic rather than being shown one good one.

Internal links spread across many pages give each less weight than they would concentrated on a few.

And a site full of thin pages presents an unclear picture of what it is actually about.

The competing-pages problem

The largest of the three for most small sites, and the easiest to verify.

Where four pages on your site touch the same subject, none of them is the obvious answer to a query about it.

You can see this directly: search for the topic with your own domain specified and look at how many of your pages come back, and which one is chosen.

Frequently the page chosen is not the one you would have chosen, which is the symptom.

Consolidating those four into one, with the useful content from each merged, gives a single unambiguous answer.

That is usually enough on its own to improve position, without any new material being written.

What to remove

• Pages with no traffic over a year.
• Near-duplicates of a stronger page.
• Thin pages created for coverage.
• Old announcements and outdated posts.
• Tag and category pages nobody uses.
• Anything you would not show a customer.

The last is the test that catches what the figures miss. A page you would be embarrassed to have somebody land on is doing nothing for you regardless of what the numbers say.

Check before deleting

Because a page with no traffic may still be doing something.

Look at whether it has links pointing to it from other sites, which is value you would be discarding.

Look at whether it converts, since some pages get very little traffic and produce enquiries from most of it.

Look at whether it is linked from your own navigation or from other pages, which tells you whether removing it breaks anything.

Where a page has external links or converts, merge it into a survivor rather than removing it, and redirect the address.

That preserves what it was contributing while removing it as a competitor.

A worked example

A site with a hundred and forty pages, of which about ninety had received under ten visits in a year.

The ninety were mostly old blog posts and location variants.

Fifty-two were deleted and redirected to the relevant service page. Twenty-one were merged into eight stronger pages. Seventeen were kept because they had external links or produced enquiries.

The site went from a hundred and forty pages to about seventy.

Over the following four months, the main service pages moved up for their primary terms and total traffic rose despite the site being half the size.

Nothing new had been written during that period.

Redirect rather than delete

The mechanical point that determines whether this helps or hurts.

Every removed address should redirect permanently to the most relevant remaining page, not to the home page.

Redirecting everything to the home page is treated as effectively a removal and wastes whatever the page was carrying.

Update internal links so they point at the survivor directly rather than through a redirect.

Update the sitemap and resubmit it, so the removals are noticed rather than discovered slowly.

Then check for errors a fortnight later, since a consolidation done at speed reliably leaves a few links pointing nowhere.

Measuring whether it worked

Since this is the part that separates evidence from belief.

Record, before starting: total traffic, traffic to the pages you are keeping, positions for your main terms, and enquiries.

Expect total traffic to fall initially, because you removed pages that were receiving some.

Watch the traffic to the surviving pages rather than the total, since that is where the improvement should appear.

Give it three months, since nothing here moves quickly, and compare against what you recorded rather than against memory.

Where surviving pages have not improved in that time, the removal was neutral rather than harmful, and the site is still simpler to maintain.

Do it once, not gradually

A practical point about sequencing.

Removing pages a few at a time over months makes the effect impossible to attribute and leaves the site half-converted throughout.

Do the analysis, decide the whole list, and execute it in one session.

That gives a clean before and after, which is the only way to learn anything from it.

It also means the internal linking is consistent afterwards rather than reflecting three different intermediate states.

The maintenance argument on its own

Worth stating separately, because it holds even if the position never improves.

Seventy pages can be reviewed annually by one person. A hundred and forty cannot, which is why the site had ninety stale ones.

Every page carries an ongoing obligation: prices to update, services to keep accurate, links to check, and claims to keep true.

A site sized beyond what anybody will maintain drifts out of date page by page, and the drift starts with the pages nobody visits and spreads.

So pruning buys a site that stays accurate, and that is worth doing whether or not anything moves in the results.

The counter-case

Pruning is not a growth strategy.

It concentrates what a site already has and does not create anything, so a site with nothing worth concentrating gains little.

The improvement is also finite and happens once, unlike writing something genuinely useful.

There is a real risk of over-deleting, particularly of pages that quietly produce enquiries from small traffic, which is why checking before deleting matters more than the deleting.

Prune once, keep anything that converts or has external links, and then spend the effort on the pages that survived.

The procedure

1. Record traffic, positions and enquiries first.
2. List pages with no traffic over a year.
3. Check each for external links and conversions.
4. Merge what has value, delete the rest.
5. Redirect to the relevant page, not the home page.
6. Update internal links and the sitemap.
7. Compare after three months.

Step one is what makes this evidence rather than faith, since total traffic falls at first and without a baseline that looks like the change failed.

Which pages qualify is covered in the page you should delete.

Frequently asked questions

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.