Deleting a Page That Ranks

Traffic is not the argument

The instinct is that a page with traffic must be kept, and it is a poor rule.

A page can bring four hundred visitors a month, produce no enquiries, and generate six telephone calls from people wanting something you do not do.

That page has a cost, in your time and in the pages it competes with, against no return.

So the question is what the page produces rather than what it attracts.

When a ranking page should go

• The service no longer exists.
• It attracts the wrong enquiries.
• It competes with a better page.
• It is out of date and cannot be fixed.
• It says something you would not say now.
• Maintaining it costs more than it returns.

The second is the most overlooked. A page ranking for a cheaper version of your service brings people who cannot afford you, and every one of those is a conversation that ends in nothing.

Merge before deleting

Since removal is rarely the best available option.

Where the page competes with a better one, move its useful content into that page and redirect.

Where it is out of date, updating frequently costs less than losing what it has accumulated.

Where it attracts the wrong enquiries, rewriting it to be explicit about price and scope filters them without losing the position.

That third option is usually the right one and it is rarely considered, because the page is framed as a problem rather than as a filter that is set wrongly.

Deleting is for the cases where none of those apply.

A worked example

A business had an article ranking well for a general question, bringing about six hundred visits a month.

It produced roughly one enquiry a quarter, and a steady stream of calls asking for advice from people in other provinces.

They rewrote it rather than deleting it: the same question answered, with the areas covered stated at the top, a price range, and a clear line about what they do not do remotely.

Traffic fell by about a third, the irrelevant calls stopped almost entirely, and enquiries from it rose to two or three a month.

The page had been ranking for the right question and answering it for the wrong audience.

Deleting it would have lost a page that turned out to be valuable.

Check what it earned before touching it

Which is the step that prevents regret.

Look at what queries it receives, whether anything links to it, and what people do after landing on it.

A page with external links pointing at it is carrying something that took years to accumulate, and that argues strongly for redirecting rather than removing.

A page nobody links to and nobody proceeds from is a much easier decision.

Write both down before making the change, since you will want the comparison in three months.

Ten minutes of checking prevents the situation where nobody can tell whether the removal helped.

Redirect rather than delete

Almost always, and for the reasons covered elsewhere.

Point it at the closest equivalent, permanently.

Where the page was the only one on its subject, the closest equivalent may be a section of another page rather than a whole page.

And where there is genuinely nothing equivalent, a gone page with routes onward is better than a redirect to something unrelated.

The people arriving from old links are the ones you are protecting, and they came for something specific.

Expect the traffic figure to fall

Which needs saying before rather than after.

Removing or narrowing a popular page reduces your traffic, visibly, in the report somebody looks at monthly.

If nobody has been told why, that fall reads as a failure and the change gets reversed.

Say in advance what you are doing, what you expect to happen to traffic, and what you expect to happen to enquiries.

Then check both after three months against what you predicted.

That is also the only way to know whether the judgement was right.

Do it deliberately, not gradually

Since a slow removal produces an unreadable result.

Where several pages are going, do them in one pass on one date and note it.

Changes spread over four months cannot be attributed to anything, and the site spends the whole period in a partial state.

One pass, one date, then three months of not touching it.

Which is difficult and is the only way to learn anything from it.

Save the content before it goes

A small precaution that costs a minute and occasionally saves a great deal.

Copy the full text of the page somewhere before removing or rewriting it, along with the date and what it was ranking for.

Reversing a decision three months later is straightforward with the original text and awkward without it, and the archived copies on the web are not always complete.

It also means the useful paragraphs can be moved into another page later, which is frequently where they end up.

Keep it with the record of what you predicted, since the two belong together when you come to review the decision.

The counter-case

Keeping it is frequently right.

A page bringing people who do not enquire may be building recognition, attracting links that help other pages, and reaching people who will need you in two years.

Traffic also has value in trades where somebody researches for months before hiring anybody.

And a page that ranks well is a real asset, which should not be discarded over a quarter's figures.

Check what it earns, try rewriting it as a filter, redirect rather than delete, and say in advance what you expect.

The decision

1. Count what it actually produces.
2. Count the calls it wastes.
3. Check what links to it.
4. Try rewriting it as a filter first.
5. Merge into a better page if one exists.
6. Redirect rather than delete.
7. Predict the numbers before changing.

Step four is the option almost nobody tries, since a page attracting the wrong people is usually a page missing a price and a statement of scope rather than a page that should not exist.

The audit version of this 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.