Pages that exist without saying anything

What counts as thin

Not word count. A short page that answers one question completely is not thin. A long page that says nothing in a thousand words is.

The test is whether somebody arriving from a search gets what they came for. If they read it and still need to look elsewhere, the page is thin regardless of length.

Typical examples: a service page with a heading and two sentences, a location page that changes only the town name, a category page listing three items with no description.

Not word count. A short page that answers one question completely is not thin; a long page that says nothing in a thousand words is.

How it accumulates

Placeholder pages created during a build and never finished. A page per service where most services got a paragraph and one got a page.

Pages generated for coverage rather than for readers, which is where a small site drifts toward the doorway pattern without intending to.

And old pages that were adequate years ago and were never revisited as the business changed.

Why it costs more than the page

Thin pages rarely rank, so the immediate cost is a page doing nothing. That alone would be tolerable.

In quantity it affects the whole site, because the proportion of useful pages is part of how a site is judged.

It also splits your own effort. Five thin pages on related subjects compete with each other and none of them wins.

In quantity it affects how the whole site is judged, and it splits your own effort across pages that compete with each other.

The wrong fix

Adding words to reach a target. Padding produces a longer page that still does not answer anything, and it is obvious to a reader within a sentence.

Generating filler text of any kind is worse, because now the page is long, empty and detectably formulaic.

The instinct comes from treating length as the measure. Length is a symptom of having something to say, not a cause.

Padding produces a longer page that still answers nothing, and it is obvious to a reader within a sentence.

The right fix, in order

First, decide whether the page should exist. Many thin pages are best deleted with a redirect to something related.

Second, consider merging. Five thin pages on adjacent subjects usually make one strong page that outperforms all five.

Third, if the page should exist alone, answer the question properly: what it covers, what it costs, what happens next, what people ask.

Delete, merge, or answer properly — and merging is usually the one that outperforms, as covered in cannibalization.

How to find yours

Sort your pages by word count and read the shortest twenty. That takes half an hour and finds most of it.

Cross reference with the performance report. A page with no impressions over months is not being shown to anybody, which is the market's verdict.

Be honest about the sets. If twenty pages differ only by one noun, that is the group to deal with first.

• Sort your pages by word count and read the shortest twenty
• Cross reference with the performance report for pages with no impressions
• Look for sets that differ only by one noun
• Look for placeholder pages created during the build
• Look for category pages listing three items with no description
• Ask of each: would somebody arriving get what they came for

How ours is prevented

Every page written by Website Builder Studio has to carry a real answer, a set of sections with substance, and questions that are actually answered before it can publish.

Pages are also measured against their siblings, so a page that is largely a copy of another with a word changed does not publish.

Every page passes a check that runs before anything publishes, validated against Google Search Essentials and modern web standards, and a page that does not reach the floor is refused rather than published thin.