What a useful page count looks like
We build websites, so we are not neutral. Our interest is in sites that earn their keep, which means pages with a job to do.
A page earns its place when someone can arrive from a search, read it, decide and act without opening another tab. That is the test. A page that cannot pass it is either too thin to help or repeating a page that already does the job.
The count comes from your customers. Write down what they must settle before they get in touch: whether you do the job, whether you do it for someone like them, what is included, how the price is worked out, and what happens next. Where the answer changes from service to service, each needs a page of its own. Where it does not, one page carries it.
Shape matters too. Keep the menu to the things people came to buy and put the detail one level below. A menu that mirrors how the business sees itself can make the visitor translate before they choose. Small questions that do not deserve a page of their own can go on an FAQ page instead.
Check the pages you already have
Start with what is there. Adding pages to a site with weak pages spreads the problem across more addresses.
Merging is the repair that gets skipped. Two pages on one subject split the proof and the links other sites point at you. One good page does the work of both.
A page that repeats the home page in different words is not a page. Neither is a page for a service you mention but do not really sell.
- Write one sentence for each page: who it is for and what they should do next. If you cannot, the page has no job.
- Find pages that answer the same question. Keep the stronger address, merge the content, and point the other address at it.
- Check that each page is linked from the page that mentions its subject. A page nothing links to is a page nobody finds.
- Read the bottom of each page. There should be one clear next step that suits the job.
- Check that any page can be someone's first page. People arrive from search, so the home page is not the only door.
When a service needs its own page
Two services that the same customer buys together, with the same questions, can share one page with a section each. Two services that attract different customers, with different proof and different next steps, can weaken each other on one page.
Locations are where page counts get out of hand. A page for a town is worth having only when the answer there is different: a different team, a different coverage area, different local proof. The same words with a swapped place name give the customer nothing. One page that lists the towns you cover does the job.
Depth beats breadth. A page that answers the follow-up questions, what is included, what happens on the day, and what you need to do first, keeps the enquiry on your site. A page that stops at the name of the service sends the customer back to search.
Questions to ask before you add or remove a page
Before a page goes up, ask what someone would type to find it, and whether another page already answers that. Then ask what would happen if you deleted it, and where its visitors would go. If you cannot answer both, the page is not ready.
Ask what decision the page is for, and whether that differs from the page beside it. Ask what proof it offers that no other page does: photographs of the work, qualifications, access details, what makes you the safer choice. Ask what links to it, and whether a customer can reach it in one click.
Ask what happens at the bottom. A page with no next step leaves the customer to invent one.
For removals, ask where the old address should point. A page that disappears without a redirect leaves broken links behind.
Mistakes that cost enquiries
The mistakes below all start the same way: a page built for the business, whether to fill a menu, to match a competitor, or to use a phrase someone told you to target. The customer is not in the room when that decision gets made. A page that does not help a customer decide is not doing a job.
- Splitting one job across thin pages, so no single page has enough to decide.
- Crowding unrelated services onto one page, so a customer cannot tell that you do what they searched for.
- A page for every town with the same words and a different place name.
- Blog posts about topics nobody searches for, while the service pages stay thin.
- Pages that nothing links to, so nobody arrives.
- Removing a page without pointing its address somewhere useful.
How to tell whether it is working
Google Search Console is the honest scoreboard for page count. Look for searches where your site appears and no page answers the question properly. Those are the pages to write next, because they come from real demand.
Then separate the problems. A page seen in search results and not clicked suggests a title and description problem. A page visited with no enquiry suggests a page problem. An enquiry asking about something you already offer points to a missing or buried page. An enquiry asking what a page already explains means people are not landing on that page.
Add pages one at a time, in the order the evidence gives you, so you can see what each one does. A batch of new pages tells you nothing about which one helped.
Results depend on your business, your area and the competition. A sensible page count makes a good result more likely. It does not guarantee one.
What to do next
Do the audit yourself first. One sentence per page, then merge, link or remove. That work costs nothing, and it can tell you more about your site than any list of page ideas.
Help is worth paying for when you cannot see which searches you are missing, or when every new page has to wait for someone else to build it. That is the work we do: we build small business websites to be found, keep them growing with new pages that answer what your customers search for, and make sure every enquiry reaches you. You change anything yourself by describing it in plain English. Hattie, our writer on the growth team, drafts those pages from real searches your site has no answer for, in your voice.