How many pages does a small business website need

The wrong version of this question is "how big should my site be." Size is an output. The input is how many distinct things people search for that you want to answer.

A restaurant with one location and a menu needs a small site. An HVAC company covering nine cities with six services needs a much larger one, and building it as a five-page brochure means it will never rank for anything except the company name.

What is the actual formula?

Count the things you sell that a customer would search for separately, then count the places you serve.

One page per service. If somebody searches "drain cleaning" and you have a page called Services listing eleven items, you have nothing that matches that search. A page that covers everything competes for nothing. This is the single most common structural failure on small business sites.

One page per location, if you actually serve it. Not one per zip code you would drive to in an emergency. One per place where you have real presence, real jobs, and something specific to say. Twelve near-identical city pages with the town name swapped is a pattern search engines have recognized for a decade.

The three standards. Home, about, contact. The about page matters more for local businesses than for anyone else, because customers hiring a person want to see the person.

Anything you get asked constantly. Pricing, financing, insurance, parking, hours. If your phone answers the same question forty times a week, that question is a page.

Add those up. That is your number, and it is defensible line by line, which is exactly what a quote should be built from.

There is evidence that the page-level detail matters rather than just the site's existence. Whitespark's Local Search Ranking Factors report, published 6 November 2025, had 47 local search experts score 187 factors. Keywords in the landing page title tag ranked 17th at 146 points, geographic keyword relevance of content ranked 18th at 142, and keywords in landing page headings ranked 20th at 135. Every one of those is a per-page property. A single Services page cannot hold six different title tags. Grade it as expert opinion rather than a controlled test, because that is what it is, but the structural point does not depend on the scores being exact.

What does the arithmetic look like?

Is more pages always better?

No, and this is where owners get talked into bad projects.

A page that exists without anything specific on it is worse than no page. It competes with your real pages, it dilutes the site, and it needs maintaining forever. Every page you publish is a page somebody has to keep true.

The maintenance cost is the part nobody prices. Hours, staff, service offerings, and photos all go stale, and a forty-page site goes stale four times faster than a ten-page one. What website maintenance actually covers is worth reading before you commit to a page count you cannot sustain.

There is a real version of a large site, and its test is whether each page could be found and used on its own. If a page would be useless to a person who landed on it directly from search, it is filler.

Should I add an FAQ page, and should it have schema?

Add the page if customers genuinely ask those questions. Do not add the markup expecting a search result out of it, and be careful who is charging you for it.

Google announced the change itself. In a Search Central post dated 8 August 2023, Google stated that FAQ rich results from FAQPage structured data would only be shown for well-known, authoritative government and health websites, and that for all other sites the rich result would no longer be shown regularly. An update appended to the same post in September 2023 said Google Search no longer shows How-to rich results on desktop and that the result type is deprecated.

That has been true for three years, and FAQ schema is still sold as a visibility tactic. Google's own generative AI guidance is equally blunt in the other direction: structured data is not required for AI features and there is no special markup that produces them.

So the honest position is a split. FAQ content is valuable and FAQ markup is not the reason. Write the answers because customers ask them, put them on the page in plain text, and put the markup budget somewhere it does something. For a local business the structured data that still earns its place is LocalBusiness with the most specific subtype for your trade, Organization, and Service, implemented so that it matches the visible text on the page.

How do I decide what the pages should be about?

Look at what your competitors are categorized as, and what people actually type.

Your Google Business Profile categories are the closest thing to a public statement of what a business is understood to do, and your competitors' categories tell you what the market thinks the service names are. What your competitors' categories reveal and how to find them is a twenty minute exercise that usually produces two or three service pages an owner had not thought to build.

There is large-sample support for taking naming seriously. Darren Shaw's analysis of 1.8 million Google Business Profiles across 4,209 categories, published on Search Engine Land in August 2026, found that the exact-match category outperforms adjacent categories across almost the whole dataset. Hair salons win as Hair salon. Plumbers win as Plumber. The same instinct applies to your page titles: name the thing the way a customer names it.

Then use the words customers use, not the words the trade uses. "Tankless water heater installation" is a page. "Hydronic solutions" is not.

Then write them properly. A service page that is three sentences and a stock photo does not rank and does not convert, and this is where most page-count plans collapse. The bottleneck is never the building, it is that somebody has to write fifteen pages. Whether to write your website copy yourself or hire it out is the decision that determines whether your page plan survives contact with reality.

What about the pages that convert rather than rank?

Different job, and they should be counted separately.

Ranking pages answer a search. Converting pages take somebody who already decided and turn them into an inquiry. Your contact page, your quote form, your booking page. There are usually only two or three of them and they carry most of your revenue.

Which means they deserve more attention than the other twelve pages combined. Baymard Institute, which sells UX research and benchmarking, published checkout research in June 2024 finding the average checkout flow runs 5.1 steps and 11.3 form fields, that 17% of users have abandoned a purchase because checkout was too complicated, and that field count matters considerably more than step count. Take the direction and leave the number: that is e-commerce checkout research, and no study with a disclosed sample exists for quote forms on local service websites. Anyone quoting you a specific conversion lift from cutting fields on a plumbing form is quoting something that has not been measured.

What you can do is look at your own. Finding out where people abandon your forms without expensive tools is the fastest way to see the money you are currently leaving on the table, and it uses data you already have.

A useful discipline: for every page, write down the one thing you want a visitor to do next. If you cannot, the page has no job and it goes on the maybe list.

Where does the one-page-per-service rule break down?

Three places.

When the services are not distinguishable to a customer. If nobody searches for the two things separately, and the copy on both pages would be the same paragraph with a word swapped, that is one page. The rule follows the search behaviour, not your price list.

When you cannot staff the writing. A page plan you will not finish is worse than a smaller one you will. Eleven published pages beat twenty drafted ones, every time.

When the constraint is elsewhere entirely. If your site converts at a rate you would be embarrassed to state out loud, more pages sends more people into the same leak. Fix the two converting pages first. The ranking pages will still be there next quarter.

What to do this week

List every service you sell, in customer language. Then list every city or neighborhood where you have actually worked in the last year. Those two lists plus home, about, and contact are your page map.

Then check what exists. Open your current site and match your real pages against that map. The gaps are your build list, in priority order.

Then run the demand filter and the maintenance arithmetic above. Both usually shorten the list, which is the point.

Then be realistic about capacity. If you can write one solid page a week, a fifteen-page plan is a four-month project. Plan the launch around the pages you can actually finish, and add the rest afterward, because what happens in the 90 days after a site launches should include publishing rather than going quiet.

Questions about your page map?

Email me at eric@seod.com with a list of the services you sell and the cities you serve. I will send back the page map I would build from it, with the pages ranked in the order I would write them and any pages I would tell you to skip. It is a short reply and it costs nothing.

More on scoping in the web design and branding library.