The short answer
Over the 28 days to October 11, 2026, 10,633 pages on websites we build and run appeared in Google search results. When we sorted them by type and by the searches that found them, three things stood out:
- 88% of impressions came from searches of three words or more. Short, broad searches were only 12%.
- Cost and pricing pages earned the most impressions per page, about 27 each, more than four times a typical industry page.
- Homepages got the clicks. They earned 2.8% of their impressions as clicks, roughly ten times the rate of other pages.
For a small business, that adds up to a simple to-do list: answer the long, specific questions customers type, publish honest cost information, and make the homepage the clear next step.
What we measured
We used the Google Search Console data for the websites we own and run, excluding the large business directory so that its hundreds of thousands of listings did not drown out normal business pages. For each page that appeared at least once in Google during the 28 days, we recorded its impressions, clicks and average position, then grouped pages by type from their address and content: homepage, cost and pricing, comparison and alternative, guides and answers, industry pages, local pages and other pages such as news articles.
For searches, we grouped the queries Google reported by length. Google hides very rare queries for privacy, so the query totals are smaller than the page totals.
Finding 1: Long, specific searches do the work
| Search length | Impressions | Share |
|---|---|---|
| 1 to 2 words | 7,354 | 12% |
| 3 to 4 words | 31,918 | 52% |
| 5 to 7 words | 18,378 | 30% |
| 8 words or more | 4,083 | 7% |
Most people do not search "website" or "plumber". They type what they actually want to know: "how much does a website cost per month", "how to price residential electrical work", "best bank for small business in San Diego". Pages that answer one of those questions directly are the pages Google shows.
This is also how people talk to ChatGPT and other assistants, only longer. A page written to answer a real question in plain words works for both.
Finding 2: Cost pages pull their weight
| Page type | Pages | Impressions | Per page | Click rate |
|---|---|---|---|---|
| Homepage | 19 | 879 | 46 | 2.84% |
| Cost and pricing | 250 | 6,723 | 27 | 0.06% |
| Comparison and alternative | 279 | 4,168 | 15 | 0.19% |
| Guides and answers | 1,023 | 9,058 | 9 | 0.24% |
| Industry pages | 2,573 | 15,503 | 6 | 0.25% |
| Local pages | 2,392 | 12,363 | 5 | 0.28% |
| Other (news, tools, misc.) | 4,097 | 49,251 | 12 | 0.21% |
Pages about cost, such as "what does a coffee shop loan cost" or "how much does a website cost per month", earned about 27 impressions each. That is more than four times an industry page and five times a local page. People research price before they contact anyone, and very few local businesses publish real numbers.
If your competitors hide their prices, a clear page that explains your price range, what changes it and what is included is one of the easiest pages to get found for.
Finding 3: Your homepage is where clicks happen
Homepages were only 19 of the 10,633 pages but had the highest click rate by far, 2.84%. Searches that reach a homepage are usually searches for the business itself or for exactly what it does. Every other page type had a click rate between 0.06% and 0.28%, because those pages are found by people still researching.
Content pages earn the impressions that make a business known. The homepage and contact page turn that into a call or a form. Both need to be good, and each page needs a clear link to the next step.
Finding 4: Most pages take time, and many never rank
Most pages with impressions sat beyond the first page of Google, at an average position in the 50s to 80s, during their first weeks. That is normal for new pages on young websites. The pages that climbed fastest answered a specific question nobody else answered well, had a clear title that matched the search, and were linked from related pages on the same site.
Publishing more pages helps only when each page is different and useful. Near-duplicate pages, such as the same text with a different city name, compete with each other and are the first to be ignored. We check every new page against the rest of the site before it goes live.
The website tasks that matter, in order
| Task | Why it matters in the data | How we handle it |
|---|---|---|
| Answer one real question per page | 88% of impressions come from 3+ word searches | Each page targets a question customers actually type |
| Publish clear prices | Cost pages earn ~27 impressions each | A pricing page and cost guides for your trade |
| Make the homepage the next step | Homepages convert at 2.8% | Clear calls, forms and booking on every page |
| Link related pages together | Linked pages rank sooner | Every page links to its parent and siblings |
| Let search and AI bots in | Blocked pages cannot rank or be quoted | Sitemap, llms.txt, schema and bot access checked at launch |
| Keep adding useful pages | More useful pages, more chances to be found | Monthly SEO adds new pages every month |
Limits of this study
These are 28 days of data from websites we own, mostly in finance, news and website services, plus the industry websites we run. Results for your trade and city will differ. Impressions show how often a page appeared in Google, not how many people visited, called or bought. Page types were assigned by address and content, so a few pages could fit more than one group. We do not guarantee rankings, traffic or leads from any page.
Sources and method
- Google Search Console, Search Analytics API (impressions and clicks by date, page and query), pulled October 12, 2026.
- Excludes the large business directory site and anonymized queries that Google does not report.
- Data window: September 13 to October 11, 2026. Websites are not named. You may quote these figures with a link to this page.
