Website Structure for SEO: How Site Depth Decides Rankings

By Garin Froehlich, Founder, GMScalezLast updated: November 202618 min read

Website structure for SEO is the set of decisions about which pages exist, how much each one covers, and how the pages connect. It sets the ceiling on what a site can rank for, because a search engine can only rank pages it finds, and each page can only be the best answer to one kind of search. A service business with one long "Services" page and a list of cities in the footer gives Google very little to match against the many things customers type. A business with a deliberate page for each service, each city and each real combination of the two gives Google a clear answer to each search.

Key takeaways

  • Search engines discover pages by following links, so a page that nothing links to is effectively invisible.
  • Each page should own exactly one search pattern; two pages competing for the same search split their strength.
  • Pages closer to the homepage tend to be crawled more often, so important pages belong within two or three clicks.
  • Internal links carry meaning through anchor text, and contextual links in the body of a page say more than menu links do.
  • Substantial means a page answers the questions a customer actually has, not that it reaches a word count.
  • City pages are legitimate when each carries something that could not exist on a sibling page.
  • Search Console and Analytics data show which pages to build, merge or rewrite next, so structure is a monthly practice.

Why structure comes before content

A search engine finds your pages by following links, and it can only rank a page once it has found and understood it. Good writing on a page that is hard to reach, or that competes with a near-identical page, earns far less than it should.

Google's documentation on JavaScript SEO basics explains that Google discovers URLs by parsing links in the href attribute of <a> elements. Google's guidance on making your links crawlable adds that links should be <a> elements with an href, and that anchor text should be descriptive, reasonably concise and relevant to the page it is on and to the page it links to. A crawl is a search engine's program visiting pages and following the links it finds.

The practical consequence is that a button which changes the page through script, with no real link behind it, gives the crawler nothing to follow. A city that appears only in a dropdown without real links may never be discovered.

The second reason structure comes first is that a page can only be the best answer to one kind of search. A page about "mobile mechanic Dallas" and a page about brake repair at your home serve different intentions. If one page tries to answer both, it answers each less completely than a focused page would. Think of structure as a filing plan for search demand: every search a customer might type should have one obvious home, and every page should have one search it exists to win.

The page types a service business needs

A service business needs a home page, one page per service, one page per city served, combination pages where demand exists, an FAQ page and a library of articles. Each type owns one search pattern and none borrows another's.

Page typeHome
Search pattern it ownsThe business name and brand search
Example patternThe business name itself
Page typeService page
Search pattern it ownsThe service plus the metro area
Example pattern"brake repair" plus the main metro
Page typeCity page
Search pattern it ownsThe business type plus the city
Example pattern"mobile mechanic" plus a city
Page typeService-plus-city page
Search pattern it ownsThat service in that suburb
Example pattern"battery replacement" plus a suburb
Page typeFAQ page
Search pattern it ownsQuestion-style searches across the business
Example patternQuestions about cost, timing and coverage
Page typeArticle
Search pattern it ownsA specific information gap with a modifier
Example patternA cost, signs or how-often question

For a service business we build a home page, one page per service at /services/{service}, one page per city at /locations/{city}, service-plus-city pages at /locations/{city}/{service}, a /faq page, a /gallery page and a /blog. The address pattern is useful in itself, because a visitor and a crawler can both read the hierarchy from the URL.

Live URLs are never renamed. When structure changes, the old addresses are permanently redirected, which tells search engines a page has moved for good, so pages already indexed keep resolving.

The rule that holds this together is keyword ownership. A service page owns the service plus the metro. A city page owns the business type plus the city. A service-plus-city page owns that service in that suburb. "Near me" is never a page target, because it describes where the searcher stands rather than something a page can be about.

Click depth and crawl depth

Click depth is the number of clicks it takes to reach a page from the homepage, and crawl depth is the same idea seen from the crawler's side. The deeper a page sits, the less often a search engine tends to reach it, so pages that matter should be shallow.

Botify, a crawling platform, reports on page click depth that "the deeper the page, the less likely Google will crawl it, especially on larger websites." Its recommendation is to keep "strategic pages at a depth no greater than 5."

For a small service site, 5 clicks is a generous allowance, and a stricter standard costs nothing. We hold every site to a rule that every money page, meaning a page meant to bring in a call or booking, sits within 2 clicks of the homepage, with a hard maximum of 3. The homepage links to the service list and the city list, and each of those links down to its children.

Depth is also about more than menus. A page reached only through a search box, or through a chain of "next page" links, is deeper than it looks to a person, because a crawler follows links and does not type queries.

Internal links are links from one page of your site to another. They let crawlers and visitors reach pages, and they pass information about what the destination is about.

Anchor text is the clickable words of a link. Google's guidance says it should be descriptive, reasonably concise and relevant to the page it is on and to the page it links to. "Brake repair in Dallas" tells a visitor and a search engine what waits at the other end, while "click here" tells them nothing.

Zyppy's study of 23 million internal links across 1,800 websites found that pages with more internal links, and more varied anchor text pointing at them, tended to receive more Google search traffic. This is a correlation, not proof of cause, but it matches the mechanism: a page linked to often, in several natural ways, is a page the rest of the site visibly treats as important.

A menu link says a page exists. A contextual link, placed in a sentence in the body of a page, says why it matters at that moment. A paragraph about a no-start problem that links to the battery replacement page, in words that describe it, gives the reader a next step and gives the crawler a reason.

An orphan page is a page that no other page links to. Crawlers that follow links will not find it, and visitors cannot reach it from anywhere on the site. Orphans usually appear after a redesign, or when a page is published without a place in the structure.

Page depth: what "substantial" means

A substantial page answers the questions a visitor has before deciding to call, completely enough that they do not need to go elsewhere. Length is a by-product of that, not the goal.

Backlinko's analysis of 11.8 million Google results found that the average first-page result contained about 1,447 words, though the study is correlational. Ahrefs studied 174,048 pages cited in AI Overviews, the generated summaries Google sometimes shows above regular results, and found a correlation of 0.04 between word count and citation position, which is effectively none. Thorough answers take space, yet padding a page does not make it more likely to be cited.

Our depth standard is therefore a list of questions rather than a word target. Every priority page answers nine: the problem, who it fits, the result, what is included, the process, how long it takes, the proof, why this market, and the next step. Each page also opens with a self-contained answer in its first 100 words, so a visitor, a search engine and an AI assistant can grasp the point before reading further. Pages built this way typically land between 800 and 1,500 words.

Where city pages go wrong

City pages go wrong when the only thing that changes from one page to the next is the city name. Google's spam policies describe scaled content abuse as "when many pages are generated for the primary purpose of manipulating search rankings and not helping users." They describe doorway abuse as "when sites or pages are created to rank for specific, similar search queries. They lead users to intermediate pages that aren't as useful as the final destination." Among Google's examples of doorways is "having multiple domain names or pages targeted at specific regions or cities that funnel users to one page."

The key phrases are "primary purpose" and "not helping users." A page that exists because customers in that city need information is on one side of the line. A page that exists to catch a search term and push the visitor elsewhere is on the other.

A legitimate location page carries something specific to that place that cannot be copied to a sibling by swapping a name: a photo of a job done in that city, a customer review that mentions it, or a first-hand local fact. The set should also be honest, reflecting cities the business genuinely serves, which is why we mirror the Google Business Profile service-area list.

Titles and descriptions

A title is the clickable headline in a search result, and a description is the short summary beneath it. They are written in the page's code, but Google decides what actually appears.

Zyppy's 2022 study of more than 80,000 title tags from 2,370 sites found that Google rewrote 61.6% of titles, and that titles over 70 characters were rewritten 99.9% of the time. Ahrefs measured 953,276 pages in 2021 and found that Google rewrote title tags 33.4% of the time. The figures differ because the samples and methods differ, but both show that a title left to chance is often replaced. The practical lesson is to write titles that give Google little reason to change them: short enough to fit, specific to the page, and matched to what the page covers.

ElementTitle length
Our rule50 to 60 characters, hard maximum 60
WhyOver-long titles are rewritten far more often
ElementTitle order
Our ruleKeyword first, then city and state
WhyThe searched words sit where readers look first
ElementHomepage title
Our ruleBusiness name only
WhyThe homepage owns the brand search
ElementDescription
Our rule140 to 160 characters, unique per page, never cut mid-word
WhyEach page needs its own reason to be clicked

FAQs built from real searches

An FAQ is only useful when its questions are ones people actually ask, and the most reliable source is data rather than invention. We draw questions from four places: the business's own Search Console history, which shows searches that earn impressions with no page answering them; sister sites in the same trade; Google autocomplete; and facts from real reviews.

Service pages carry 4 to 5 questions and city pages carry 3 to 4, placed where a visitor on that page would ask them. The main FAQ page holds 12 to 20 grouped questions assembled from those. Answers run 40 to 60 words with one real fact each, short enough to scan and specific enough to quote.

Using Search Console and Analytics to decide what to change next

Search Console shows how a site appears in Google, and Analytics shows what visitors do after they arrive. Read together, they tell you which pages to build, rewrite or merge.

Google's help page, What are impressions, position and clicks?, explains that position is the topmost position a link from the site occupied, averaged across the searches where it appeared. An average position is therefore a blend of many searches, and looking at one query at a time gives a clearer picture. Position also does not equal visits: Ahrefs reports that when an AI Overview appears, an AI Overview correlates with a 58% lower average click-through rate for the top-ranking page in its latest dataset, up from a 34.5% reduction in its April 2025 analysis.

Search Console history is collected daily and reviewed weekly for four patterns, each pointing to a specific change.

Pattern in Search ConsoleA page at positions 4 to 15 with real impressions
What it suggestsThe page is close to earning more clicks
Typical changeDeepen the content and add contextual links
Pattern in Search ConsoleClick-through rate out of line with position
What it suggestsThe title or description is not doing its job
Typical changeRewrite both
Pattern in Search ConsoleTwo pages competing for one search
What it suggestsOwnership of the pattern is unclear
Typical changeMerge them or sharpen each focus
Pattern in Search ConsoleA city with searches but no page
What it suggestsDemand with no home
Typical changeBuild a city page with local proof

Analytics key events (call taps, text taps, form starts and submits) show which pages turn visits into contact. A page with strong traffic and no contact needs a clearer next step.

Plain-language phrasing is another source of demand. The phrase "mechanic that comes to you" shows Any Hour Mobile Mechanic at an average position of 1.1 in Search Console, and the same theme earned DFW Mobile Auto Mechanic Pros 1,562 impressions at an average position of 10. Customers describe a mobile service in their own words, and pages that use those words meet them where they search.

On-site articles fill the remaining gaps. We publish two a month per business, at 800 words or more, each aimed at a real search gap. Article titles carry a modifier such as cost, signs or how often, so they support the service page instead of competing with it.

What the page list says about the business

Giving each service its own page forces a useful question: does this service deserve a place on the menu? Search Console shows how many people look for it, and the call and text taps in Analytics show which pages turn visits into work. A service nobody searches for and nobody calls about is worth a second look, and a service that draws steady calls may deserve better pricing, more hours or a technician of its own.

The city list works the same way. The towns that send calls show where the service area is working and where the drive time is not paying off. Working with us includes this kind of business advice, so the page structure becomes a way to decide which services to lead with and where to work. Our business growth advice page covers the topics, and business model and service area goes deeper on the service mix and the towns an owner serves.

How to apply this

  1. List every service and every city you serve.Match the city list to the service area on your Google Business Profile.

  2. Assign one search pattern to each page.If two pages share one, merge them or change one pattern.

  3. Build the core set first.Create a page per service, a page per city, an FAQ page and a blog, and add service-plus-city pages only where search data shows demand.

  4. Check the click path.Keep every money page within 2 clicks of the homepage and never beyond 3.

  5. Use real links.Make every link an <a> element with an href and write anchor text that describes the destination. Give every orphan page a place in the structure.

  6. Add 3 to 8 contextual linksin the body of each page, connecting related services, cities and articles.

  7. Deepen priority pages.Answer the nine questions and open each page with a self-contained answer in the first 100 words.

  8. Give every location page proof:a job photo from that city, a review that names it, or a first-hand local fact.

  9. Write unique titles and descriptions,titles at 60 characters or fewer and descriptions at 140 to 160.

  10. Review monthly.Check pages at positions 4 to 15, mismatched click-through rates, competing pages and cities without a page, then use Analytics to see which pages bring contact.

Frequently asked questions

There is no fixed number. The right count is one page per service, one per city served, and a combination page wherever search data shows real demand. A smaller site with focused pages usually outperforms a larger one whose pages repeat each other.

No. Building every combination produces many pages that attract no searches and add nothing a visitor needs. Create one when search data shows demand, and make sure it carries local proof no sibling page could share.

It can be, provided the old addresses are permanently redirected to the new ones. Without redirects, indexed pages stop resolving and visitors reach errors. The safest approach is to avoid renaming live URLs at all.

Yes, a menu link is a real link and crawlers follow it. But menu links only say a page exists. Links in the body of a page, in a sentence about a related idea, tell search engines and visitors why the connection matters.

Search engines need to recrawl and reprocess changed pages, so results arrive gradually rather than on a fixed date. Watching impressions, position and clicks by page in Search Console shows whether a change is being picked up.

Start with the services customers actually call about, then add others as search data shows demand. Page data also tells an owner which services are worth leading with, pricing carefully or dropping from the menu. That is business advice we include when we work with a business.

Closing

At GMScalez we apply these principles to the mobile and service-area businesses we work with, starting from the services each business sells and the cities it serves. Every site is built around one search per page, shallow paths to the pages that bring in calls, and depth where it helps a customer decide. From there the work is tailored to the business and adjusted month by month, using Search Console and Analytics data to decide which page to build, sharpen or merge next. The conversation starts with a free one on one marketing consultation.

Related guides and pages

Call or text us at (712) 251-1967. Email us at garinfroehlich37@gmail.com.

Last updated: November 2026

Sources

  1. Google Search Central, JavaScript SEO basics
  2. Google Search Central, Make your links crawlable
  3. Google Search Central, Spam policies
  4. Botify, Page click depth and ranking
  5. Zyppy, Internal links study
  6. Zyppy, Google title rewrites
  7. Ahrefs, Title tags study
  8. Backlinko, Search engine ranking study
  9. Ahrefs, Short vs. long content in AI Overviews
  10. Google Search Console Help, What are impressions, position and clicks?
  11. Ahrefs, AI Overviews and clicks
CallTextFree consultation