Of everything that influences how a clinic site performs in search, the internal link graph is the input most fully under the site owner's control and the one most often left to chance. A theme is installed, a navigation menu is populated, pages are added over three years by four different people, and the resulting structure is an accident that nobody designed.
Two separate jobs are being done by internal links, and it is worth separating them because they have different requirements.
Job one: reachability
A crawler discovers URLs by following links in HTML. If a page is linked only from a menu that is built in the browser, or only from a search results page, or only from a paginated archive twelve pages deep, it is hard to reach and will be crawled rarely or not at all. The practical standard for a clinic site is that every page you want indexed should be reachable from the homepage in three clicks or fewer, through links present in the server-rendered HTML.
Test this by crawling your own site with any crawler that can be set to ignore JavaScript, then comparing the URL list against your sitemap. Pages in the sitemap that the crawl never reached are orphans. On clinic sites the usual orphans are old treatment pages, location pages added for a second site that never got a menu entry, and blog posts published before a category structure existed.
Job two: describing what a page is about
An internal link is also a statement. The anchor text and the surrounding sentence tell the system what the destination page covers, and because you write both, it is a rare case of being able to describe your own page from the outside. "Click here" and "read more" waste that entirely. A link that reads "how consultation records affect follow-up scheduling" tells a system exactly what it will find.
The most valuable internal links are in-body links inside prose, not links in a navigation template. Template links appear on every page and carry correspondingly little signal. A link placed in the third paragraph of a relevant article is a deliberate editorial statement about relatedness.
The hub structure that suits clinics
Clinic sites usually have three natural page families: treatment or service pages, location pages if there is more than one site, and informational articles. The structure that works reliably is a hub per treatment area, with the hub page covering the area broadly and linking down to the specific pages, and each specific page linking back up to the hub and sideways to its two or three closest relatives.
| Level | Page type | Links out to |
|---|---|---|
| 1 | Homepage | Every hub, the primary location page, the main service pages |
| 2 | Treatment area hub | Every page in its area, the relevant articles |
| 3 | Specific treatment page | Its hub, two or three sibling pages, the booking route |
| 3 | Article | The relevant hub, one or two specific pages, one or two other articles |
This is not complicated and it does not need a plugin. It needs someone to decide what belongs where and then to enforce it when new pages are added, which is the part that fails.
Finding the orphans you already have
Three checks find nearly all of them. First, crawl and diff against the sitemap. Second, pull the Search Console Pages report and look for URLs with a discovered status, since they are often reachable only from the sitemap. Third, list every page created in the last two years and check by hand that at least one in-body link points to each. That third check is tedious and it always finds something.
Your own website
HIGH CONTROL- Every byte a crawler receives, and the speed it arrives at
- The entity claims: name, address, identifiers, sameAs, service list
- Canonical URLs, hreflang if used, and the internal link graph
- Whether the page answers the question it was built to answer
- How the answer is displayed once it leaves your server
- Whether a third party republishes an outdated version of your facts
- Crawl the whole site and compare the URL list against the sitemap
- Validate every JSON-LD block against the Rich Results Test
- Measure field data, not only lab scores, for Core Web Vitals
Depth, pagination and archives
Blog archives paginate, and by page seven the links are effectively out of reach. If older articles still have value, surface them from somewhere other than pagination: a curated index page, a related articles block driven by topic rather than by date, or links from newer articles that build on them. If they no longer have value, consolidate them into a better page and redirect.
Faceted navigation is the opposite problem. Filters that generate URLs create very large numbers of near-identical pages, which consumes crawl capacity and produces duplication. On a clinic site the answer is usually to make filters operate without changing the URL, or to block the parameter patterns from crawling while keeping the unfiltered pages fully reachable.
Anchor text without over-engineering
Vary the phrasing naturally. Using the identical anchor phrase for every link to a page reads poorly to humans and provides no more information to a machine than three natural variants would. The useful discipline is that the anchor should describe the destination accurately enough that a reader could decide whether to follow it without reading the sentence around it.
The navigation menu is not the architecture
A dropdown listing thirty treatments is a usability problem and a link-equity problem at once. It tells the system that all thirty pages are equally important, which is never true. Keep the menu to the areas rather than the individual pages, and let the hub pages do the distribution. This also makes the mobile menu usable, which matters more than the link structure on a site where the majority of clinic traffic arrives on a phone.
Breadcrumbs earn their place
Breadcrumb trails do three useful things: they give every page an upward link, they express
hierarchy explicitly, and with BreadcrumbList markup they can change how the URL is
displayed in results. They are cheap to implement and they solve the orphan problem structurally
rather than by inspection, because every page gains a parent by construction.
A quarterly link review
Once a quarter, take the ten pages that matter most commercially and count the in-body internal links pointing at each. If a page you consider critical has one, and a page you consider marginal has fourteen, your site is arguing against you. Redistribute deliberately: add links from relevant articles, from the hub, and from sibling pages, and remove links that exist only because a template generated them.
This work is unglamorous and it compounds. It costs nothing, it requires no external approval, and on most clinic sites it produces more movement per hour spent than any other single activity.