Tier keywords by intent, not by volume
The standard mistake is optimizing for volume. In this category it is particularly costly, because the highest-volume terms are informational health queries dominated by publishers with authority you will not match — and that traffic would not buy from you anyway.
The terms worth owning are commercial and transactional. Lower volume, dramatically higher value, and frequently under-optimized because everyone else chased the big numbers.
| Tier | Intent | Query shape | Page type | Build order |
|---|---|---|---|---|
| Transactional | Ready to act | buy X, X pricing, X near me | Commercial service page | First |
| Commercial investigation | Comparing | best X, X vs Y, X cost | Comparison or cost page | First |
| Solution aware | Knows the problem | how to X, X checklist | Guide | Second |
| Problem aware | Researching | what is X, X explained | Educational article | Third |
| Brand | Looking for you | your name, your products | Home, product pages | Always |
One page, one intent
Every page needs a single primary intent and a canonical URL. Two pages targeting the same intent compete with each other, split their signals, and both underperform what one good page would have achieved. Keyword cannibalization is the most common problem we find in audits, and it is almost always the result of publishing without a keyword map.
Build the map before you write anything. A spreadsheet with one row per page: URL, primary keyword, intent tier, and the pages it links to and from. It takes an afternoon and it prevents a year of accidental self-competition.
Commercial pages get the most depth. A page targeting a transactional query should be genuinely substantial — the specific information a buyer needs, not a thin page with a keyword swapped in. Thin commercial pages are the fastest way to look like every low-effort competitor.
Technical foundations, in priority order
- Crawlability and indexation. A sitemap with honest modification dates, robots exclusions for admin, API, and personalized result pages, and noindex on anything private. Getting this wrong wastes crawl budget on pages that should never rank.
- Canonicals and duplication. One canonical per page, self-referencing. Parameterized and paginated URLs handled deliberately rather than by accident.
- Core Web Vitals. LCP under 2.5s, INP under 200ms, CLS under 0.1. Not because it is a large ranking factor — it is not — but because a slow page loses the visitor before the ranking matters.
- Structured data that reflects reality. Organization, Service, Offer, Article, and Breadcrumb markup describing what is genuinely on the page. Marking up things that are not there is a manual-action risk, not a shortcut.
- Internal linking. The most underused lever in nearly every site we audit, and it costs nothing. Every supporting page should link to the commercial page it supports, with a descriptive anchor.
- Metadata written for the click. Unique titles and descriptions that match what the page delivers. Overselling produces clicks that bounce, which is worse than a lower click rate.
E-E-A-T in a scrutinized category
Health-adjacent queries fall under heightened quality evaluation. Practically, anonymous content with no named author, no credentials, and no clear organization behind it struggles regardless of technical quality. This is not something technical SEO can compensate for.
The fixes are organizational rather than technical, which is why they get skipped. Put a real named author on content. Name the reviewer where clinical accuracy matters. Publish who actually runs the business, with real detail. Make contact information genuine and findable. Use accurate publish and modified dates rather than cosmetically refreshing them.
Then keep every claim inside the line. A page that overclaims is a page quality raters mark down — a separate problem from any regulatory question, and one that shows up in rankings.
- Named authors with substantive bios, not “Admin”
- Named reviewers wherever accuracy requires a credential
- An About page that describes actual people and actual operations
- Real, verifiable contact details in the footer of every page
- Accurate dates — a modified date should mean something changed
- Citations to primary sources rather than to other marketing pages
- Policy pages linked sitewide and genuinely reviewed
Local SEO for clinics and practices
For a single-location practice, local search usually returns more than national content, and it returns faster. The work is unexciting and mostly mechanical.
Claim and complete the Google Business Profile with accurate categories, hours, and services. Build a location page with genuine local content — not a template with the city name swapped. Ensure name, address, and phone are byte-identical across every directory listing. Establish a process for requesting reviews from satisfied clients, and respond to all of them.
The multi-location version adds a location page per site with distinct content, and a decision about whether locations should rank independently or funnel to a central page. That decision should be made deliberately, not discovered after six months.
Tactics to avoid, and why
The uncomfortable truth about the tactics below is that some of them work for a while. They also produce a business whose entire acquisition channel can vanish in a single update — and in a category where you cannot fall back on paid, that is not a risk worth carrying.
- Purchased links, private blog networks, and link exchanges. Manual-action risk lands on your domain, not on whoever suggested it.
- Guest posting on sites that exist only to host guest posts. The links carry no value and the association carries risk.
- Bulk AI-generated content published without human writing and editing. It is detectable, it is thin, and it dilutes the pages that are good.
- Pages built to rank that offer nothing to the person who lands on them. Every one of these raises your site-wide quality problem.
- Structured data for content that is not on the page. This is misrepresentation with a schema validator attached.
- Cloaking, doorway pages, and expired-domain redirects. These are not grey areas.
What to expect, and when
| Month | Realistic expectation |
|---|---|
| 1 | Technical fixes live. Indexation improves. Little traffic change. |
| 2 | First commercial pages indexed. Long-tail impressions begin. |
| 3 | Meaningful impression growth. Occasional first-page positions on low-competition terms. |
| 4–6 | Commercial pages start ranking. First organic inquiries. |
| 7–12 | Compounding. Pages published in month two are now mature and pulling steadily. |
Anyone promising faster is either describing branded traffic you already had, or using tactics that carry risk you will inherit.
Search compounds. Start the clock.
Technical SEO is included in every build. Content SEO comes as a pack or an ongoing plan.