Everything We Ship, In the Order It Has to Ship

Before anything is proposed, we establish what is actually true about your visibility, your site, and your market.

• Technical and content audit. Crawl access, indexation, speed, structured data, and the state of every page that currently earns anything

• Local surface read. Google Business Profile completeness, category strategy, review volume and recency, citation consistency, and geo-grid rank across your real catchment rather than one check from your own office

• AI visibility baseline. Whether ChatGPT, Gemini, Perplexity, and AI Overviews name your practice today, and who they name instead

• Competitive gap analysis. The treatments, questions, and neighbourhoods your competitors answer and you do not

• Compliance baseline. Tracking on intake pages, accessibility failures, claim language, and any state board disclosure your market requires

A plan with named priorities and named numbers, agreed before work starts. Not a package with your logo on it.

• Priority sequence. What gets fixed first, second, and third, ordered by what changes patient volume rather than what is easiest to report on

• KPI agreement. The numbers we will be judged on, defined before we start so nobody redefines success in month six

• Service and market scope. Which treatments and which neighbourhoods the campaign targets, based on verified demand rather than a keyword list

• What we are not doing. Named explicitly, so the plan has edges and the budget is not spread across everything at once

A custom site engineered around the booking path, with speed, accessibility, and compliance as build constraints rather than later fixes.

• Custom architecture. Server-rendered so your content is in the HTML on first fetch, which is the only version AI crawlers read

• Booking as the spine. Online scheduling wired to your practice management system, click to call on every screen, and a form for patients who will not do either

• Treatment page system. One page per treatment, structured to answer cost, candidacy, downtime, and what happens on the day

• Provider pages. Named clinicians with credentials, certifications, and the specialty detail that decides whether a page is trusted

• Accessibility built in. WCAG 2.2 AA in the markup itself. No overlay widget, which plaintiff firms now treat as a signal worth investigating rather than a defence

• HIPAA-aware tracking. Advertising tags kept off intake, booking, and portal pages, consent gating everywhere else, and conversion measurement that does not depend on them

The machine-readable version of your practice, generated from one source of truth so it cannot drift.

• Clinical business types. MedicalClinic, Physician, Dentist, and the specialty subtypes rather than a generic LocalBusiness, so Google knows which category of near me search you belong in

• Complete location data. Address, coordinates, opening hours, telephone, service area, and accepted insurance where it applies, on every location

• Provider entities. Each clinician as a Person or Physician node with credentials, specialty, and links to their professional profiles

• Services and procedures. Your treatment list as structured offerings rather than prose a machine has to interpret

• One source of truth. Every schema field generated from the same data as the visible page, so the two can never disagree

Being the source an AI engine quotes, which is a different job from ranking and is increasingly the one that matters first.

• Crawler access, agent by agent. The AI citation agents explicitly allowed, the training agents set to whatever you decide. Most sites conflate the two and lose the citations while trying to protect the content

• Answer-first structure. Every treatment and condition page opens with the citable factual answer, then earns the rest of the read

• Content in the initial HTML. AI crawlers do not execute JavaScript. If the answer arrives after hydration, it does not exist to them

• Question and answer markup. The cost, downtime, pain, and candidacy questions patients actually ask, marked up and visible on the page rather than hidden behind a script

• Corroboration. The same facts stated consistently across your site, your profile, and your listings, because agreement across sources is what an engine treats as confidence

The searches people speak rather than type, which are longer, more specific, and much closer to booking.

• Speakable markup. The passages that make sense read aloud, marked as such

• Question-shaped headings. Content structured around how much, does it hurt, how long, and am I a candidate

• Open now and near me. Hours in structured data so an assistant can answer whether you are open, and neighbourhood language in plain sentences

• One-tap calling. Every number a tel link, because a voice search that ends in a memorised phone number ends

The three results above the fold on a phone. For a practice patients have to physically reach, this is the most valuable real estate in search.

• Profile optimization. Primary and secondary categories, services, attributes, booking link, and the storefront or service-area configuration that matches how you actually operate

• Site to profile parity. Name, address, phone, and hours verified identical across your site, your schema, and your profile. Drift here is the most common cause of a stalled map pack

• Citation consistency. The directories that matter for your vertical, corrected and kept consistent

• Review systems. Compliant velocity with no gating and no incentives, HIPAA-safe response templates that never confirm someone is a patient, and a recovery process for legitimate negatives

• Geo-grid measurement. Rank across a grid of points around your practice, so you can see which neighbourhoods you are invisible in rather than one flattering number

Treatment and condition content written to answer completely, attributed to a named clinician, and safe to publish.

• Pillar and cluster architecture. A page per treatment, supported by the condition and question pages that feed it, linked programmatically rather than by hand

• Named clinical authorship. An author and, where it applies, a medical reviewer. Anonymous health content is the weakest form there is

• Claim-safe by construction. Claims narrowed to what your evidence supports rather than broad claims propped up by a disclaimer, which is the approach the FTC actually asks for

• Results imagery done properly. Consistent lighting and angles, advertising-specific written consent per image, and typical-result disclosure rather than results may vary

• Freshness cycles. Scheduled re-verification, because a page that has not been reviewed in two years is a liability in a medical vertical

Traffic that does not book is a cost. This phase is about the distance between arriving and being on the schedule.

• Booking path engineering. Taps from landing to booked, measured and reduced. Every additional step is a share of patients who do not finish

• Speed to lead. Form and call routing so an enquiry reaches a person fast, because the practice that answers first usually wins the patient

• Objection handling on the page. Cost, financing, downtime, and what happens on the day, answered before the consult rather than during it

• Call tracking and attribution. Calls attributed to channel so cost per booked patient is a measured number rather than an estimate

• Testing. Structured tests on the paths that carry the most patients, with revenue per visitor as the measure rather than bounce rate

The campaign is not finished at launch. It is finished when the numbers move, and the numbers are yours.

• Booked appointments by channel. Reported monthly in your numbers. Not impressions, not lead counts, not a dashboard screenshot

• Geo-grid heatmaps. Map pack position across your catchment, tracked window over window

• Search Console and rank tracking. Clicks, impressions, and page-level movement, with competitor overlays

• Profile actions. Calls, direction requests, and booking clicks originating from your Google Business Profile

• AI visibility radar. Mentions and citations across the assistants your patients now ask first, tracked as they move

• Scheduled checkpoints. Reviews at 30, 60, and 90 days against the KPIs agreed in phase 01, not against goalposts moved afterwards