Why original data gets cited and opinions do not
A page with something no other page has, such as your measured wait for a new patient visit or the questions your phones get most, gives search engines and AI assistants a reason to point to you rather than to pages repeating common knowledge. Google asks directly whether content offers original information or analysis. Nobody outside the AI companies can prove what they cite.
What Google says about originality
Google's guidance on helpful content opens its list of self assessment questions with originality: does the content provide original information, reporting, research or analysis? Does it provide substantial value compared with other pages in search results? Its ranking systems guide also describes original content systems, which help ensure Google shows original content prominently.
Its 2026 guide to generative AI features goes further. It contrasts commodity content, based on common knowledge, with non commodity content that offers unique expert or experienced takes, and says creating content people find unique, compelling and useful will likely influence your presence in generative AI search more than any of its other suggestions.
That is as close as Google comes to saying what it wants. For a practice, original usually means information that comes from your own patients, schedule and clinicians.
Commodity content versus what only you know
Most practice websites are full of commodity content. "Seven signs you need to see a dermatologist" or "What is physical therapy?" can be found on thousands of sites, and an AI assistant can write its own version in seconds. Nothing about it points to your practice.
Now compare statements only you can make, shown here as hypothetical examples:
- "New patients at our clinic are usually seen within two weeks; in the last quarter the median wait was nine days."
- "The question our front desk hears most in spring is whether allergy shots can start mid season. Here is our answer."
- "Our surgeons performed over 400 of these procedures last year. Here is what recovery has looked like for most of them."
Each is checkable, specific and tied to you. If true, each gives a patient, a journalist or a searching system a reason to cite your page rather than any other.
Does ChatGPT recommend your practice?
The Surge Score measures five things about your practice website: how often ChatGPT, Gemini and Perplexity recommend you, your PageSpeed, your structured data, your indexing and your reviews, each with its evidence and date. Free.
Data most practices already have
You do not need a research department. Most practices already hold useful, non identifying information:
- Access: time to the next new patient appointment, by specialty or location
- Volume: how many of a given procedure your team performs each year, if you can state it accurately
- Questions: the most common questions callers ask, from phone logs or front desk notes
- Seasonality: when certain visit types peak at your practice, such as sports physicals or allergy visits
- Process: how long results usually take to come back, how long a first visit lasts
- Cost: typical self pay prices for common services, if you publish them
Choose numbers that help a patient decide, not ones that only flatter the practice.
Publish it so it can be trusted
A number without context is a claim. A number with its method is evidence. For each figure you publish, say:
- What was measured, and how
- The time period, and when you last updated it
- How many patients, visits or calls it covers
- Any limits, such as one location only, or excluding urgent visits
Publish only aggregate figures with no patient details, and have your privacy or compliance lead review anything drawn from patient records before it goes live. Outcome claims raise more questions than access or volume figures, because advertising and medical board rules apply to them; see health claims rules before publishing any.
Finally, keep it current. A wait time from two years ago that is now wrong is worse than no number at all. Our own methodology page shows the kind of disclosure we mean.
Opinion is not worthless; generic opinion is
The title of this guide needs a qualification. Clinical judgment, explained with reasons, is valuable. Google's Search Quality Rater Guidelines treat first hand experience and expertise as core parts of E-E-A-T, and its AI guide describes first hand perspectives as more useful than summaries of existing content.
What fails is opinion without anything behind it: "We believe every patient deserves the best care." What works is reasoning from experience, as in this hypothetical orthopedic example: "We start most patients with knee arthritis on physical therapy for six weeks before discussing injections, because in our experience many improve enough to avoid them. Here is how we decide." That is a view only your practice can offer, with a basis a reader can weigh. Our guide on physician reviewed content and E-E-A-T covers presenting that expertise.
What we cannot promise
It would be easy to claim that publishing data guarantees AI citations. No one outside the AI companies can show that. Their documentation says nothing about preferring pages with data, and their answers vary from one run to the next.
What can be said honestly: Google says originality matters to its systems, including its AI features; pages that repeat common knowledge give any system no particular reason to choose yours; and specific, checkable facts are what patients use to choose a practice. Those reasons are enough to do it. Present your figures clearly, near the top of the relevant page, in plain sentences, as described in answer first content.
Checklist
- Each key page includes at least one fact only your practice could state
- Every published number says what was measured, when and over how many cases
- Only aggregate figures are published, with no patient details
- Your compliance lead reviews any figure drawn from patient records
- Outcome claims are checked against advertising and medical board rules
- Every figure has an update date and an owner who keeps it current
Frequently asked
Is it safe to publish numbers from our practice data?
Aggregate figures with no patient details, such as median wait times or annual procedure counts, are generally low risk, but have your privacy or compliance lead review them first. Avoid small numbers that could identify individuals, and be careful with outcome claims.
What if our numbers are not impressive?
Publish the ones that help patients decide, and be honest. A realistic wait time builds more trust than a vague promise. If a number would hurt you, that is useful information for improving the practice, and you are not obliged to publish it.
Do AI assistants really cite practice websites?
Assistants that search the web show links to their sources, and those can include practice websites. Which pages they choose is not published. Original, specific content is the most defensible bet, because it is the one thing other pages cannot copy.
Sources
- Creating helpful, reliable, people-first content, Google Search Central
- A guide to Google Search ranking systems, Google Search Central
- Optimizing your website for generative AI features on Google Search, Google Search Central
- Search Quality Evaluator Guidelines, Google
Book a strategy call with the team.
Twenty minutes. We'll walk through the specific opportunities in your market and what a MedPracticeGrowth engagement would look like for your practice.