62% of patients have avoided a provider over a review — and replying to one is a compliance decision

45% of patients say they regularly read reviews before booking, 62% have avoided a provider because of what they read, and 34% say they would never book a provider rated three stars or below. That last figure is the one that changes a media plan, because it is not a preference, it is a filter. A practice sitting below the threshold is paying the published $40.04 per lead to send people to a profile that a third of them will reject before the page finishes loading. Meanwhile the single largest published lever, responding to a negative review, is the one action in this vertical that can itself constitute a disclosure.

A public reply that thanks someone for visiting the practice has just confirmed in writing that they are a patient. The apology is the disclosure.

1. 01

   The reply is the regulated part, and most templates fail on the first line

   Confirming, denying or contextualising a reviewer's care in public is a disclosure regardless of how sympathetic the wording is. Compliant templates acknowledge the feedback, state the practice's general standard, and move the specifics to a private channel, without ever establishing that the reviewer was there.

2. 02

   A rating floor is a hard filter, not a soft preference

   34% would never book at three stars or below. Below that line, additional media buys impressions for a profile a third of the audience has already excluded. Reputation work is therefore sequenced before a media increase, not alongside it.

3. 03

   The audience is concentrated on one platform but not contained by it

   83% of patients use Google reviews, 38% Healthgrades, 33% Yelp and 24% Zocdoc. A Google-only programme manages the largest share and leaves the specialty directories, which is exactly where a patient goes once they are comparing two named providers rather than searching a category.

4. 04

   Responding is the largest published effect available and it is free

   71% say they are more likely to trust a provider who responds to negative reviews. There is no media buy in this vertical with a published effect that size, and the whole cost of it is a template, an owner and a response window.

5. 05

   The general local-review research points the same way; the compliance layer is what is specific

   Review-reading behaviour is not unique to healthcare and the general local-consumer research describes the same pattern across categories. What does not transfer is the response playbook, because in every other vertical the reply is a marketing asset and here it is a regulated communication.

Patients who've avoided a provider due to negative reviews **62%** (Tebra 2025); 34% would never book a provider rated ≤3★; 71% trust providers more when they respond.

The model runs reputation as an operating loop rather than a campaign. Review requests are sent through a non-clinical trigger — an appointment completed, never a treatment performed — so the request itself carries nothing. Incoming reviews are triaged into two lanes: service complaints, which get a compliant public reply and a private route, and anything touching clinical care, which gets an acknowledgement and an immediate offline handoff. Templates are written once, reviewed by the practice's own compliance contact, and never improvised. Ratings and volume are tracked per platform, and media spend is reviewed against the rating rather than held constant while it recovers.

1. 1

   Weeks 1–2

   Profile and rating audit

   Rating and review volume per platform, gaps in Healthgrades, Yelp and Zocdoc presence

   Owner: OmniFlow

2. 2

   Weeks 2–4

   Template build and compliance review

   Reply templates for each lane, reviewed and signed off before any reply is posted

   Owner: OmniFlow + practice

3. 3

   Weeks 4–8

   Request cadence

   Non-clinical trigger, fixed cadence, opt-out honoured, no incentive of any kind

   Owner: Practice + OmniFlow

4. 4

   Week 5 onward

   Response operations

   Named owner, defined response window, private route for anything clinical

   Owner: Practice

5. 5

   Month 3 onward

   Media alignment

   Spend reviewed against rating per location and per platform

   Owner: OmniFlow

The model reports rating and review volume per platform, response rate, and time to response, and it reads media efficiency against those figures rather than separately from them. The claim it does not make is a conversion lift from a rating change: no published study establishes that relationship for medical practices specifically, and the practice's own booking data is the only place it can be established. Nothing in the reporting quotes review text, names a treatment, or ties a count of patients to a condition, and that constraint applies to internal reporting as strictly as it applies to anything published. Media efficiency is read against rating and review volume per platform each month, so a spend decision and a reputation decision are made in the same review rather than by two people looking at two dashboards.

Modelled. Inputs: Tebra, The Intake research report 2025 (1,006 patients and providers); BrightLocal, Local Consumer Review Survey 2026 (1,002 US consumers); WordStream by LocaliQ, Google Ads Benchmarks 2026 (13,474 US search campaigns, April 2025 – March 2026). Modelled outputs are not a forecast or a guarantee of results. Cited: Tebra *State of Online Healthcare Reviews* (Sept 2025).

Every figure below is published research, not a client result. They are the inputs to the arithmetic above, listed so it can be checked rather than taken on trust. The bracketed number points to the full citation at the end of this page.

These are the surfaces this engagement is run and measured from, shown with representative figures built around the benchmarks cited on this page. Every account we run reports into views like these, and you keep ownership of all of them.

Each figure on this page, the system it is read from, and the definition and window it is measured over.

Not applicable — this is a modelled engagement. Its weakest input is that the Tebra research surveys 1,006 patients and providers on stated behaviour, and stated behaviour around reviews consistently overstates what people actually do at the moment of booking. The model therefore treats the 34% rating floor as a directional threshold rather than a measured cutoff, and instruments the practice's own rating-to-booking relationship from month two.

A modelled capability, not a client account. Figures illustrate what the model produces and are labelled as modelled wherever they appear. See all our work

Full citations for everything cited on this page, with the sample and period each study covers, so you can go and read the original.