Most practice owners we talk to have been burned by a marketing agency before. Sometimes more than once. The pattern is recognizable: a long contract, a slick monthly report, vanity metrics dressed as wins, and a calendar that stays empty. The owner cancels at the contract end, picks a different agency, and the pattern repeats.
The pattern is fixable. The fix is in the questions you ask before you sign. Below is the framework we recommend to any practice owner evaluating an agency, including practices that are not evaluating us.
Question one: what is the only metric that matters in your reporting?
If the answer includes the words "impressions," "clicks," "engagement," or "reach," walk away. If the answer is "booked consults" or "new patient appointments," continue. The agencies that get paid on activity report on activity. The agencies that get paid on outcomes report on outcomes. The first one always sounds smarter. The second one actually moves your calendar.
Question two: how do you measure that?
The follow-up matters more than the initial answer. Anyone can say they care about booked consults. Few can explain how they measure them. A real answer involves either a conversion pixel that fires when the booking confirms in your booking system, or a weekly export from the booking platform matched back to ad clicks by phone or email. If the answer is "we track form submissions," they are measuring a proxy, not the real number.
Question three: what is your guarantee?
If the answer is "we do not offer guarantees," the answer is honest but it tells you the agency is selling effort, not results. If the answer is a money-back guarantee on impressions or clicks, it is a guarantee on activity. The guarantees that matter are on outcomes. We offer 10 booked patient consults in 90 days, or we keep working free, on the Patient Flow System. We covered the mechanics in the article on the 10-consult guarantee.
Some agencies cannot offer outcome guarantees because the work they do does not produce reliable outcomes. They are not lying. They are correctly assessing their own work. The question for you is whether that agency is the right one to hire when other options exist with stronger guarantees.
Question four: who is on the call?
Ask who you will be talking to during the engagement. The answer should name actual people, not a tier or a department. "The strategist who handles the account will be on the kickoff and every monthly review" is a better answer than "Your dedicated success manager will coordinate with the strategy team." The first answer means the person making the decisions is in the room. The second means a layer of customer-service translation is in the middle.
A 1 to 15 provider practice rarely needs a six-person agency team. Usually one strategist plus one or two specialists is enough. Larger agency teams add overhead, slow decisions, and dilute the relationship without adding much value.
Question five: what happens in the first 30 days?
Ask for a specific 30-day plan before you sign. Not a long-form proposal. A short plan with three to five deliverables and a check-in point. A real plan looks like: "Week 1: Discovery + audit. Week 2: Site rebuild draft + ad account setup. Week 3: Photo shoot + ad campaign launch. Week 4: First conversion review." Agencies that cannot describe the first 30 days specifically are agencies that improvise. Improvisation in the first 30 days produces no booked patients.
The reference check that most owners skip
Ask for three references from current clients. Call them. Ask the references one question: "How many new patients did you book last month, and is that more or fewer than the month before you started with this agency?" A real, current reference will give you a specific number. A vague answer means the work is not producing clear results.
If the references are uniformly enthusiastic with no specifics, the agency cherry-picked them. Ask for one more reference from a client that finished an engagement without renewing. Most agencies cannot produce one. The ones that can are the ones that handle hard conversations well.