Field notes
A dental referral program that tracks itself
Three fixed moments where the source gets captured, the monthly math per source, and what changes when the count covers nine offices instead of one.
A dental referral program that tracks itself is one where the referral source is captured at three fixed moments, the new patient phone call, the intake form, and the first visit, written into one structured field in the practice management system, and read back once a month as a per-source count: referrals received, first visits completed, treatment accepted, referrers thanked. The gift card is optional. The capture and the monthly read are not. Most programs are built backwards from the reward: the office buys the cards, prints the sign, and six months later nobody can say which patients came from patients, because the question was never asked in a way the system could count.
This page is the capture-and-count side of referrals: where the source gets recorded, the four numbers that get read every month, and why internal referrals and specialty referrals need two different loops. It is written for a single office with one front desk and for a group where the count has to survive nine of them. The gift-card question is at the end, where it belongs.
Why "how did you hear about us?" is not tracking
Almost every office already asks the question. The answer dies in one of three places. It gets asked once, on the phone, and never checked again, so "my sister" turns into "Google" by the time the chart is made. It gets typed into a free-text field, so the month-end count needs a human to decide whether "Instagram", "insta" and "saw you online" are one source or three. Or it gets captured cleanly and read never, which is the most common ending: the data exists, and no meeting looks at it.
A program that tracks itself closes all three holes with rules, not effort:
- One structured source field per patient, picked from a short fixed list, never free text
- The source captured at first contact, then verified twice before the first visit ends
- A monthly read with a named owner and the same formula every month
Where does the referral source get captured?
Three moments, three owners. Each later moment exists to catch what the earlier one missed, because patients tell the scheduler one thing and tell the chair the truth.
| Moment | Owner | The ask | Where it lands |
|---|---|---|---|
| New patient call | Whoever answers the phone | "Who can we thank for sending you in?", asked before the insurance questions, while the caller is still telling their story. | A referral attached to the new patient in the OpenDental Family module, picked from the referral list. |
| Intake forms | Front desk at check-in | A fixed pick list plus one name line: "If a person referred you, who?" | Checked against the phone answer; a mismatch gets resolved at check-in. |
| First visit | The clinical team, relayed at checkout | No script. Patients volunteer the real story in the chair; the assistant passes the name to the desk. | The record corrected the same day, before the thank-you goes out. |
The named person matters more than the channel. "A friend" cannot be thanked and cannot be counted as a repeat referrer, so the fixed list keeps the channels honest and the name line keeps the people countable. When one patient names both a person and a channel, the person wins; write that rule down once, or the attribution argument restarts every month. The same discipline that keeps outreach ROI honest applies here: a count nobody trusts changes nothing.
What does the monthly referral math look like?
Four numbers per source, computed the same way every month. Each needs a formula in words, a place it comes from, and the decision it triggers, the same definition-contract discipline as the dental KPI short list. A number missing any of the three is a decoration.
| Number | Formula in words | Where it comes from | What it decides |
|---|---|---|---|
| Referrals received | New patients this month whose record carries this source. | Family module referral entries, counted by entry date. | Whether the ask is alive. A month of zeros means it died at the desk, not that patients stopped talking. |
| First visits completed | Referred patients whose first completed visit fell in the month. | The same records, counted by first-visit date. | Which sources send patients who actually show up. |
| Treatment accepted | Dollar value of treatment accepted by this source's patients within 90 days of the first visit. | Treatment plans, read against the referred cohort. | Which sources send patients who say yes; use the treatment acceptance rate definition, not a looser one. |
| Referrers thanked | Thank-yous sent this month, against named referrers this month. | The commlog entry made when the thank-you goes out. | Whether the loop closes. A referrer who never hears back has no reason to do it again. |
Illustrative, not a benchmark: an office counts 14 referrals in March, 9 from named patients and 5 from the website. By June the patient-referred nine have completed 8 first visits and accepted $11,200 of treatment; the website five completed 3 visits and accepted $1,900. The office did not learn that websites are bad. It learned that its best marketing channel was sitting in its own chairs, and that the March thank-you list was nine names long and took twenty minutes to work.
The gift card is optional. The capture and the monthly read are not. An office that asks every caller and thanks every named referrer has a referral program, whatever it spends.
How are internal and specialty referrals different?
One word covers two loops that share nothing but the field they are recorded in.
Patient-to-patient referrals
The internal loop runs on named people and small numbers. In most offices a short list of patients produces most of the names; the monthly read tells you who they are, and the thank-you keeps them doing it. The natural moment to invite a referral is the same checkout moment that carries the Google review ask: the patient is happy, and the words are plain. "We are taking new patients, and we would love to see your family." One more rule: the desk should know, before asking, whether this patient has referred before. Asking a three-time referrer whether they have heard of your referral program is how three-time referrers stop.
Specialty referrals, out and in
The specialty loop is practice-to-practice, and the money leaks on the way back. A patient sent out for an extraction is supposed to come back for the restorative work, and the ones who never return are the quietest hole in general-practice production. So track outbound referrals as their own source records: who was sent, to whom, for what, and whether a comeback appointment exists. That comeback list is unscheduled treatment with a referral attached, and it belongs on the daily call list, not in a drawer. Inbound runs the same loop in reverse for the specialist: one source record per referring practice, read monthly, because a referring doctor who goes quiet in April should be a conversation in May, not a mystery in December.
Should you attach a reward?
A thank-you note is always safe and almost always skipped. A reward needs a check before the cards get printed: state dental boards treat referral incentives differently, fee-splitting rules can reach further than offices expect, and patients covered by federal programs change the analysis entirely. That is a short question for a lawyer who knows your state, not a policy to copy from a Facebook group. The operational finding is simpler: a reward multiplies a loop that already works. An office that asks every caller, records every name, and thanks every referrer has a working referral program before it spends a dollar. An office that skips those three steps has a gift-card budget.
Who reads the number across more than one office?
In a single office the owner is the office manager, once a month, on the same page as new patients by source. In a group the count has to exist per office before it is a roll-up, because referral culture is local: one front desk asks every caller, another stopped in March, and the group average hides both. The month this breaks is usually the month the count moves into a spreadsheet that one person updates by hand.
That per-office accountability is the shape MyDentalForce is built around. Managers see each office's activity and results without asking, the Referral Suite runs in-house referral capture on the same per-office footing, and New Patients & Marketing ROI puts referral counts next to every other source of new patients, office by office. The daily call list each office works is rebuilt every morning from that office's own OpenDental data, due and overdue hygiene recall, unscheduled treatment, lapsed patients, which is exactly where the specialty comeback list lands, and every call outcome is logged and written back to the practice management system. The referral count stops being a spreadsheet and becomes a number somebody reads.
Start with the field, not the budget. Clean up the source list this week, brief the morning huddle on the one-line ask, and run one full month of capture before deciding anything about rewards. A month of honest counts will tell you who your referrers are and what they are worth. When keeping that count honest by hand gets expensive, across one office or nine, automate the counting, not the gratitude.