Dental Lead Tracking: Build a Pipeline the Clinic and Agency Both Trust

Write a one-page measurement agreement

An unverified Reddit account expresses distrust of marketing reporting and uncertainty about appointment results. That is an individual concern, not proof of agency misconduct. The constructive response is to make the counting rules visible to everyone involved.

Define the reporting period, which sources are included, how duplicates are handled and when a record changes stage. Name the source of truth for each event. An ad platform may report a submission, while reception confirms booking and the clinic records attendance. Do not let one system imply knowledge it does not have.

Treat stages as events with evidence

Use a short stage list that the clinic can maintain reliably. The following is a proposed internal model, not a claim about an Allie integration. Add complexity only when it supports a real operational decision.

• Received: the unique new inquiry reached the clinic, with a receipt time.
• Reached: a two-way conversation occurred, with a recorded time.
• Ready for reception: enough nonclinical routing context is available.
• Accepted by reception: a named human owns the next action.
• Booked: the appointment is confirmed by the scheduling team.
• Attended: attendance is recorded by the clinic.
• Closed or pending: the known reason or outstanding next step is stated.

Keep source, status and reason separate

Source describes where the inquiry came from. Status describes where it is now. A reason explains a known obstacle or closure. Mixing these fields creates misleading categories such as “Facebook bad lead,” which combines acquisition channel with an unsupported judgment.

Use factual reasons such as service unavailable, timing mismatch, duplicate or person declined. Keep “not reached” distinct from “not interested.” Allow unknown. If a person does not reply, the record does not establish why.

Retain the original inquiry cohort for downstream review. A booking that happens later should remain connected to that inquiry rather than being reassigned to whichever campaign ran during the booking month. Note attribution uncertainty instead of trying to eliminate it with guesswork.

Reconcile exceptions together

Choose a regular review that fits the inquiry volume. Compare aggregate totals, then investigate discrepancies inside the clinic’s approved systems. Look for duplicate submissions, late bookings, missing handoff acceptance and records left in an outdated status.

For example, a hypothetical clinic and agency disagree by three inquiries. The review finds two duplicates and one existing-patient request. The right outcome is a documented reconciliation, not retroactively changing a definition to preserve a headline number. This example is illustrative, not a typical discrepancy rate.

Assign responsibility for corrections. If no one owns maintaining a field, remove it from the headline report until a reliable process exists. Reporting should expose uncertainty rather than reward invented completeness.

Keep patient information out of a casual reporting workflow

Begin with an access-controlled internal pipeline and an appropriately limited aggregate view for marketing discussions. This article does not recommend sending patient-level records or treatment information to advertising platforms. Any external sharing needs separate review and authorization.

Use platform terminology carefully, but do not confuse a platform label with evidence that the clinic reached or treated someone. A pipeline is useful when its stages mean what the team says they mean.

Allie focuses on new treatment-form follow-up, qualification and warm transfer to reception. Join the waitlist if the unowned gap between inquiry and human handoff is the part of your pipeline that needs attention.

Sources

• Marketing
• Google Ads: qualified leads and converted leads
• Google Ads customer data policies