Before you blame the leads, find the broken link.

Use three separate questions instead of one quality score

Ask whether the request is relevant, whether the team can act on it and what happened next. Those questions describe different parts of the journey. Combining them into one score can make a promising enquiry look worthless simply because the appointment has not happened yet.

Relevance concerns the public invitation and administrative fit: the requested service, the location and the action the person intended to take. Actionability concerns whether the team has enough usable information and an appropriate route to respond. Progression records the actual conversation, appointment and later outcome. A record can be relevant and actionable while its progression remains open.

Agree who can determine each fact. The campaign owner can check whether the ad named the correct location. The front desk can establish whether the contact route works and whether an appointment was arranged. The clinician owns clinical assessment and recommendations. A marketing review should not infer clinical suitability from a form submission or use a care decision to judge a person’s worth.

Do not use income, homeownership, appearance or an assumed attitude as a shortcut for quality. These guesses fail to describe the actual service request and can send the team toward the wrong repair. A person asking a reasonable price question is not evidence of a poor enquiry. It may simply mean the page has not explained the appointment clearly enough.

The strongest working definition is one the team can apply to the same record and reach the same conclusion. If two people cannot agree, inspect the definition before treating the reported quality rate as a business fact.

Count the open work before counting the loss

A stage report needs a place for people who are still progressing. Someone may be waiting for an answer, considering an appointment, arranging transport or holding a confirmed time next week. The practice does not need to guess their private reasons. It needs to record the known status accurately.

For each acquisition group, show the current stage and whether the next action belongs to the practice, the person or an unresolved investigation. This makes the report operational. “Awaiting practice response” creates a task. “Awaiting the person’s reply” calls for the agreed follow-up approach. “Reason unknown” creates a measurement gap. None means the same thing as a confirmed service mismatch.

Keep cancellations and rescheduling distinct where the records allow. A person moving an appointment is still engaged in an arrangement. If the team reports the cancelled date as a permanent loss while ignoring the replacement, it understates progression and creates unnecessary pressure to “recover” someone who already has a visit.

Use counts beside rates. Three unresolved records in a group of ten deserve attention, but they do not establish a stable population-wide pattern. A small sample can identify a concrete broken link. It cannot automatically tell you how often that link fails across every campaign or month.

Distinguish the wrong request from the wrong expectation

A wrong-service request and a misunderstood invitation need different corrections. If a practice offers chiropractic appointments and repeatedly receives requests for an unrelated service, inspect the creative and destination for ambiguity. If people want the advertised service but misunderstand its price or appointment process, the problem may be missing decision detail.

Compare what the person actually asked with what a reasonable reader could infer. Do not use the campaign team’s intention as the only standard. A large “free” headline with a small qualification can create a broader expectation than the team meant. A photograph of one location can cause confusion when the form sends people to another. A button labelled “confirm” can mislead when no appointment exists.

Look for contrary examples. If most people understood the invitation and two did not, investigate the two records before rebuilding the campaign. They may have arrived through an older link or seen a different post. A clear record of the source prevents a repair to the wrong asset.

When the public wording is accurate but the request is outside the practice’s offering, a respectful answer can close the conversation. You do not need to persuade every person to become a patient. A clear explanation of what the practice can help arrange protects their time and the team’s attention.

Read the first useful answer, not just the first timestamp

A fast acknowledgement establishes receipt. A useful answer advances the person’s actual question. Keep both visible when reviewing response. A system can record an immediate message while the practical question remains unanswered for days. That is a coverage issue hidden behind a reassuring speed metric.

Read the message itself. Does it identify the practice? Does it recognise the request? Does it answer what can be answered administratively? Does it give a practical next step without pretending to provide clinical advice? If a clinician needs to answer, is the handoff clear and owned?

Check coverage across the times enquiries actually arrive. A page may invite people to contact the practice during an evening when nobody is monitoring that route. The answer is not necessarily to have the owner respond all night. Set accurate expectations, assign the next staffed review and keep the waiting request visible.

When a person asks to stop or declines the next step, follow the practice’s approved process and record the known outcome. Repeated contact is not proof of care. Intimacy in the R.O.I. Method is a humanly attentive relationship: someone notices what was asked, answers appropriately and respects the response.

Finally, inspect availability. If no suitable time exists, record that constraint. A polished reply cannot create a slot the practice does not offer. Bring scheduling evidence into the review so the campaign team and the practice can make a coordinated decision.

Use a reason code that tells the next person what to do

A short reason list works when it describes observed events. It fails when it becomes a set of opinions. “Not serious,” “cheap” and “bad attitude” do not identify a repairable part of the journey. Replace them with what the record can actually support.

Keep one primary reason for a particular stalled transition, plus a short note for relevant context. Otherwise the same person can appear in several loss categories and make the total larger than the group being discussed. A record may also change as new information arrives. Preserve the date and reason for a material correction.

Rank potential repairs by the strength of the evidence, the number of people affected, the consequence of the issue and the effort required to fix it. The largest category is not automatically the first action. An inaccurate confirmation may affect fewer people but require immediate correction. A large unresolved category may justify better record handling before a major campaign change.

End with a concrete assignment. “Improve lead quality” cannot be checked. “Replace the inaccurate appointment-confirmed message, test the received request and review whether that misunderstanding recurs” can. This is the difference between labelling a problem and taking responsibility for the work.