When people ask for appointments a practice cannot offer, their requests contain more than scheduling work. They can reveal a gap between the service people need and the choices available to them.
Understanding that gap can help the practice decide where to invest its attention: availability, the options it presents, or the way it helps someone who cannot find a suitable visit. The following synthetic example shows how to turn a recurring request into a useful business question.
Find the need behind the request
Imagine that later appointment times keep coming up in a practice’s service conversations. Some people accept an alternative. Others finish the exchange without a suitable option.
Start with the requests and their outcomes. Which times did people ask for? What choices were they offered? Did they accept one, agree to follow-up, or leave the need unresolved? Keep the exchanges connected to those details so the interpretation can be checked.
The first finding is modest but useful: some requests for later times are not being met by the options offered. That is enough to investigate. Estimating the scale of demand or the value of extending hours needs more context.
The same pattern can point to different decisions
| What might explain it | What to examine | What that could change |
|---|---|---|
| Available hours do not fit some people’s needs. | Unique requests, preferred times, alternatives accepted, and needs left unresolved. | Whether different availability is worth evaluating. |
| Suitable appointments exist but were not presented. | Availability at the time and the options actually offered. | How staff or software find and explain choices. |
| Several exchanges concern the same request. | The relationship between contacts and requests, using permitted information. | The estimate of demand and the effort caused by repeated contact. |
| A temporary disruption changed the choices. | Closures, staffing changes, or other relevant conditions. | The response to that disruption, rather than normal operating hours. |
Domain knowledge makes the difference here. A person who understands the practice can identify which explanations fit its operating reality and which facts are missing. An AI-generated theme helps organize the question; it does not supply that local knowledge on its own.
Let the finding inform both the business and the product
For the practice, the question is whether to change the service. It might assess different appointment options against staffing and capacity, or investigate why existing choices are not reaching the people who want them. The practice owns that decision.
For the product, the question is how to help the next person more effectively. It might ask about timing earlier, explain suitable alternatives more clearly, or preserve an unresolved request for useful follow-up.
These responses address different problems. Better questions can improve an interaction without adding capacity. New appointment options can address an access need without fixing an unclear explanation. Evaluate each against what it was intended to achieve.
Put the finding in front of a decision
A useful review note for this example could contain:
- Finding: some requests for later appointments remain unresolved, with the relevant exchanges and options attached.
- Question to resolve: is this unmet demand, an issue with how options are presented, or a temporary condition?
- Owner and timing: practice leadership, before its next availability review.
- Possible response: assess a change in appointment options; separately test a better product response to timing preferences.
- What to look for afterward: suitable options accepted, needs left unresolved, and repeat contacts for comparable requests.
The timing gives the information a purpose. A pattern can inform the next service decision; an individual unresolved request may need attention while the person is still waiting.
Check whether the response helped
State the intended improvement before changing anything. Then examine subsequent work alongside changes in request mix, staffing, and availability. A simple before-and-after difference may have more than one explanation.
Combine operating measures with direct feedback. A model’s interpretation of sentiment and a person’s reported satisfaction are different evidence. The GOV.UK Service Manual similarly recommends combining measures with user research and service data—a useful general principle, rather than a medical-practice requirement. Service measurement guidance
The same reasoning can help with repeated preparation questions or uncertainty about follow-up. The aim is to give the practice a better question, enough context to act on it, and a way to learn from the result.
This is a worked reasoning example, not a customer finding or a demonstration of available Koltra analytics.
Explore how Koltra connects service work and business understanding