Opening a Second Clinic Branch: Prove Demand, Capacity and Record Boundaries
Evaluate a second clinic with incremental economics, branch-specific schedules and billing, controlled record access and a reversible opening trial.
Choose clinic KPIs that lead to a specific decision, and define their timestamps, units, denominators, exclusions and data-quality limits before building a dashboard. Combine patient outcomes or experience with process and balancing measures. Waiting, follow-up completion and unresolved work are useful only when the team can explain exactly what was counted and what action the result supports.
Indian clinic owners and administrators selecting a small operational dashboard or reviewing reports from clinic software.
'Improve efficiency' is too broad to define a measure. A more useful question is whether the evening session starts late, whether planned reviews are being completed or whether unresolved results lose an owner. Name the decision, the people affected and the operational change that might follow from the evidence.
IHI distinguishes outcome, process and balancing measures in improvement work. The distinction helps a clinic avoid optimising one number while creating another problem. For example, a change intended to reduce waiting should also be reviewed for rushed communication, unfinished work or staff burden. The specific measures below are original clinic examples, not mandated targets.
Use a small set that staff can collect and interpret consistently. A dashboard with twenty unexplained percentages can be less useful than three well-defined measures and a short discussion. Include patient and staff observations so that a tidy chart does not conceal a confusing or inaccessible process.
Choose the timestamps that answer the question. Arrival-to-consultation time measures the patient's elapsed wait after arrival. Scheduled-time-to-consultation delay measures how late the consultation starts relative to the appointment, using a clear rule for early starts. These measures differ when a patient arrives early or late.
In a fictional case, the patient arrives at 09:40 for a 10:00 appointment and the consultation begins at 10:15. Arrival waiting is 35 minutes; start delay is 15 minutes. Another patient arriving at 10:10 for the same scheduled time and starting at 10:15 waits five minutes, while the appointment delay is still 15. Neither value by itself assigns blame.
Choose a consistent summary. For five fictional waits of 5, 10, 15, 20 and 60 minutes, the median is 15 minutes and the mean is 22. The long wait matters even though most waits are shorter. Review the distribution and outliers rather than presenting the preferred statistic as the whole story.
Record missing timestamps and exceptional pathways separately. Do not enter zero for an unknown consultation start, and do not silently remove a person who left before being seen. An apparent improvement can result from losing the most difficult cases from the data. Define walk-ins, late arrivals and priority handling in the measure without allowing the metric to dictate clinical urgency.
A follow-up measure needs a defined group and time window. Distinguish visits clinically planned, appointments actually scheduled, reminders attempted and reviews completed. Sending a message is not the same as a completed review, and creating an appointment does not establish that the person attended.
In a fictional cohort of 40 planned reviews due in a defined period, 24 are completed, six are rescheduled, four are cancelled and six have unresolved status. The categories total 40. Completion among the full original cohort is 60%. If a report excludes the four cancellations, 24 out of 36 becomes 66.7% when rounded to one decimal place. That change comes from the definition, not improved care.
State how a reschedule, a review completed elsewhere or a clinician-changed plan is handled. Preserve the original cohort when comparing over time, or explicitly describe the revised method. Avoid treating every missed booking as treatment failure or sending repeated reminders where the care plan no longer calls for them.
Use an appropriate owner to review unresolved cases. The operational dashboard should identify missing status and handoff needs; the responsible clinician determines clinical priority and next steps. Do not invent a uniform recall interval or escalation deadline merely to make the chart easy to calculate.
An open-work count should reconcile. Start with the items open at the beginning of the period, add new and reopened items, and subtract genuinely closed items. If an item was transferred to another person, decide whether it remains open for the clinic rather than counting a handover as resolution.
For a fictional week, 18 items are open at the start, 12 new items arrive, three are reopened and ten close. The ending count is 23. A report showing only 'ten completed' misses the growth in unresolved work. Define the item itself so that three messages about the same issue do not become three separate clinical tasks.
Pair counts with age, responsible owner and appropriate priority. A small backlog can still contain an important unresolved concern. Protect identifiable details in management views and avoid turning a clinical exception list into an unrestricted staff dashboard.
On a narrow screen, scroll the table horizontally to read all columns.
| Measure | Definition to settle | Decision it can support |
|---|---|---|
| Arrival waiting | Arrival to actual consultation start; missing and left-before-seen cases reported. | Review reception flow and the patient's experience. |
| Appointment start delay | Scheduled time to actual start, with an early-start rule. | Review session starts, overruns and capacity assumptions. |
| Planned-review completion | Completed reviews within a defined original cohort and window. | Investigate unclear plans, scheduling barriers or missing status. |
| Unresolved work | Opening items plus new and reopened minus closed. | Assign ownership and examine growing or ageing work. |
| Collection status | Amounts collected for defined charges and periods, with refunds distinguished. | Investigate financial exceptions without equating billing with care completion. |
| Balancing observation | A named effect such as staff overtime or patient clarification needs. | Check that an improvement elsewhere is not creating a new problem. |
Ask the software provider or analyst to show how one visible value is built from synthetic source events. Can the clinic trace the timestamp, status and inclusion rule? A polished chart is difficult to trust if two team members produce different counts from the same cases.
Keep a versioned definition. If the clinic changes what 'completed' means, note the change on the trend rather than comparing the old and new figures as if they were identical. Identify data-entry gaps and timezone assumptions, especially when events are recorded after an interruption.
Do not collect more personal information simply because a dashboard can display it. Group results at an appropriate level, restrict detailed exception views and retain only what the purpose requires. A small clinic may need to consider whether tiny subgroups make individuals obvious even without names.
Plot consistent measures over time and annotate operational changes, such as an added receptionist or a revised session start. Compare similar services and periods. A festival week, a doctor's leave or a change in appointment mix can alter the result without proving that the latest software change caused it.
Use the data to select a small testable action. If waiting rises only after a late hospital handover, changing the registration form may not address the cause. If follow-up status is missing because another clinician completed the review, increasing reminder frequency may make the experience worse.
Close the review with an owner, action and observation plan. Avoid imposing a target that encourages staff to manipulate timestamps, close unfinished tasks or discourage complex patients. A useful KPI improves the decision process; it is not a certificate of clinical quality or a substitute for professional judgement.
A fictional clinic completes 24 of 40 planned reviews. One report shows 60%; a second report removes four cancelled reviews and shows 66.7%. No additional review occurred. The owner reconciles the cohort and publishes the denominator rule beside the figure, while separately investigating the six unresolved statuses. In the same week, ten tasks close but open work rises from 18 to 23 after new and reopened items are included. The combined review prevents two reassuring headlines from hiding unchanged follow-up completion and a growing backlog.
Use these prompts in your clinic’s own notes. Record the result, responsible person and next action for each item.
Choose a few linked to current decisions, such as waiting, appointment delay, planned-review completion and unresolved work. Define them precisely and add a balancing observation. There is no universal dashboard or target appropriate for every clinic.
They answer different questions. The median describes the middle observed wait; the mean is affected by long waits. Report the definition and examine the distribution and missing cases rather than selecting whichever number looks better.
Decide and disclose the rule according to the question being asked. Excluding cancellations changes the denominator and can raise the percentage without another completed review. Keep the original cohort and status categories traceable.
Not necessarily. Reconcile opening items, new arrivals, reopened items and closures. A transfer is not automatically a resolution, and the ending backlog can grow even during a busy week.
Ask for the metric definition and trace a value through synthetic source events, including exclusions and missing data. Check that the same rules apply across periods and branches, and identify what the metric cannot establish.
Updated on . All numerical examples are fictional. The measures are operational tools, not clinical quality certification, validated benchmarks or recommended clinical review intervals. No independent clinical or legal review is claimed.
Improvement-method guidance on outcome, process and balancing measures; not an OPD benchmark or mandated reporting set.
Evaluate a second clinic with incremental economics, branch-specific schedules and billing, controlled record access and a reversible opening trial.
Give a clinic administrator clear decision rights, access limits, spending boundaries and exception ownership without sharing the doctor's login.
Turn vendor feature claims into scoped evidence, exception tests, dependency checks and purchasing decisions using a practical capability record.
Ask Bractis to calculate waiting, follow-up completion and open work from a small synthetic event set. Include an early arrival, cancellation, reopened task and missing timestamp, and compare the report with the written definitions.
Request a clinic demo