The best branch may be solving a different problem

Is the group deciding where to add cover, which branch should test a service timetable or whether a workflow is ready to spread? Each decision needs a different comparison. A service-development question might use completed eligible activity per available consultation session. A staffing question might examine demand by hour and the experience of the team. Neither is answered adequately by prescription volume per employee across the whole month.

Payment references can help establish consistent terminology, but branch economics still depend on local costs. Separate incremental contribution from allocated central overhead, and explain how shared staff or delivery costs are apportioned. A branch can make a useful contribution without looking attractive after an arbitrary overhead allocation. Conversely, a high activity total can conceal overtime or repeated administration. The owner needs both operational and financial views before moving resources between sites.

Present each branch's completed service activity beside its available service sessions, professional cover and unresolved exceptions. Display numerator and denominator together. If one branch delivered twenty sessions and another delivered five, their totals should not be compared without that context. Avoid collapsing every dimension into a composite score: management needs to see whether a difference concerns demand, conversion through the pathway, capacity or recording quality before selecting an action.

Suppose one branch has a higher completion rate. Examine whether referrals arrive earlier, a trained colleague is consistently available or patients can reach the pharmacy more easily. The next test should focus on the plausible mechanism, not replicate every feature of the branch. PHARMORIS could help organise this chain from observation to explanation to bounded change, with the evidence and uncertainty available to the people expected to implement it.

A sample comparison would put completed eligible consultations beside available sessions, referral timing and recorded exceptions for each participating branch. It would also show whether the underlying records are complete enough to compare. The owner could distinguish a branch receiving fewer suitable referrals from one losing work between booking and completion. The proposed output would finish with a specific choice to test, such as adjusting a session timetable, rather than a single winner-and-loser score.

Agree the limits in advance: additional administration, service delays, staff confidence and data completeness. Stop a comparison that depends on incompatible records or makes teams reluctant to report exceptions. GPhC standards support a safe, well-managed service; a performance process should reinforce that environment (General Pharmaceutical Council, n.d.). Improvement is not credible if it depends on hiding difficult cases, excluding inconvenient shifts or allowing one branch to absorb work that another can no longer handle.

A small group can move quickly when the question is specific and local knowledge is respected. The result need not be a national benchmark. It can be a well-supported decision to change a timetable, provide training, avoid an unsuitable rollout or repeat a promising test. Bring a branch difference you want to understand and the management decision it could change, and we can examine whether a focused evaluation would help.