One-Click Refill Versus Six-Field Verification
Two stakeholder groups on a health-plan portal modernization disagree openly. Member services wants a one-click prescription refill from the member dashboard, arguing it will reduce refill call handle time, currently 6.4 minutes across 2,900 calls per month. Pharmacy operations insists on a six-field verification step because a state audit last year produced three findings about missing refill authorization evidence. Compliance states that audit evidence must be preserved in any redesign. IT estimates one-click at two sprints and full verification at one sprint plus integration work. Ten weeks remain at fixed sprint capacity. The backlog holds 240 items, three of which directly contradict one another.
1. Is this a genuine requirements conflict? Analyze it and propose how to resolve it.
Read the worked answer
Test whether the conflict is at the need level or the solution level. It is at the solution level:
- Member services' need: reduce member effort and call handle time for refills.
- Pharmacy's need: produce defensible evidence that refill authorization was verified.
- Compliance's constraint: audit evidence preserved.
Neither side's need requires the other's solution. A one-click button is not inherently evidence-free, and a six-field form is not inherently slow if the fields arrive prefilled.
Candidate resolution: prefill the six verification fields from the member's profile, coverage record, and claims history; the single click becomes the member's confirmation of pre-filled data. Log the field values, the timestamp, the member identifier, and the source of each prefilled value. That produces stronger audit evidence than a form the member types quickly, and it removes the typing effort that drives the 6.4-minute handle time.
Do not declare this resolved on your own authority. The auditor's three findings define what counts as evidence. Bring the proposed evidence record to compliance and, if possible, to the auditor who wrote the findings, and get explicit confirmation before designing to it. The risk here is a technically elegant solution that still fails the audit.
Also confirm the handle-time premise: if much of the 6.4 minutes is identity verification or payment, one-click refill will not deliver the expected reduction, and that changes the priority fight.
2. What requirements model would you use to expose where verification is genuinely required, and why that model rather than another?
Read the worked answer
The underlying disagreement is conditional, not universal. Verification depth plausibly varies by situation: whether the member's profile data is current, whether the drug is controlled, whether the refill is too soon, whether coverage changed since last fill. A single yes/no argument about "one-click" hides those conditions.
Use a decision table. Rows: the condition combinations (data freshness, drug schedule, refill timing, coverage change, authorization on file). Columns: verification depth (none/prefilled confirmation/full manual), evidence captured, and resulting member experience. Benefits:
- It makes the conflict concrete and countable - pharmacy can point at the specific rows where they insist on full verification, and member services can see that most refill scenarios fall in the one-click rows.
- It produces testable acceptance criteria for UAT and gives compliance a readable statement of what evidence is captured per scenario.
- It is compact enough to review with both groups in one session, which matters with ten weeks left.
Support with a process model of the refill flow end-to-end (member action → verification → authorization → dispensing → evidence log) so both groups see where their step sits, and a state model for prescription status if approvals and rejections need explicit states. Avoid use cases here: they describe interaction, not the condition logic that is actually in dispute.
3. With 240 backlog items, ten weeks, and three contradictory requirements, how do you prioritize - and what must you not do?
Read the worked answer
Sequence by constraint type, not by stakeholder volume:
- Mandatory first. Compliance/audit-driven requirements and anything contractually or legally fixed. These are not negotiable against value; they define a floor. The verification-evidence requirement sits here.
- Resolve the three contradictions before scoring anything. Ranking an item against a contradictory peer produces a meaningless order and will resurface later as rework. Take the three to a joint session with the two stakeholder leads and the decision owner; if unresolved in that session, escalate with the options and consequences written out, not with a request to "decide."
- Score the remainder on a defined model - business value, risk reduction, effort (sprint cost from IT's estimates), and dependency. Publish the scores so priorities are arguable on evidence rather than seniority.
- Make scope explicit against capacity. Ten weeks at fixed capacity cannot absorb 240 items. Produce a committed set and a deferred set, and have the sponsor accept the deferred list in writing.
What not to do: do not let sprint capacity silently determine scope. If the BA simply drops what does not fit, the sponsor never owns the trade-off and the deferred items will reappear mid-sprint as escalations. Also do not let "mandatory" expand unchecked - challenge whether each compliance-driven item is truly required by the finding or merely associated with it.