Start with the customer task the demonstration did not complete.

I would introduce the fictional account-opening prototype I approved in 2004, then ask participants whose work the demonstration leaves out. The pilot’s 120 businesses included fifty-six with two-person internal approval mandates, and thirty-eight of those were blocked by the single-approver flow. The workshop would explore pending approval, decline, retry and an assisted route as parts of the product rather than inconvenient exceptions. I would retain my responsibility for favouring the easy path and explain why the correction did not establish that every blocked business subsequently completed opening. Participants would work with an invented journey, without sharing customer applications or actual bank procedures. The proposed discussion would not prescribe a universal approval rule or suggest bypassing relevant checks. It claims no delivered event. Its purpose is to help a team define completion through the customer’s real task, including necessary participants who do not act at the same time.

I would examine the fictional accelerated-settlement proposal at Serravon, where research found different merchant needs and the commercial sponsor preferred paid default inclusion. The sixty-merchant pilot eventually recorded twenty-two choosing acceleration and thirty-eight retaining standard settlement. The proposed session would ask what successful choice means before adoption becomes the dominant measure. Participants would consider how to keep timing, price and conditions visible while preserving a credible standard route. I would distinguish product scope from treasury and operating decisions about what can actually be delivered. The example would contain no live offer, financial recommendation or universal settlement promise. It is a programme concept with no prior audience or appearance claimed. The intended contribution is a more precise discussion of whom an optional service serves, why a customer might reasonably decline it and how commercial ambition can be assessed without making every non-selection look like a product failure.

I would use the fictional invoicing retirement to examine what a customer needs in order to leave a feature usefully. Of fourteen hundred businesses, 960 could use the standard export while 440 required additional history fields. The discussion would follow the choice of an assisted extension and the later position in which 360 of that group had migrated while eighty remained supported. I would ask participants to consider the platform sponsor’s concern about continuing cost as well as the customer dependency. The session would distinguish a managed exception from an indefinite promise and avoid inventing a legal deadline or a completed saving. Supplied material would be fictional, with no actual customer records requested. This is a proposed event rather than a delivery credit. Its purpose would be to make the remaining work part of the retirement account, so a clear calendar cannot substitute for an exit the affected businesses can actually use.

I would present two fictional tests of a business cash-view prototype. Sixty-seven of 240 owners initially interpreted forecast receipts as spendable funds; after a redesign, fourteen of 180 still did so. The session would examine why an improved result can support further work rather than automatic release. I would discuss the decision to separate available balance from forecast information and remove a combined headline despite pressure for a smaller explanatory note. Participants would consider the difference between technical wording and the meaning users take from the whole presentation. The format would provide no banking advice, actual account information or claim of a full launch. It is an unpublished speaking proposal. The intended value is to help leaders keep unresolved understanding visible, assess the evidence with the relevant product and control owners, and resist turning a more favourable comparison into a statement that the underlying problem has disappeared. Participants would identify the next evidence needed before recommending a wider release.