Work through the choice before asking for a confident answer

I ask participants to decide what evidence they would require before renewing two channel allocations with equal opening line counts. The fictional exercise follows two cohorts of 2,400 lines and their different ninety-day active totals. We examine the observation window, the distinction between a line and a person, and the treatment of introductory support and commissions. The discussion deliberately avoids turning the observed difference into a causal claim. Participants compare a decision to preserve the allocation, a bounded reallocation and a request for further evidence, including the opportunity cost of waiting. I explain why finance should own the contribution definition while marketing remains accountable for the allocation proposed. The session is intended to improve the decision account rather than teach a universal retention benchmark. It closes by asking how the opening goal and later review could have been agreed together, so colleagues are not judged retrospectively against an event they were never asked to manage.

This workshop puts a proposed communication saving beside a customer choice that the simplification may obscure. Participants work through a prepaid mobile proposition and a proposed combined brand, considering commitment and replenishment without assigning preferences to demographic stereotypes. The fictional research findings are useful but limited: selected participants express a preference, not a forecast of churn. I ask the group to write the decision that follows from those findings and the evidence that could change it. We then price a six-month extension and define the point at which management must return to the choice. The exercise includes a finance challenge about whether an annual saving is being compared fairly with a temporary cost. The aim is neither to defend every existing brand nor to favour consolidation automatically. Participants should leave able to explain what a separate proposition does for the customer, why the company might pay to preserve it and when that argument must be tested again.

I use a twelve-location assisted-choice pilot to examine the difference between a convenient measurement path and the customer's actual decision. The exercise contains 360 sessions,228 completed applications and 132 encounters without an application. Participants must state what those counts establish before recommending another period of investment. We consider the possibility that some people would have applied digitally anyway, and the possibility that an assisted conversation serves a need the online report misses. Neither possibility is treated as a measured result. I explain the division between marketing authority and distribution's responsibility for people and operations. The group develops an evaluation question that could help the next decision without pretending the original pilot was a controlled experiment. This is a discussion of commercial judgement, not an argument that human assistance is always superior or that digital acquisition is inadequate. Its purpose is to make competing explanations visible before the budget adopts one of them by default.

I examine a disputed C$2.6 million allocation in which the final recorded click was asked to carry a larger investment conclusion than the model could support. Participants distinguish an observed conversion path, an attribution convention and an estimate of additional business. The case then becomes more difficult: rejecting the narrow ranking does not establish that the existing brand programme deserves protection. I describe the bounded decision to retain C$1.8 million and transfer C$800,000, with the assumptions and six-month review kept visible. The roundtable asks what an executive should document when judgement remains after analysis, and what a board should question without choosing channels itself. I also discuss why the absence of an attributed return at review must remain in the public account. The session gives participants a way to challenge false certainty while retaining responsibility for expenditure. It does not supply a replacement model or promise that a particular allocation method will improve returns.