Conversations that keep management answerable for its interpretation.

I would examine the point at which an analytical estimate changes meaning because of the audience receiving it. The discussion begins with a fictional turnover pilot covering 686 employees and the proposed release of individual rankings. It asks participants to distinguish a pattern worth investigating from a judgement about a named person. I would explore the strongest argument for using such tools: managers already make imperfect decisions, and analysis may expose what intuition misses. The session then considers the intended decision, plausible misuse, specialist responsibilities and the employee’s ability to question the account. It does not provide a legal assessment or a technical model certification. The practical output is a better set of questions for the management team before a pilot moves into use. Participants should bring a general decision context, not identifiable employee records or confidential model results. The aim is to make ownership of the employment consequence as explicit as ownership of the analytical method.

I would use a vacancy register to show how one exact total can combine several different decisions. The fictional example begins with 164 entries and ends with 92 current requirements, 48 obsolete requests and 24 duplicates. Participants would examine why the lower active count is a correction rather than evidence of 72 salaries saved. The session connects reporting definitions to the work executives actually need to decide, including hiring priorities, expense commitments and the distinction between direct employees and partner workforces. I would also discuss how to explain a change in category without presenting it as an operating improvement. This is a management workshop, not a replacement for financial accounting or specialist employment advice. Teams should use anonymised or invented material that they are authorised to share. The useful outcome is a clearer account of the population, purpose and limits of a measure, together with the responsibility for reconsidering it when the business changes.

This session examines how a survey becomes a management story. I would begin with the fictional 2016 exercise in which 586 responses from 880 invitations across nine units were initially summarised too confidently. The discussion asks what participation can tell us, what it cannot establish and how a material minority concern can remain visible without exposing the speaker. I would invite leaders to consider the tension between a useful summary and an account that erases context. The session includes management’s response after listening: the need to describe an action, an inquiry or a reason for not proceeding, rather than treating a presentation as the end of the conversation. Privacy and handling decisions require appropriate specialist guidance and are not settled by the workshop. No private comments should be shared. The intended value is a more careful relationship between evidence, interpretation and the explanation employees receive after they have been asked to contribute.

I would discuss why common titles and salary categories can conceal differences in responsibility after a transaction. The fictional case examines 126 roles, with 91 supported matches, 23 revised descriptions and 12 matters left open for further review. Participants would consider the cost of a ten-week extension and the pressure to publish one completion date. The central question is which differences matter to the decisions that the mapping is intended to support. I would also explore the opposite risk: allowing classification work to remain indefinitely open without a responsible owner or next step. This session distinguishes administrative integration from an employee’s experience of change and from the transaction’s commercial result. It does not interpret acquisition agreements or individual employment obligations. A sponsor should identify the general operating question in advance and avoid supplying confidential deal or personnel material. The discussion should help leaders recognise when a finished-looking report is concealing a decision that has not actually been made.