The study puts the same three questions to four leaders inside one company, none of whom sees the others’ answers. The finding is not any single answer. It is the distance between them, scored from 0 to 100, and that score is the Decision Delta.

What is being measured

Not performance. An organization can perform well for a period and be unable to say why, which is a different condition from performing badly and knowing exactly why.

Decision Delta measures whether a leadership team holds a common account of its own marketing: the same record of what happened, the same explanation for it, and a shared view of what follows. Where those three diverge, the organization is deciding by seniority rather than by evidence.

Who is asked

Four people per organization, chosen by role rather than by availability: the executive accountable for revenue, the leader accountable for marketing, the person closest to the reporting, and one operator running the work.

They answer separately and without seeing each other’s responses. That constraint is the whole instrument. Any version of this survey that lets a team confer measures the meeting rather than the organization.

Ask four people the same question separately, and the distance between the answers is the finding.

What the early responses suggest

The pattern that recurs in pilot interviews is that agreement decays as the question moves from fact to decision. Teams converge on what changed. They diverge on why. They diverge further on what to do next, and the divergence is rarely visible to them, because the question is seldom asked in a form that would expose it.

These are field notes rather than findings. Nothing here is a published result, and none of it should be cited as one until the first edition is out with its method and its raw data.

Status

Fieldwork opens September 2026. Edition one publishes in January 2027, with the full method, the anonymized dataset and the scoring model. Participating organizations receive their own Decision Delta score before publication.