Yes — when their mandate says evaluate, they do. In a 2026 study with 31 Emirati managers, evaluators asked to independently assess an initiative that senior leadership had publicly backed requested evidence on whether it works at the same rate as when no decision had been signalled: 58% of their requests, against 56%. A change of mandate, from evaluating to implementing, is what shifted what they asked for.
Every leader who champions an initiative and then commissions an independent review faces the same quiet question: will the review still be a review? The concern is reasonable. The data from this study offer a practical answer, and a simple way to check it in your own organization.
Why does it matter whether evaluators keep evaluating after leadership has backed a project?
It matters because the evaluation is usually the last point at which an initiative can be adjusted cheaply. Once rollout begins, every correction costs more. Dan Lovallo and Olivier Sibony’s study of 1,048 major corporate decisions found that the decision process — including whether viewpoints contradicting the senior leader’s were openly discussed — mattered for outcomes, not only the depth of the analysis. They also noted that discussion is harder to keep open when everyone in the room already knows the decision maker’s view.
That is the situation of an evaluator reviewing a project leadership already supports. It connects to a wider pattern in organizational governance: why human review so often misses what it was set up to catch, and what happens when approval continues after examination has stopped. The question is whether evaluators, given a clear mandate, keep examining.
What did the study with 31 Emirati managers find?
The managers held the evaluation mandate when leadership had publicly backed the initiative, and shifted their search when the task became implementation.
Each participant worked through six short business situations — outsourcing internal IT support, a new performance-management system, an AI-based customer-service pilot, and others. In each, they could request three of six additional pieces of information. Two items offered evidence on whether the initiative works, two offered guidance on how to implement it, and two offered background context. Each situation came with one of three framings:
- Open evaluation: leadership has not decided; evaluate and recommend.
- Backed evaluation: leadership has publicly expressed strong support; evaluate independently anyway.
- Implementation: the decision is made; plan how to implement it well.
Under open evaluation, 56% of requests were for evidence on whether the initiative works. Under backed evaluation, the share was 58%. When the task became implementation, requests for implementation guidance rose to 41%, from 33% under backed evaluation — the one comparison in the study that held under every statistical test applied.
The study tracks what managers asked to see, not what they finally recommended.
Why does the mandate shape what people look for?
Because people tune their attention to the task in front of them. Peter Gollwitzer and colleagues showed in 1990 that people weighing a decision attend to information about whether an option is desirable and feasible, while people planning its execution attend to when, where and how. The original experiments involved individuals and personal decisions. The managers in this study showed the same task-congruent pattern in organizational situations: asked whether, they looked for evidence; asked how, they looked for implementation guidance.
Even in implementation mode, 47% of requests remained evidence-focused — much of it the kind of performance information an implementer needs to track whether a rollout is working.
The practical reading is that the mandate carries real weight. Evaluators respond to the question they are actually given.
What should leaders do when they want an honest review of a project they support?
State the question in writing, and state which question it is. Four practices follow from the data:
- Name the mandate explicitly. “Assess whether we should proceed” and “determine how we implement” are different assignments. Evaluators in this study treated them differently; give them the wording that matches what you need.
- Keep support and mandate distinct. Public backing and an open evaluation are compatible. The managers in this study treated them as compatible, requesting evidence at the same rate whether or not leadership had signalled support.
- Ask for the information request before the recommendation. The first three things an evaluator asks for show which question they believe they are answering.
- Set the point at which evaluation closes. A defined date after which “whether” becomes “how” protects both the evaluation and the rollout.
How can you tell whether a review is still an evaluation?
Look at the first requests. Evidence requests — comparisons with the current system, consequences if a trend continues, results from further testing — signal evaluation. Rules, procedures and rollout options signal implementation. Both are legitimate; what matters is whether they match the mandate.
| Mandate given | First requests mostly evidence | First requests mostly implementation |
|---|---|---|
| Evaluate whether | Review is working as intended | Clarify the mandate before the recommendation |
| Plan how | Check whether the decision is settled | Review is working as intended |
The table takes a minute to apply and needs no new process: the request list already exists in every review.
Frequently asked questions
Does visible support from senior leadership make independent evaluation pointless?
No. In this study, managers asked to evaluate a publicly backed initiative requested evidence on whether it works at the same rate as when no decision had been signalled — 58% against 56%.
What is the difference between evaluation information and implementation information?
Evaluation information tells you whether to proceed: comparisons, consequences, results. Implementation information tells you how to proceed: rules, procedures, escalation paths, rollout options.
Is it a problem if evaluators ask for implementation details?
Not when the mandate is implementation. When the mandate is evaluation, a request list that leads with implementation details is a prompt to restate the question before the recommendation arrives.
How was the study conducted?
Thirty-one Emirati managers enrolled in a UAE business-school program completed six scenario-based decision situations in 2026. Each scenario rotated across the three framings between form versions, so every participant contributed data under all three.
Sources and further reading
- Gollwitzer, P. M., Heckhausen, H., & Steller, B. (1990). Deliberative and implemental mind-sets: Cognitive tuning toward congruous thoughts and information. Journal of Personality and Social Psychology, 59(6), 1119–1127. https://doi.org/10.1037/0022-3514.59.6.1119
- Lovallo, D., & Sibony, O. (2010). The case for behavioral strategy. McKinsey Quarterly, March 2010.
