Meetings Without Decisions
There is a phrase that has quietly done real damage to how institutions make decisions, and it is this: data-driven decision-making.
Not because data is the problem. Because driven implies agency that data does not have. It suggests that if you collect enough of it, clean it sufficiently, and present it in a compelling format, the decision will make itself. It will not. It never did. And as the AIM committee meets this week in Washington. It’s the final session of negotiated rulemaking on Accreditation, Innovation, and Modernization.
The AIM committee is the Department of Education’s negotiated rulemaking process for reforming how accreditors evaluate institutional quality. The framework has five focus areas, but the one with the most direct implications for IE professionals is this: accreditors would be required to evaluate institutional quality using program-level student outcomes data — retention, graduation, employment, and median salary, down to the six-digit CIP code. As I wrote in What’s the Difference Between IE and IR?, IR can pull every one of those numbers. IE is what determines whether anyone does anything meaningful with them.
What no regulation touches is whether institutions know how to make a decision once the data is in the room.
Structured Decision Making, or SDM, is an approach rooted in the decision sciences that has been applied everywhere from natural resource management to public policy to complex organizational governance. The U.S. Geological Survey defines it plainly: every decision consists of management objectives, decision options, and predictions of what each option would produce.
The data’s job… its only job… is to model those predictions. It does not select among the options. It does not determine the objectives. It illuminates consequences and then humans decide.
That last sentence matters more now than it did five years ago. Because the question I hear more frequently, especially as AI tools become standard in institutional workflows, is whether the analysis can just tell us what to do. It cannot. Neither can the dashboard. Neither can the model. Data surfaces consequences. AI can accelerate the analysis. But the moment someone in the room has to say this is what we are going to do, and own it…that is a human moment. It always will be. And most institutions have not built a process that makes that moment possible.
The PrOACT framework, developed by Hammond, Keeney, and Raiffa in Smart Choices and widely used in SDM practice, structures decisions around five elements: the Problem, the Objectives, the Alternatives, the Consequences, and the Trade-offs. Notice what comes first: the problem (not the data). You cannot usefully model consequences until you have named the alternatives. You cannot name the alternatives until you have articulated the objectives. And you cannot articulate the objectives until you have defined, precisely, what decision you are actually here to make.
Most institutions skip the entire front half of that process.
They bring data to a meeting and let it do the work of defining what the meeting is about. The data becomes the agenda. The agenda becomes the conclusion. Someone raises a concern. Someone else cites a different number. A third person asks for a follow-up report. The meeting ends. No decision was made, because no decision was ever named. What happened was a data review, which is a useful thing, but it is not a decision-making process.
I have been in rooms where this happens at every level: program review committees, cabinet retreats, board finance sessions. As I explored in More Data, Less Clarity, the volume of institutional data has never been higher and the confidence in what to do with it has rarely been lower. That gap is not a technology problem. It is a decision-making problem.
The IE professional who understands SDM brings something specific to that room that no dashboard and no AI tool can provide. They can open the meeting with a question that reframes everything that follows: What decision are we here to make?
That question is not procedural; it is foundational. It forces the group to name the decision before examining the evidence, which means the evidence gets evaluated against actual alternatives rather than in the abstract. It separates the people who are there to model outcomes from the people who are authorized to choose among them. It makes the data’s role explicit, not the authority in the room, but the intelligence that informs the authority in the room. That is Signal #4 (Executive Leadership) and Signal #5 (Operational Quality) working in concert: leadership that can define the decision, and operational systems sufficient to carry it through.
This is not a new idea. It is, in fact, the original idea behind institutional effectiveness. IE did not emerge as a reporting function. It emerged because accreditors in the 1980s started asking not just what institutions were producing but what they were doing about it. That is a decision question, not a data question. The data describes the gap. IE closes the loop between evidence and action. That loop has always required someone who knows how to structure the decision, not just deliver the report.
AIM is creating conditions where program-level outcomes data will be a live accountability input, not a self-study artifact pulled together every ten years. As I wrote in Program Outcomes as a Strategic Asset, that cadence shift changes what IE needs to be ready to do. The institutions positioned to navigate this well are not the ones with the most data. They are the ones with people who know how to turn data into a decision and the process to back them up.
This week on the EdUp Institutional Effectiveness podcast, I spoke with Christy England, Founder and President of Authentic Insights and contributing author of the forthcoming Priority Partners: Turning Vendor Spending into Mission Strength (IEHE, Summer 2026). Her chapter centers on accreditation as a strategic partnership, and her core counsel applies well beyond the accreditor relationship: the only wrong answer is a dishonest answer. Name the problem accurately. Bring the decision to the room, not just the data. The episode is available on Spotify and Apple Podcasts.
Data does not improve institutions … decisions do. A meeting without a clear decision to be made at the beginning of the meeting is just an expensive conversation.
When did your last meeting name the decision, surface the options,
and let the data inform rather than decide?
