# How to Build a Research Brief That Produces a Decision

> A practical method for turning a broad research request into a decision, evidence plan, scope and usable definition of done.

- Canonical page: https://mtfinstitute.com/insights/research-brief-that-produces-a-decision/
- Content type: Article
- Editorial category: Guides & Frameworks
- Publisher: MTF Institute of Management, Technology and Finance
- Author: [Dr. Alex Amoroso](https://mtfinstitute.com/about/faculty/alex-amoroso/)- Published: 2026-07-26
- Updated: 2026-07-26
- Living edition: 1.0
- Language: English
- Topics: Business Analysis, Product Strategy, Research, Decision Making

A useful research brief does not begin with a method. It begins with a decision. Before commissioning interviews, a survey or a market report, the team should be able to state who will decide what, by when, and what uncertainty prevents that decision today.

That distinction matters because “understand our customers” can produce months of activity without changing a product, investment or operating choice. A decision-ready brief narrows the work without pretending that the answer is already known.

## Start with the decision statement

Use a sentence with four parts:

> By **date**, **decision owner** must decide **choice**, using evidence about **uncertainty**.

For example: “By 30 September, the product director must decide whether to redesign onboarding for small accounting firms, using evidence about where qualified users abandon setup and why.”

This statement identifies timing, ownership, choice and uncertainty. It also exposes requests that are not yet ready for research. If no one owns the decision, or no action can change, more evidence may not create value.

## Separate questions from assumptions

List what the team believes, then mark each statement as a fact, an interpretation or an assumption. Research questions should target assumptions that could change the decision.

| Brief element | Useful question |
| --- | --- |
| Behavior | What do people currently do, and at which point does behavior change? |
| Motivation | What outcome are they trying to achieve? |
| Constraint | What prevents or delays that outcome? |
| Variation | Which segments, contexts or roles behave differently? |
| Consequence | What would the organization change if the finding were true? |

Avoid questions that merely invite agreement with an internal idea. “Would users like this feature?” is weaker than “Which existing task would this feature replace, and under what conditions?”

## Define the evidence boundary

The brief should name evidence already available, evidence that is missing and evidence that is out of scope. It should also identify the population, relevant contexts and important exclusions.

A small study can be rigorous when its claims are appropriately bounded. A large dataset can still mislead when it does not represent the decision population or when the measures do not answer the research question.

## Choose a method after the question

Interviews can reveal reasoning and context. Observation can reveal behavior that participants do not report. Surveys can estimate patterns in a defined population. Product analytics can show where behavior occurs but may not explain why. Mixed methods are justified when the decision requires both explanation and estimation.

The brief should explain why the proposed method can answer the question and what it cannot establish.

## Make the output decision-ready

Define the deliverables before work begins:

- findings linked to evidence;
- material differences between segments or contexts;
- limitations and unresolved uncertainty;
- decision options and implications;
- recommended next actions;
- an appendix that preserves method and traceability.

The definition of done is not “deliver a presentation.” It is “provide enough traceable evidence for the named owner to make the named decision within the agreed boundary.”

## A five-test quality check

Before approving a brief, ask:

1. Is there a real decision and a named owner?
2. Could a plausible finding change that decision?
3. Does the method match the research question?
4. Are population, context and limitations explicit?
5. Is the final output defined in decision terms?

If any answer is no, improve the brief before increasing the sample or adding methods.

## Related MTF resources

Explore the [Research and Business Analysis Practice](/for-business/practices/research-and-business-analysis/) and the faculty profile of [Dr. Alex Amoroso](/about/faculty/alex-amoroso/).


## Editorial version

This is living edition 1.0. MTF Institute maintains this article with the named faculty author. Material revisions receive a new version and an updated publication date.

## Citation

When citing or summarizing this material, link to the canonical HTML page: https://mtfinstitute.com/insights/research-brief-that-produces-a-decision/