MBA Capstone Projects: Choosing Research Questions That Matter

How to choose an MBA capstone or dissertation research question that is feasible, defensible, and actually useful to a real business. Twenty years of supervising the choice.

The wrong question makes the rest impossible

More MBA dissertations fail at the question stage than at any other point. A question that is too broad cannot be answered in 15,000 words. A question that is too narrow has no managerial relevance. The judgement is in the calibration.

The three-test filter

1) Is it answerable with data you can realistically collect in the time available? 2) Is the answer non-obvious — i.e. could a reasonable manager disagree about it before reading your dissertation? 3) Would a real business actually act differently after reading the answer?

Where good questions usually come from

Tension in the practitioner literature. Disagreement between a practitioner publication and an academic finding. A change in regulation, technology, or market structure that has not yet been studied in your context. Your own employer's actual decision the strategy team is stuck on.

Questions to avoid

'What is the impact of social media on consumer behaviour' — too broad, done a thousand times. 'How does company X improve sales' — consultancy, not research. Anything answerable with a single survey and a regression — fine for a project, not for a capstone.

Translating question to design

Once the question is sharp, the design follows almost mechanically. Causal claim → experiment or quasi-experiment. Exploratory claim → qualitative or mixed methods. Descriptive claim → survey or secondary-data analysis. Mismatched design is the second-most-common reason capstones underperform.