Strategic Choice Structuring (What Would Have to Be True?)
When you need this method
Two or three paths sit side by side in the leadership team, and the discussion goes in circles. Everyone argues from a finished conclusion while the assumptions underneath stay unspoken. In the end rank or stamina decides, not the better argument. Or a soft consensus forms that nobody defends once the first resistance appears.
Approach
- 1Frame the question as a genuine choice between at least two mutually exclusive paths, not as a yes or no to a single proposal.
- 2Collect several possibilities and tell each one as a short story: where would we play, how would we win? Nothing is judged at this stage.
- 3For each possibility separately, write down what would have to be true about market structure, customer value, your own capabilities, costs and the competitive response for it to hold.
- 4Rank the conditions by how confident the group is. The condition with the least confidence is that possibility's barrier condition.
- 5Agree on a test for every barrier condition and let the person who doubts most set the standard of proof. Otherwise the result gets renegotiated later.
- 6Test in reverse order of confidence: start with the condition whose failure kills a possibility outright. The expensive analyses come afterwards.
- 7Choose the possibility that survived the hardest test and carry the still open conditions forward as named pieces of work.
Typical application
A typical B2B SaaS in the HR space has been arguing for months about the next move: more sales coverage in its home mid-market, a second product, or an entry into France. Instead of continuing to argue, the leadership writes down for each path what would have to be true. For the market entry the top condition is that sales must work without a dedicated local leader. That is precisely what the sales director doubts most, so it gets tested first. The agreed test is twelve first conversations with French target accounts, run by the existing team, and the sales director himself sets how many of them must go well. Ten break down over language. The option is closed after three weeks, with no market study and no recruiting, and the group now argues about two paths instead of three.
Limits and counter-indications
The process presupposes a genuine choice. Where only one possibility is seriously on the table, it creates work without insight, and for day-to-day operating decisions it is too heavy. Two limits weigh more. First, a condition that does not hold today can sometimes be made true; Roger Martin added an assessment stage to his own process for exactly that reason, because otherwise good possibilities die too early. Second, the process depends on people exposing their own chain of reasoning and challenging that of others. Where acceptability becomes the yardstick, or where a dominant leader signals a preference early, the uncomfortable formulations that matter most are the ones that get withheld.
How to measure impact
Measure the process, not the outcome: how many genuine possibilities were on the table, how many barrier conditions were named per possibility, how many of them were tested before the decision, and how many options failed on a test rather than in execution.
Related methods
Sources
- 1.Roger L. Martin: Strategic Choice Structuring, Whitepaper, 1997 (opens in a new tab) · Roger L. Martin (Rotman School of Management, University of Toronto) · 1997 · practitioner source · describes the methodCarries the full five-step procedure: frame the choice, brainstorm possible options, specify the conditions necessary to validate each option, prioritize the conditions that create the greatest barrier to choice, and design valid tests for those barrier conditions. It states explicitly that conditions are derived backwards from the option, that they are examined in reverse order of the team's confidence, and that the standard of proof is set by the most demanding member. Limit: a consulting white paper built on a single narrated case, with no effectiveness measurement and no comparison group.
- 2.Baer, Dirks & Nickerson: Microfoundations of Strategic Problem Formulation, Strategic Management Journal 34(2), 2013, S. 197–214 (opens in a new tab) · Wiley / Strategic Management Journal (Volltext über University of Illinois) · 2013 · academic and scholarly literature · supports the underlying mechanismCarries the mechanism behind the procedure: formulating a strategic problem is distinct from solving it and determines which problem gets solved at all. The paper cites an analysis of 33 case histories in large US companies in which 75 percent of problems had to be recycled from the solving process back into formulation. It also warns against using acceptability as the success criterion, because that is exactly when uncomfortable formulations get withheld. Limit: a theory paper with an illustrative process example, not an empirical test of any specific method.
- 3.Lafley, Martin, Rivkin & Siggelkow: Bringing Science to the Art of Strategy, Harvard Business Review, September 2012 (opens in a new tab) · Harvard Business Review · 2012-09 · practitioner source · provides the contextEstablishes that the method is jointly advanced by practice and academia: A. G. Lafley, Roger L. Martin, Jan W. Rivkin (Harvard Business School) and Nicolaj Siggelkow (Wharton), published September 2012. Limit: authorship, title and publication date were verified, but the full text sits behind a paywall and could not be inspected. The substantive weight of this entry therefore rests on the 1997 white paper, not on this article.
- 4.Roger Martin: The Evolution of the Strategic Choice Structuring Process, Medium, Juni 2025 (opens in a new tab) · Medium (Roger Martin) · 2025-06-23 · practitioner source · limits the methodQualifies the method in the originator's own words. Because successful strategies sometimes emerge despite unmet conditions, and because a condition can be made true after the fact, Martin added a dedicated assessment stage and described the sequence as a loop rather than a one-way street. Limit: a retrospective self-account with no data and no independent review.
Origin: Roger Martin