Problem-First Research

Introduction

Most innovation fails for the same reason: premature commitment to a solution.

• Startups pivot endlessly because they built before they understood.
• Corporate R&D departments chase technology-led roadmaps and shiny objects that ignore the structure of the problems they claim to solve.
• Policy interventions backfire because the problem was framed as a symptom rather than a cause.

The pattern is universal, and it is full of waste. Problem-first research inverts this default sequence trap.

Where conventional innovation follows Idea, Build, Test, Fail, Pivot, problem-first innovation follows Understand, Investigate, Map, Design. The operative principle: you cannot solve a problem you do not understand, and you cannot understand a problem without understanding its history.

This is not a single school of thought. It is a convergent methodology emerging independently across design (Stanford d.school’s Empathize-Define phases, the British Design Council’s Double Diamond), engineering (DARPA Grand Challenges, Canadian Engineering Grand Challenges), policy (wicked problems theory from Rittel and Webber), and commercial innovation (Larry Smith’s Problem Lab at the University of Waterloo, XPRIZE’s solution-agnostic challenge design). These traditions arrive at the same conclusion through different paths: deep problem understanding is the highest-leverage activity in any innovation process.

Practitioners who master this methodology gains a structural advantage. A well-researched problem is itself a strategic asset: it defines the parameters of any viable solution, eliminates entire classes of doomed approaches, and becomes the foundation for an Innovation Challenge that mobilizes diverse solvers toward measurable outcomes.

The Four Dimensions

The most systematic articulation of problem-first methodology comes from the University of Waterloo’s Problem Lab, founded by Larry Smith. His framework decomposes problem understanding into four interdependent dimensions, each building on the previous.

Scale asks: who is affected, how many, and how critical is this problem to them? A problem worth deep investigation must be mission-critical to its stakeholders, not merely inconvenient. Scale analysis maps the affected population, estimates economic magnitude, determines relative priority (is this their first-order problem or their thirty-third?), and identifies whether the problem is business-facing or consumer-facing. The scale dimension prevents researchers from wasting rigour on trivial problems and ensures the resulting design brief targets genuine leverage.

Context asks: what is the causal structure? A problem is either primary (it causes other problems) or secondary (it is caused by other problems). Primary problems offer higher leverage because solving them cascades benefits downstream. Context analysis maps root causes, environmental conditions (regulation, technology shifts, competitive dynamics), interdependencies with adjacent systems, and boundary conditions that define the problem space. The critical output is a causal chain: upstream factors leading to this problem leading to downstream consequences.

History asks: how has this problem evolved through time? This is where problem-first research diverges most sharply from solution-first approaches. Smith’s directive is unambiguous: “Unless you understand the history of the problem, you do not understand it.” Historical analysis is not background reading performed for academic completeness. It is the primary analytical method. The researcher constructs a temporal narrative documenting when the problem emerged (distinct from when it was recognised), how its scale has changed, how those affected have shifted, how causes and effects have mutated, and how environmental conditions have altered the problem’s nature. The problem timeline is the analysis, not an appendix to it.

Failures asks: what has already been tried, and why did it fail? This is the most important dimension and should consume forty to fifty percent of total research effort. Smith’s key intellectual contribution is the reframing of failure from a defensive concern (“learn from mistakes to avoid them”) to a generative tool (“study mistakes to find your starting point”). He calls this the Actionable Mistake: a specific error from a past failed attempt that reveals not just what not to do, but where to begin with a new approach.

The actionable mistake must be specific. “They did not do enough research” is not actionable. “They failed to use available cross-disciplinary information from materials science, applying only their domain-native chemical engineering lens” is actionable. Smith identifies over twenty categories of failure, grouped into information failures (ignored available data, disregarded cross-disciplinary insight, believed information without verification), cognitive failures (leaped to conclusions, overvalued comfortable information types), and methodological failures (poorly designed experiment, never replicated apparent success, used inappropriate tool). Each category points toward a different starting condition for new approaches.

The generative orientation of failure archaeology distinguishes problem-first research from preventive failure analysis frameworks like pre-mortems, root cause analysis, or failure mode effects analysis. Those frameworks produce constraints: “do not repeat X.” Problem-first failure analysis produces direction: “begin HERE, because past failure X reveals gap Y.”

The Temporal-Futures Extension

Classical problem-first methodology, including the Problem Lab’s framework and most design thinking implementations, terminates at historical analysis. Once the problem timeline and failure record are documented, the researcher moves to solution ideation. This is a missed opportunity. Historical analysis produces a trajectory. That trajectory can be projected forward with the same rigour applied to the past.

We extend problem-first research into futures studies through a four-type scenario projection framework:

Actual documents the factual timeline as it occurred. This is the historical analysis output: verified events, confirmed failures, measured condition changes. It carries the highest confidence (0.92 to 1.0) and serves as the baseline against which all other scenarios are evaluated.

Alternative History asks: what if a specific past failure had succeeded instead? This is the counterfactual validation step. If changing one historical failure would not have materially altered the outcome (because other blocking failures existed upstream), then that failure is not truly actionable. The right actionable mistake is one where the alternative history diverges significantly: “If THIS had been done differently, the trajectory would have materially changed.” Alternative histories reveal sensitivity and help distinguish decisive failures from inevitable ones.

Forecast asks: given current trajectories, what happens if nothing changes? This requires identifying which conditions are currently shifting (technology maturation curves, regulatory momentum, demographic transitions) and projecting the problem’s evolution through those changes. A forecast uses the same evidentiary standard as historical analysis: project from documented trends, not from speculation.

Speculative Future asks: if the actionable mistake is corrected, what becomes possible? This is the design fiction component. It treats the correction of the identified failure as a “supposed future” and works backward to determine what conditions, resources, and partnerships would be required. The speculative future produces the Scenario Gate: the specific condition that must be true for a new attempt to succeed where previous attempts failed.

This extension transforms problem-first research from a retrospective methodology into a full temporal analysis spanning past evidence through present conditions into projected futures. The problem is understood not as a static object but as a temporal phenomenon with momentum, path dependencies, and multiple possible trajectories.

From Problem to Brief

The four dimensions and temporal extension produce a large volume of analysis. The Design Brief is the compression format that makes this analysis actionable. A design brief is a structured problem articulation template that forces solvers to define the problem on its own terms without prescribing solutions. It is solution-neutral by construction.

A problem-first design brief contains:

• Context: One-paragraph problem statement situating the problem in its domain and explaining why it matters now.
• Scale: Who is affected, how many, how critical. The economic and social magnitude of the pain.
• Causal Structure: Whether the problem is primary or secondary. The upstream causes and downstream consequences.
• Temporal Arc: When the problem emerged, when it was recognised, its current phase, and its trajectory (growing, stable, or shifting).
• Failure Record: The top three to five past attempts with specific reasons for failure, classified by failure type.
• Actionable Mistake: The single most generative failure that becomes the launch point for new approaches.
• Scenario Gate: The condition that must be true for a solution to succeed this time, derived from alternative history and speculative future analysis.
• Constraints: Non-negotiable boundaries that any solution must respect.

The brief defines the parameters and qualities of any successful solution attempt without specifying what that solution should be. It tells solvers: “Here is what success looks like. Here is what has already been tried and why it failed. Here is where to begin. Here are your boundaries. Now solve it your way.”

This is the bridge between understanding and action. A problem that has been researched to this depth does not need to be solved by the researcher. It can be presented as an Innovation Challenge: a structured call for diverse solvers to apply their unique expertise toward a rigorously specified outcome.

From Brief to Challenge

A design brief produced through problem-first research is a strategic asset. It represents weeks or months of temporal analysis, failure archaeology, and scenario projection compressed into a document that any capable solver can act upon immediately.

This asset enables a specific format for mobilising action: the Innovation Challenge. Where a design brief defines the problem, an Innovation Challenge presents that problem to diverse solvers with evaluation criteria, timelines, and incentive structures that reward the most viable approaches. The Compelling Question framework complements this by fracturing conventional thinking about the problem: fusing bold ambitions with significant constraints to force lateral solutions.

The integration is deliberate. Problem-first research produces understanding. The design brief compresses it. The Innovation Challenge deploys it. Each layer builds on the previous, and each references the same underlying temporal analysis.

References

• British Design Council. The Double Diamond Framework.
• Engineers Canada. Canadian Engineering Grand Challenges. May 2022.
• HeroX. Open Innovation Challenge Design.
• Morgan, Adam, and Mark Barden. A Beautiful Constraint: How to Transform Your Limitations Into Advantages, and Why It’s Everyone’s Business. Wiley, 2015.
• Rittel, Horst W. J., and Melvin M. Webber. “Dilemmas in a General Theory of Planning.” Policy Sciences 4, no. 2 (1973): 155-169.
• Smith, Larry. Problem Lab Methodology. University of Waterloo, Centre for Business, Entrepreneurship and Technology. https://uwaterloo.ca/entrepreneurship/problem-lab/our-methods
• XPRIZE Foundation. Grand Challenge Design Methodology. https://www.xprize.org

Cross-links

• Compelling Question
• Innovation Challenge