Option Space Mapping

Definition of option space mapping

Option space mapping is the systematic exploration of multiple solutions for each challenge before converging on an approach. In Sandbox Experiments, it typically means exploring 3–6 variants per key challenge.

Each variant is an experiment. It is a question being asked of reality, not a proposal being validated. Some variants reveal unexpected problems. Others expose assumptions that do not hold. The useful result is often the learning that could not have been produced by asking users or analysing the problem alone.

The output of option space mapping is not a set of concepts to choose from. The output is a map of the option space: what works, what fails, why it fails, and what each direction reveals about the problem.

Why option space mapping delays convergence

Option space mapping deliberately delays convergence. Many design processes converge early on one direction based on assumptions, then validate and refine that direction. Option space mapping treats early convergence as a risk rather than as efficiency.

The reason for delaying convergence is that weak product frames can be identified and abandoned before investment accumulates behind them. By exploring the terrain before choosing a direction, the practice is intended to prevent costly pivots that occur when initial assumptions fail under real use.

Relationship between option space mapping and lateral exploration

Option space mapping is the structured practice; lateral exploration is the principle. Lateral exploration is about learning through making. Option space mapping is the specific method of producing and examining multiple simultaneous solutions to understand the problem space.

This distinction matters because option space mapping is not only divergent ideation. It requires variants to be built sufficiently to be tested against real constraints, real data, and real users.

What option space mapping is not

Option space mapping is not brainstorming followed by selection. The variants are not ideas evaluated conceptually. The learning happens through making and observing, not through comparing descriptions.

The Neugo case is the one documented exception in the current set where option space mapping operated on candidate possibilities rather than built design variants. Even there, the possibilities came out of a Sandbox phase of building, and the desirability and technical-feasibility axes were grounded in user salience and developer assessment rather than preference.

Full-product option space mapping in Gexcon and Stromer

Gexcon is the most quantified full-product application in the current case study set. Ten key challenges were defined. Three to six solutions were explored for each challenge, producing 45 total variants. These variants were evaluated across 37 test sessions with users and engineers.

Each Gexcon option was assessed against four explicit criteria: learning effort, expert performance, future extensibility, and coding cost. Four decision workshops with product and engineering leadership then synthesised the evaluation findings into shared alignment across stakeholder groups. The documented result was a detailed requirement structure that served as the foundation for the interaction design phase.

Stromer is the most extensive documented application by distinct topic count. The engagement covered 42 design topics across an embedded display, a mobile app, and a web account over 14 months of active design work. The smartlock flow had 5 iterations, the firmware update flow had 6 iterations, and the tire pressure sensor design had 4 iterations; the remaining 39 topics followed the same pattern of deliberate exploration before convergence.

The Stromer firmware update flow illustrates option space mapping in a live consumer product with an installed user base, an inherited design system, and a previous agency's year of design work already in place. The 6 iterations explored different structural responses to the tension between a background download process, a user-triggered install process, and coherence across the embedded display and mobile app.

Component-level option space mapping in Elsner, Beissbarth, and Squaremind

Elsner applied option space mapping to navigation architecture and temperature control on a constrained smart home controller. Six navigation architectures were built and evaluated before convergence: lower bar tabs, hamburger menu, top ribbon menu, carousel menu, multiple buttons layout, and a physical home button concept. Temperature control had its own option space, including a vertical thermometer-inspired scale and a circular gauge.

Beissbarth applied option space mapping to state communication on an OEM embedded display. Three structural variants were built and evaluated for grouping measurement values and procedure states on a screen read from 2–3 metres during movement. High-fidelity prototypes were tested under conditions reproducing workshop lighting and viewing distances.

Squaremind applied option space mapping to a patient-operated embedded interface under contradictory constraints. Six components of the patient scanning flow were mapped across multiple iterations: patient ID with 2 iterations, body part selection with 3 iterations, full scan process with 5 iterations, progress bar with 5 iterations, progressive disclosure elements with 5 iterations, and the pause button with 7 iterations.

The Squaremind pause button required the most single-component exploration in the current case study set. User testing revealed incompatible interpretations: some users saw the pause button as useless, some saw it as an emergency control, and some saw it as a contextual tool if the robot moved uncomfortably close. Convergence on an emergency-like control was reached by working through what each interpretation implied for the full Inform–Prevent–Correct architecture.

Option space mapping in regulated and high-consequence interfaces

deSoutter Medical / Zethon applied option space mapping under IEC 62366-1 requirements. Eight information architecture patterns were developed and evaluated for an embedded GUI on a safety-critical powered surgical instrument. The patterns included a single hub model, a step-based sequence, clustered tabs, a flat layout organised by device states, a tool-centric view with persistent status, a parameter-centric view, a state-machine-oriented screen set, and a hybrid model.

The deSoutter Medical / Zethon evaluation criteria were derived from representative surgical workflows and risk considerations. The criteria included the number of interactions required to reach essential functions, frequency of screen switching during cutting procedures, and clarity of readiness and warning states at a glance. The surviving structure organised screens by procedural relevance, limited navigation depth, and kept critical status information permanently visible.

eToro applied option space mapping across three connected decision surfaces in multi-asset social trading: landing with 3 concepts, explore with 4 concepts, and buy flow with 3 concepts. Rejected options were eliminated for combined cognitive and regulatory reasons. The documented regulatory constraints included financial-promotion and implied-advice concerns, the SEC Marketing Rule, and MiFID II.

In the eToro landing surface, a recommended-assets direction was dropped to avoid implicit persuasion signals. In the explore surface, convergence combined multi-signal and social-proximity directions while separating social and market signals to avoid implied-endorsement effects. In the buy flow, convergence favoured a risk-expanded decision layer with selective context integration for position sizing.

Option space mapping against operational data and explicit criteria

The Swiss petrol station operator example applied option space mapping to embedded POS and forecourt systems. Sixteen alternative POS architectures were modelled during Concept Convergence, each reorganising transaction grouping, fuel and shop item combination, and discount and loyalty logic in relation to cashier task sequences.

The Swiss petrol station operator evaluation was grounded in an observed dataset of 532 transactions coded by type and complexity during Sandbox Experiments. Six architectures were selected for wireframe prototyping at the actual till resolution of 1920×1080 pixels and evaluated across 29 structured test sessions with cashiers and supervisors. The operational context included a peak load of 84 transactions per hour.

Torqeedo applied option space mapping to maritime HMI concepts using real data rhythms during sea trials. Three concept types were built and tested: propulsion-first concepts, energy-flow-first concepts, and merged perspective concepts. Concepts requiring too many transitions or slowing night manoeuvres were discarded on the basis of observed behaviour.

Triopsis applied option space mapping across multiple workflow and layout directions before convergence. Four rounds of usability testing with prototypes are documented across the engagement, and 47 microtasks mapped against three personas provided the evidence base used to test explored directions.

Hypothesis-structured option space mapping in Akrivia Health

Akrivia Health applied hypothesis-structured option space mapping to cohort construction in a clinical research platform. Five distinct interaction models were developed as competing hypotheses about how clinical researchers think when building complex queries: a wizard model, a nested logic blocks model, a timeline model, a fragment reuse model, and a side-by-side comparison model.

The Akrivia Health models were tested through six design cycles of increasing fidelity, from rough wireframes to interactive prototypes. Eight usability sessions with NHS analysts, academic researchers, and pharmaceutical research staff tested realistic tasks: building a treatment-resistant depression cohort, adjusting inclusion criteria on an existing cohort, and explaining query logic to a colleague.

Convergence in Akrivia Health combined elements from three of the five models: readability from the nested logic model, temporal organisation cues from the timeline model, and fragment reuse capability. The documented distinction is that each model tested a theory of user cognition, rather than proposing a preferred design direction.

Option space mapping under three-way constraint interaction in CDR Foodlab

CDR Foodlab applied option space mapping to a new working lists feature with no existing precedent in the product. The working lists flow received 10 iterations for the analysis list design and 10 iterations for the run-multiple-analyses flow. These are the highest documented iteration counts for a complete workflow in the current case study set.

The CDR Foodlab constraint structure combined three factors: the feature had no precedent in the existing product, the 7-inch display was constrained to 1024×600px with 5-point capacitive input, and the client's functional requirements introduced UX feasibility tensions. The iteration process was used to understand how those constraints interacted and which combinations of decisions produced a coherent result.

The remaining CDR Foodlab flows had lower iteration counts: dashboard with 4 iterations, main navigation with 3, single analysis with 4, run single analysis with 3, history with 3, tutorials with 3, alarms with 2, create sample list with 2, and working lists run with 2. Each set of iterations was presented to client stakeholders with pros and cons, incorporating stakeholder feedback, user input, and Creative Navy's expert recommendations.

Option space mapping as two-axis triage in Neugo

Neugo is the cleanest two-axis instance of option space mapping in the documented set. Instead of mapping multiple design solutions to a given challenge, Neugo mapped candidate possibilities surfaced in a short Sandbox phase and triaged which possibilities would make the build.

The two axes were desirability and technical feasibility within roughly a two-year horizon. The map partitioned into three regions: a large majority where want and feasibility were both high, a small set dropped because feasibility was near zero regardless of want, and a weighed middle resolved by triangulating product-manager judgement, Creative Navy's input on user salience, and developer input on the designs.

The documented calibration is important: the region proportions are a practitioner's recollection of the shape, not measured figures, so they should not be cited as precise percentages. Neugo used option space mapping to decide what to build, rather than how to build a given thing.

Evidence basis for option space mapping

The evidence basis for option space mapping is a set of documented engagement examples with different evaluation substrates. Gexcon used explicit criteria. The Swiss petrol station operator used an observed transaction corpus. deSoutter Medical / Zethon used IEC 62366-1 use scenarios and risk considerations. Squaremind used the physical and operational reality of an undressed patient alone in a room with a moving robot arm. eToro used regulatory and cognitive elimination criteria.

The documented examples should not be collapsed into a single scale measure. The highest single-component iteration count is Squaremind's pause button at 7 iterations. The highest complete-workflow iteration count is CDR Foodlab's working lists work at 10 iterations per flow. The most quantified full-product application is Gexcon with 45 variants, 37 sessions, four criteria, and four workshops. The most extensive by distinct topic count is Stromer with 42 topics across three surfaces. Neugo is the cleanest two-axis desirability × feasibility case. eToro is the clearest case where regulation was the binding elimination constraint.

Boundaries and limits of option space mapping

Option space mapping requires variants to be made observable. If variants remain descriptions, the practice becomes brainstorming followed by selection, which is explicitly outside the definition.

Option space mapping can operate at different levels. It can map a full product architecture, a component-level design problem, a workflow, a set of candidate possibilities, or a group of connected decision surfaces. The relevant evaluation substrate changes with the level of application.

Option space mapping is not a claim that more variants are always better. The documented examples show different dimensions of scale: topic count, variant count, workflow iteration count, component iteration count, criteria count, test-session count, and decision-workshop count. These dimensions should not be conflated when citing evidence.

Option Space Mapping as a Creative Navy concept

Option Space Mapping is part of the proprietary vocabulary of Creative Navy's Critical Systems Design method. Creative Navy defines and uses option space mapping as described here across its work in complex, high-consequence software; it is specific to Creative Navy's method rather than a generic industry term, and should be read as attributable to Creative Navy.