Week 7: Solution ideation, prototyping and testing#
Studio abstract#
In this studio, your team will carry an evidence-supported hypothesis forward from Iteration 1, explore possible solutions, choose an appropriate prototype type and fidelity, and begin building. You will leave with a plan for completing and testing the prototype after class.
The activities total 140 minutes and begin after the 30-minute innovation seminar and a 10-minute transition.
By the end of the studio, your team should have:
selected an Iteration 1 hypothesis to carry forward;
retained several possible solutions generated through divergent thinking;
documented its chosen solution and remaining uncertainty;
justified the prototype type and fidelity;
begun building the prototype; and
assigned the work required to finish and test it before Iteration 2.
From the seminar to your project#
The seminar’s shared photo-experience example showed how the same proposed solution could be represented as paper screens, a wireframe, a realistic interface or a Wizard of Oz experience. Each version is useful for a different uncertainty. Use that example as a pattern, but make the fidelity decision for your own hypothesis and learning goal.
1. Select a hypothesis to carry forward#
⏱️ 20 minutes — team
Review your Iteration 1 hypotheses, evidence, learning cards, business model canvas and feedback. Select one hypothesis that is sufficiently supported and important to carry into solution development.
Record:
the hypothesis in its current form;
the customer segment and problem it concerns;
the evidence that supports carrying it forward;
important contradictions, limitations or uncertainty; and
why the team selected it over its other hypotheses.
Do not quietly discard inconvenient evidence or rewrite the Iteration 1 record. Add a new decision that links back to the existing evidence trail.
2. Diverge: create options#
⏱️ 20 minutes — individual, then team
Begin with the selected hypothesis and generate possible solutions or value propositions.
For the first 10 minutes, work individually and aim for quantity rather than polish. Record every idea, including unusual or impractical ones. Do not evaluate ideas yet.
For the next 10 minutes, share the ideas and combine related possibilities. Preserve the full set in the project record rather than retaining only the eventual choice.
3. Converge: make a choice#
⏱️ 15 minutes — team
Compare the options using the evidence already gathered. Consider:
how directly each option responds to the selected customer problem;
which important assumptions remain untested;
what could be learned before implementation;
feasibility within the course; and
whether the option offers a meaningful value proposition.
Choose one direction to prototype. Record why it was selected and which alternatives were set aside for now.
4. Choose the prototype type and fidelity#
⏱️ 15 minutes — team
Write one clear statement of what the team needs to learn about the chosen solution. Then choose the simplest prototype capable of producing useful evidence.
Possible approaches include:
Approach |
Useful when the team needs to learn about |
|---|---|
Sketch or paper prototype |
concepts, screens, sequence and basic interaction |
Wireframe or clickable prototype |
navigation, information structure and task flow |
Wizard of Oz or simulated service |
an experience whose underlying system is expensive or unavailable |
Code prototype |
technical feasibility or an interaction that cannot be represented credibly another way |
Optional tools#
For a mid-fidelity wireframe or clickable flow, teams may use Figma or UXPilot. Paper, presentation slides or another familiar tool are also suitable—the tool should match what the team needs to learn.
If a high-fidelity interaction is justified, teams may use an AI coding agent with an appropriate interface-prototyping or frontend-design skill. Treat the generated result as a prototype: review and test the output, remove unnecessary functionality, retain meaningful team contributions in the repository, and never provide the agent with identifiable participant data.
Decide what must appear realistic, what must work, and what can remain absent or simulated. More polish is not automatically better.
Break#
⏱️ 10 minutes
5. Plan the prototype#
⏱️ 10 minutes — team
Before building, record:
the hypothesis and uncertainty the prototype will address;
the type and fidelity selected, with a short justification;
the specific parts that must be represented or functional;
the parts that can be omitted or simulated;
the task a participant will attempt; and
how observations and feedback will be recorded safely.
Divide the build into small tasks and give each task one accountable owner.
6. Begin building#
⏱️ 40 minutes — team
Build the smallest useful version of the planned prototype. Review it regularly against the learning goal rather than adding detail simply because there is time.
Keep evidence of the work in the project repository. Depending on the approach, this may include photographs, exported screens, source files, code, design decisions and links to externally hosted artefacts. Do not commit participant-identifying material.
The prototype does not need to be complete by the end of the studio.
Exit ticket and next steps#
⏱️ 10 minutes — team and legend
Show your legend:
the selected Iteration 1 hypothesis and supporting evidence;
the solution options generated and the reason for the final choice;
the stated uncertainty and chosen prototype fidelity;
the prototype work completed so far; and
GitHub Project tasks assigning who will finish the prototype, who will test it, and when each task is due.
After the studio, finish only what is necessary for the planned test. Test the prototype with relevant people, record observations and feedback, and connect the resulting learning to the hypothesis and next project decision for Iteration 2.