Product Validation
Build validation into the solution design workflow
Step 1Map the solution, find the hidden bets
You've picked your opportunity in the Customer workflow. Now generate solution ideas, compare them, and surface the assumptions baked into each one. Every solution has bets about desirability, usability, feasibility, viability, and ethics.
Step 1.1
Brainstorm solution ideas
Start from the opportunity. What are the different ways to solve this problem?
Step 1.2
Optimize UX in a story map
Which solution best serves the customer journey? Story map each option to see the full experience.
Step 1.3
Surface hidden assumptions
Every solution makes bets. The AI pulls out what you're assuming about desirability, usability, feasibility, viability, and ethics.
Step 2Rank assumptions, find the gaps
Not all assumptions carry the same risk. Rank them, collect the evidence you already have, and identify the ones that are high risk but low evidence. Those are the bets worth testing.
Step 2.1
Desirability assumptions
Will customers want this and do what's needed to get value from it? Testing demand alone isn't enough. You need to verify customers will engage with the friction points.
Step 2.2
Usability assumptions
Can customers find it, understand it, and do what we need them to do? Discovery, comprehension, and capability are three separate questions.
Step 2.3
Feasibility assumptions
Can we build it? Extends beyond technical capability to include legal, compliance, security, and organizational constraints.
Step 2.4
Viability assumptions
Will this create value for the business? Not just alignment with goals but whether benefits offset delivery costs.
Step 2.5
Ethical assumptions
Is there any harm in building this? Covers data practices, transparency, ideal customer decisions, and potential social or relational harms.
Step 3Test and iterate
Test the riskiest assumptions with the smallest experiment you can run today. If an assumption holds, great. If not, tweak your solution to avoid it or pick a different approach.
Step 3.1
Prototype tests
Simulate a moment in the experience and watch what customers actually do. Show a wireframe or clickable mockup. Observe real behavior instead of asking people to predict their own.
Step 3.2
One-question surveys
A single targeted question about past behavior or current intent. Deploy in-product or via email. Ask about what people already did, not what they would do.
Step 3.3
Data mining
Use data you already have to evaluate an assumption. Behavioral analytics, support tickets, search queries, sales notes. Often the fastest path to evidence.
Step 3.4
Research spikes
Engineering builds a time-boxed proof of concept. Throwaway code is fine. The goal is to answer a technical question, not ship a feature.
Start validating
You can build anything now. Make it worth using.
14-day free trial, no credit card required