Conducting user research goes beyond mere data collection; it necessitates a methodical approach encompassing the validation of hypotheses at various stages. Every product concept relies on a foundation of assumptions. It can be helpful to initially comprehensively list these assumptions and break them down into their most basic forms to be able to evaluate the product through the lenses of feasibility, desirability, and viability. 

1. Distinguish Between Macro and Micro Hypotheses

IJudd Antin, former research leader at Airbnb and Meta, defines three types of research:

• Macro: Big-picture, strategic, forward-looking research
• Middle range: Directional insights for product areas
• Micro: Tactical, usability-focused investigations

Example:

• Macro hypothesis: “Startup founders struggle to consistently prioritize product features.”
• Validate with in-depth interviews and pattern analysis (ethnography, exploratory research).
• Micro hypothesis: “Changing the label on the ‘Submit’ button will increase response rates.”
• Test this with A/B testing, heatmaps, and usability testing.

Takeaway: Start at the strategic level. Don’t waste time optimizing the wrong thing.

2. Separate Problem Space from Solution Space

Dan Olsen’s Product-Market Fit Pyramid is clear: problem validation comes before feature validation. Problem space includes your target user, their needs, and your value proposition. Solution space is about features and UX.

Example:

• Problem space hypothesis: “B2B marketers waste time switching between tools to run campaigns.”
• Validate with user interviews, behavioral shadowing, and pain point analysis.
• Solution space hypothesis: “An integrated dashboard will solve this problem.”
• Validate with clickable prototypes and usability testing.

Takeaway: Keep these two validations separate. Mixing them introduces bias and wastes effort.

‍

‍

3. Use Generative Research to Find Real Problems

According to Jared Spool, there are two kinds of research:

• Generative: Uncover problems and opportunities
• Evaluative: Test specific solutions

Start with generative research to identify what’s broken before building fixes. Use open-ended interviews, job shadowing, or customer journey mapping.

Example:
Use The Mom Test approach:

“Tell me about the last time you ran into that issue.”
“What did you do to try solving it?”

Takeaway: Don’t test solutions before deeply understanding the problem. You’ll miss what really matters.

4. Frame and Structure Hypotheses with the Right Questions

From resources like Maze and the Board of Innovation, we get structured question sets to uncover meaningful insights. Good interviews require more than casual conversations.

Example:
Instead of asking, “Would you use this?” ask:

“Can you describe step-by-step how you currently solve [X]?”
“What’s the most frustrating part of that process?”
“Have you tried to fix this problem before?”

Pair these with tools like the Problem Validation Script and Solution Validation Script for consistency.

Takeaway: Good questions expose pain. Leading questions create fiction.

5. Use the “5 Whys” to Dig Deeper

Borrowed from Toyota’s lean manufacturing, the 5 Whys technique is a powerful tool to uncover the root cause of user pain.

Example:

“Why don’t you use our notification system?”
“Because I don’t trust it.”
“Why not?”
“Because I missed a message last week.”
“Why did that happen?”
“Because the mobile app didn’t alert me.”
“Why didn’t it alert you?”
“Because I never enabled push notifications.”

Takeaway: You often need to go 5 layers deep to find the actual blocker. Surface-level problems are symptoms, not causes.

6. Prioritize Which Hypotheses to Test First

There are too many assumptions to test at once. Use prioritization frameworks like:

• Opportunity Solution Tree (Teresa Torres)
• RICE (Reach, Impact, Confidence, Effort)
• Kano Model (Delighters vs. Must-haves)
• Value vs. Complexity Matrix

Example:
Let’s say you’ve identified five potential user problems. Map them in an Opportunity Solution Tree to see which ones align with business outcomes. Then score them using RICE to find the one with the biggest upside and lowest cost.

Takeaway: Prioritize assumptions that are high-risk and foundational to your product’s success.

Borrowed from Toyota’s lean manufacturing, the 5 Whys technique is a powerful tool to uncover the root cause of user pain.

Example:

“Why don’t you use our notification system?”
“Because I don’t trust it.”
“Why not?”
“Because I missed a message last week.”
“Why did that happen?”
“Because the mobile app didn’t alert me.”
“Why didn’t it alert you?”
“Because I never enabled push notifications.”

Takeaway: You often need to go 5 layers deep to find the actual blocker. Surface-level problems are symptoms, not causes.

7. Falsify, Don’t Just Validate

Judd Antin’s mantra: “We don’t validate, we falsify.”
Your goal is to disprove assumptions, not confirm them. If a hypothesis survives scrutiny, it’s stronger.

Example:
Rather than asking users, “Would this feature be helpful?”
Run a usability test where the feature is present—but unannounced. See if users find and use it naturally. If they don’t, the hypothesis likely needs revision.

Takeaway: Look for reasons you might be wrong. Be rigorous. Be skeptical. That's where the insights live.

Final Thoughts

Hypothesis validation is not a checkbox—it’s a practice. It keeps you honest, user-centered, and focused on what matters.

Here’s how to put it all together:

• Start with macro, not micro.
• Separate problem from solution.
• Uncover root causes, don’t patch symptoms.
• Test risky assumptions first.
• Try to break your own ideas.

Do this, and you’ll dramatically increase your odds of building a product users truly want—and of reaching product-market fit smarter and faster.