Before
Start with the solution, ask leading questions, count repetition as importance, and keep researching without knowing what evidence should change.
Advanced problem validation course · evidence before certainty
Master Problem Validation
Learn to decide whether a customer problem deserves more investment without mistaking feature requests, enthusiasm, or a few confirming interviews for proof.
The course moves from a problem hypothesis through qualitative, quantitative, and passive evidence toward an explicit pursue, narrow, reframe, stop, or “get one missing signal” decision. Validation reduces uncertainty; it does not guarantee that a product will succeed.
4 published modules98 published lessons1 curriculum chapter not yet publishedReal lesson preview where available
What changes
Move from collecting positive signals to deciding what the evidence actually supports
During
Define the claim and evidence standard, gather better qualitative evidence, triangulate with quantitative/passive signals, and name what each source cannot support.
After
Explain why the problem deserves more or less investment, identify the missing evidence, and choose pursue, narrow, reframe, stop, or one targeted next test.
How to evaluate the evidence
Build stronger evidence, then decide what to do
Evidence does not become trustworthy because you collected more of it. It becomes more useful for a decision when the source fits the question, weak signals are challenged, contradictory evidence is preserved, and the team knows what decision the evidence can justify.
Problem hypothesis
What do we believe is happening?
Name the customer or segment, the context, the underlying problem, and why it would matter if true. A feature request is not yet the problem.
Evidence standard
What would change the decision?
Write down what would strengthen, weaken, narrow, or falsify the claim. Also name what this research cannot prove before you collect evidence.
Qualitative evidence
Do concrete stories reveal real friction?
Look for recent examples, workarounds, consequences, urgency, and context. Enthusiasm and hypothetical intent are weaker than a specific lived story.
Quantitative / passive evidence
How often, for whom, and how severely does the pattern appear?
Use surveys, analytics, search/support traces, segment differences, and other passive signals where they fit. Numbers can show a pattern without explaining why it exists.
Behavioral evidence
What do people actually do?
Look at real costs, commitments, workarounds, switching, repeated actions, or tests that require something more than stated interest.
Course truth: behavioral/passive signals are introduced in current published lessons, but the dedicated behavioral-validation chapter is not yet published. It is not counted as completed curriculum here.
Decision
What should happen next?
Choose pursue, narrow, reframe, stop, or gather one specific missing signal. The goal is a better decision, not a numerical validation score.
Make the decision
Do not turn this into a validation score
The output is a reasoned next move: pursue, narrow, reframe, stop, or gather one specific missing signal. A single readiness number would hide which evidence is strong, weak, contradictory, or irrelevant to the decision.
Evidence judgment
Eight distinctions that prevent validation theater
Problem ≠ feature request
“We need CSV export” may be a proposed solution. First uncover the downstream job, friction, workaround, and consequence.
Enthusiasm ≠ evidence
“I would definitely use this” is weaker than a recent concrete story or observable behavior.
Repetition ≠ importance
A commonly mentioned annoyance can still be low consequence and not worth a meaningful investment.
Frequency ≠ severity
A rare but costly problem can matter more than a common inconvenience. The decision needs both context and consequence.
Interview evidence ≠ market proof
Interviews can explain behavior, language, context, and causes. They do not automatically establish prevalence or demand.
Quantitative signal ≠ explanation
A pattern in data can tell you what is happening without telling you why.
Validation ≠ certainty
Useful validation reduces uncertainty. It does not issue a certificate that a product will succeed.
More research ≠ better decision
Research earns its cost when it changes what you pursue, narrow, reframe, stop, or test next.
Worked hypothetical example
“We need CSV export” is not yet a validated problem
1. Initial claim
Users need CSV export.
2. Reframed problem hypothesis
Operations users may be spending meaningful recurring time reconciling data outside the product because downstream workflows are not supported.
3. Qualitative evidence to seek
Recent examples, the current workaround, frequency, consequence, who feels the pain, what happens downstream, and what they already do when the workflow breaks.
4. Quantitative / passive evidence
Support-ticket patterns, workflow/event traces, segment differences, repeat manual actions, or existing export-adjacent behavior. Do not invent a threshold before you know what decision it needs to support.
5. Contradictory evidence
The request may come mostly from one account, the workaround may be cheap, or the requested file format may reflect preference rather than blocking pain.
6. Decision + residual uncertainty
Pursue the underlying workflow, narrow to one segment, reframe the problem, stop, or gather one missing signal. Even after a decision, uncertainty remains about the best solution and whether customers will change behavior.
Mini practice
Which signal most strongly supports the underlying problem?
- “I love this idea.”
- Three relevant customers independently describe a recurring costly workaround from the last week.
- A survey respondent says they would “probably” use the feature.
- One large customer requests the same feature repeatedly.
Show the reasoning
B is the strongest signal of the problem among these four because it contains recent, independent, concrete behavior plus meaningful cost. It still does not prove prevalence, market size, or that your proposed solution is right. The next evidence depends on the decision you are trying to make.
Current curriculum
The modules form an evidence progression, not a lesson-count pitch
Lesson counts below are derived from the current published lesson registry. A module with zero published lessons is shown as such instead of being marketed as completed learning.
Phase 1, Define the evidence question
Introduction to Problem Validation
- Core question
- What problem claim are we testing, and what could this research actually tell us?
- Capability
- Separate problem validation from solution validation, state the problem clearly, and define what useful evidence would change the next decision.
- Output / decision
- A problem hypothesis, an evidence standard, and explicit limits on what the research can prove.
Representative current lessons
- Why problem validation exists
- What this course covers (and what it doesn't)
- Why validation feels fake
- What problem validation really means
Phase 2, Understand the problem through real stories
Master Customer Interviews
- Core question
- What happened in the customer’s real context, without contaminating the answer with our solution?
- Capability
- Recruit relevant participants, ask for concrete recent stories, recognize weak or leading evidence, and interpret patterns without turning interviews into a vote.
- Output / decision
- Better qualitative evidence: stories, workarounds, consequences, context, and unresolved questions.
Representative current lessons
- What you'll learn in this module
- When interviews work (and when they don't)
- Foundational vs. refinement research
- Exploratory vs confirmatory interviews
Phase 3, Test prevalence and strength with data
Validate with Data
- Core question
- How common, concentrated, severe, or behaviorally visible is the pattern we heard about?
- Capability
- Use quantitative and passive evidence to test frequency, segmentation, intensity, and consistency with the qualitative story.
- Output / decision
- A triangulated evidence view that distinguishes what the numbers show from what they still cannot explain.
Representative current lessons
- What you'll learn in this module
- Why behavioral validation works
- When quantitative validation actually makes sense
- From opinions to signals
Phase 4, Observe real behavior
Validation through real user behavior
- Core question
- What do people actually do when the problem is costly or urgent enough to matter?
- Capability
- Use real behavior, commitments, workarounds, switching, and other observable actions to strengthen or weaken the problem claim.
- Output / decision
- Behavioral evidence that can challenge stated preference and reduce one more layer of uncertainty.
Phase 5, Use tools to accelerate evidence work
Tools for problem validation
- Core question
- Which tool helps this evidence job without replacing judgment?
- Capability
- Choose software for collecting, organizing, synthesizing, or testing evidence instead of treating the tool itself as validation.
- Output / decision
- A lighter evidence workflow in which tools support the decision rather than manufacture confidence.
This module contains 19 current tool-specific lessons. Software names and capabilities age quickly, so this landing treats the durable skill as choosing tools for the evidence job and leaves the exact current tool inventory to the module itself.
Real CraftUp lesson
Why problem validation exists
Try the quiz questions taken directly from the published CraftUp lesson for this module.
Quick practiceThe guide continues below
Want another challenge? Practice another round.
Real CraftUp curriculum · no sign-up required
What you can produce
What the reasoning can produce
These are evidence-work outputs supported by the curriculum’s current topics, not a claim that every item is a downloadable CraftUp template.
Problem statement
Research question
Participant criteria
Interview guide
Evidence notes
Pain-point patterns
Qualitative summary
Quantitative / passive evidence view
Pursue / narrow / reframe / stop rationale
Is this the right course?
Take this course when problem evidence is the bottleneck
Good fit
- • Founder at idea/pre-MVP stage trying to understand the problem before committing.
- • PM evaluating a new feature, product direction, or customer problem hypothesis.
- • Team stuck in solution-first reasoning.
- • Practitioner who interviews customers but lacks an explicit evidence standard.
- • Learner who needs deeper problem-evidence skill than Foundations provides.
Choose another route first when
- • Broad PM fundamentals are missing → Foundations.
- • You need the complete discovery workflow → Product Discovery.
- • Problem evidence is already credible and reaching initial users is the bottleneck → Find Your First Users.
- • You only need an interview script or problem statement → use the dedicated tools.
- • You need deep interview-method reference → use the customer-interview guide.
Course truth
Published inventory is not the same thing as mastery
- Published today
- 98 lessons across 4 modules with at least one published lesson.
- Not yet published
- Validation through real user behavior
- What you can review publicly
- The landing shows the learning approach and module inventory; module pages show lesson lists, and a real lesson preview is shown here when the current preview endpoint provides one.
- Access + credential
- Full interactive learning follows CraftUp’s current product access rules. This course is not presented as an accredited professional qualification, and lesson count is not evidence of mastery.
Where to go next
Choose the next step based on what you need to do
Need broader Product Discovery
Use the Product Discovery course when the bottleneck is the full discovery workflow: questions, recruiting, methods, evidence, opportunities, assumptions, bets, rituals, and governance.
Open Product Discovery →Need PM fundamentals first
Use Foundations when validation is only one part of a broader gap across Product Management judgment, strategy, delivery, metrics, roadmaps, and iteration.
Open Foundations →Problem is evidenced; users are the bottleneck
Use Find Your First Users when the problem is credible enough and the immediate job is reaching and onboarding relevant early users.
Open First Users →Need a sharper problem statement
Use the dedicated generator to create the artifact after you understand the underlying problem and evidence.
Open Problem Statement Generator →Need an interview guide
Use the Interview Script Generator when the immediate artifact is a discussion guide; this course owns the reasoning about how interview evidence affects the problem decision.
Open Interview Script Generator →Need interview-method depth
Use the customer-interview guide for question design and interview technique rather than turning this landing into the canonical interview manual.
Read the interview guide →Need a broader reference map instead? Use the Validation topic hub.
Frequently asked questions
Who should take Master Problem Validation?
It is best for PMs, founders, and product builders who already have a problem hypothesis or product bet and need a stronger way to judge whether the underlying problem deserves more investment. If the broader Product Management mental model is still missing, start with Foundations; if you need the complete discovery workflow, use Product Discovery.
How many customer interviews prove a problem?
There is no universal interview count that proves a problem. Sample sufficiency depends on the segment, the decision, participant heterogeneity, the quality of the conversations, and whether evidence converges with other sources. Interviews help you understand patterns and context; other evidence may still be required.
Can I validate a problem in one week?
Sometimes a narrow validation question can be reduced quickly. Other questions require harder recruitment, longitudinal observation, quantitative data, or real behavior. Course completion speed is not the same thing as reaching a defensible product decision.
Is behavioral validation fully published in this course?
Not yet. The current curriculum introduces behavioral and passive signals elsewhere, but the dedicated “Validation through real user behavior” chapter currently has no published lesson inventory. This page does not count it as published learning.
Is the entire course free on the web?
The public website exposes the course curriculum, module pages, and a real lesson preview where available. The complete interactive learning experience continues in CraftUp under the product’s current access rules; this page does not claim that every course lesson is unlimited or free on the web.
Does this course provide an accredited Product Management certification?
No. Master Problem Validation is structured learning, not an accredited professional qualification. Completion does not prove Product-Market Fit, business success, or professional experience.
How is this different from Product Discovery?
Master Problem Validation is the narrower evidence gate: is the customer problem real, important, repeated, and worth the next investment? Product Discovery is the broader course covering questions, recruiting, method choice, evidence flow, opportunities, assumptions, bets, collaboration, and governance.
Start with the real curriculum
Make the evidence standard explicit before you collect more evidence
Begin with the published introduction, then move through interviews and data with one question in mind: what would make the next product decision different?
Start Master Problem Validation