Stop Thinking, Start Discovering: Validate Your Idea in 3 Steps

The AI-powered playbook that turns ideas into revenue-validated evidence

I’ve watched hundreds of non-tech founders make the same mistake: they either research themselves into paralysis or build themselves into bankruptcy.

The ones who research endlessly? They spend months perfecting market analysis, competitor deep-dives, and persona workshops. They know everything about their market, except whether anyone will actually pay them money.

The ones who build immediately? They skip validation entirely, spend six months coding (or hiring developers), then discover their “revolutionary” idea already exists or solves a problem no one cares about.

Both approaches kill startups.

The founders who succeed do something different: they discover what customers actually need through real conversations, then validate demand before writing a single line of code. They turn hunches into evidence in days, not months.

This isn’t about thinking less. It’s about discovering more. Moving from assumptions in your head to evidence from real customers. Every step in this playbook has one purpose: to gather concrete proof that people will pay for what you’re building.

This guide shows you how to leverage AI to compress what used to take expert researchers weeks into what you can do in 2-3 days. Then you’ll spend 5 days discovering whether real customers will actually pay you.

How to Use AI Research (Without Getting Burned)

AI tools like ChatGPT, Claude, and Perplexity can significantly accelerate your research. But they can also lead you astray if you’re not careful.

Critical rules for AI-powered research:

1. Always verify top claims. Pick 3-4 specific facts from any AI output (competitor pricing, market size, user complaints) and manually verify them. Visit actual websites, read real reviews, and check pricing pages.

2. Use retrieval-based AI when possible. Tools like Perplexity or ChatGPT’s search mode cite actual sources. This makes verification easier and reduces the likelihood of hallucinations.

3. Treat outputs as 80% drafts. The AI gives you a strong starting point, not the final answer. Your job is to refine, verify, and add the 20% that comes from real-world validation.

4. Push back on vague responses. If the AI says “various spreadsheet tools,” respond with: “Name specific companies with exact URLs and current pricing.” Good research requires good follow-up questions.

5. Watch for “too perfect” answers. If every competitor has neatly categorized strengths and weaknesses, or all user complaints fit into clean buckets, the AI is probably guessing. Real markets are messier.

Which AI tool should you use?

• ChatGPT (with search enabled) or Perplexity: Best for competitive research, user frustrations, and market trends because they can search current information

• Claude: Best for analyzing uploaded documents, synthesizing information, and working with structured data

• Any of the above: Fine for brainstorming, positioning, and structured prompts

The 3 Steps to Evidence

Here’s what we’re building toward: by the 3rd step, you’ll have concrete evidence that people will pay you money to solve their problem. Not assumptions. Not hopeful projections. Actual evidence from real customers.

That’s your validation. Everything else is just thinking without discovery.

Step 1, Find Your Wedge: Identify the specific problem and customer segment where you can win.

Step 2, Define Your Target: Create a detailed profile of your ideal first customer.

Step 3, Get Paid: Find 3 people willing to pay you before you build anything.

Let’s begin.

Step 1: Find Your Wedge

Most founders confuse having a solution with having a business. Here’s the truth: if you can’t clearly explain why someone would pick you over their current approach (including doing nothing), you don’t have a business idea, you have a hobby.

This step kills bad ideas fast and cheap. Better to learn this in 2 days than in month 6.

Why This Matters

Your real competition isn’t just other products. It’s:

• Direct competitors: Other products that solve the same problem in the same way

• Indirect competitors: Different solutions to the same problem

• Do nothing: The status quo, ignoring the problem entirely (often your biggest competitor)

If you can’t win against all three, you don’t have a wedge; you have a weak market position that will bleed money.

Map Your Real Competition

First, ensure your idea is concrete enough to research. A strong idea summary should answer:

• The problem: What specific pain or inefficiency are you solving?

• The people: Which roles, industries, or customer segments feel this problem most?

• Your solution: What’s your approach, and how does it work at a high level?

• The outcome: What measurable value will customers gain?

Avoid vague statements like “helps people save time.” Instead: “helps finance managers close monthly books 5 days faster by auto-categorizing expenses from credit card feeds.”

Copy-Paste AI Prompt:

Act as a competitive research analyst. Take my idea and identify all competitors.

# My Idea

- Problem: [WHAT SPECIFIC PAIN OR INEFFICIENCY ARE YOU SOLVING?]
- Customer Segment: [WHICH ROLES, INDUSTRIES, OR CUSTOMER SEGMENTS FEEL THIS PROBLEM MOST?]
- Solution: [WHAT IS YOUR APPROACH OR PRODUCT CONCEPT?]
- Outcome: [WHAT MEASURABLE VALUE WILL CUSTOMERS GAIN?]

# Research Task

Identify competitors in three categories:

1. DIRECT COMPETITORS (products/services that solve this exact problem)
2. INDIRECT COMPETITORS (different approaches to solving the same problem)
3. DO NOTHING ALTERNATIVE (what happens if customers ignore the problem)

For each category, provide a table with these columns:

- Name (specific company/solution, not “various tools”)
- Value Proposition (in one sentence)
- Pricing (actual current pricing from their website)
- Main Strength (from user perspective)
- Source URL (so I can verify)

List 3-5 options per category. Be specific about real companies and solutions.

If you’re unsure about any pricing or details, say “VERIFY MANUALLY” rather than guessing.

After you get results:

1. Verify the essentials. Pick your top 3 competitors and visit their actual websites. Check three things: Does the product exist as described? Is the pricing accurate? Do they actually solve the problem the AI claimed? If you find discrepancies, trust the website over the AI.

2. Watch for AI hallucination patterns. If all competitors sound eerily similar, or if pricing information is consistently vague (”$10-50/user/month”), the AI is guessing rather than researching. When you see red flags like broken URLs or generic descriptions, set those results aside and search manually.

3. Fill the gaps yourself. You probably know competitors the AI missed, tools you’ve heard about, solutions friends mentioned, or workarounds people use. Add them to your map. Your lived experience matters more than AI completeness.

Extract User Frustrations

Now find out what people actually hate about their current options. This is where your opportunity lives.

Instead of reading hundreds of reviews yourself, let AI analyze user feedback about your specific competitors, then verify the patterns manually.

Copy-Paste AI Prompt:

Act as a user research analyst. Based on the competitors I identified:

1. DIRECT COMPETITORS: [LIST FROM YOUR COMPETITION MAP]
2. INDIRECT COMPETITORS: [LIST FROM YOUR COMPETITION MAP]
3. DO NOTHING ALTERNATIVE: [DESCRIPTION FROM YOUR COMPETITION MAP]

Research user frustrations by analyzing:

- G2/Capterra/Trustpilot reviews for each direct competitor
- Reddit discussions comparing these solutions
- LinkedIn posts about challenges with these tools
- App store reviews (if applicable)

Extract and synthesize:

1. TOP 10 FRUSTRATIONS across all competitors (ranked by frequency mentioned)
2. MOST EMOTIONAL COMPLAINTS (look for strong language: “hate,” “waste,” “impossible,” “frustrating”)
3. UNSOLVED PROBLEMS (frustrations that appear across ALL competitors, suggesting no one solves this well)
4. EXPENSIVE PROBLEMS (complaints mentioning costs: “this costs us $X” or “losing revenue because...”)
5. DO NOTHING PAIN POINTS (specific problems with ignoring the issue entirely)

For each frustration, provide:

- Which competitors attempt to address it
- How well they solve it (based on user feedback)
- What users say is still missing
- Economic impact mentioned (if any)
- 2-3 actual user quotes showing this frustration

Create a “Frustration Opportunity Matrix”:

- HIGH frustration + LOW solution coverage = BIG OPPORTUNITY
- HIGH frustration + HIGH solution coverage = CROWDED SPACE
- LOW frustration + ANY coverage = NOT WORTH PURSUING

Recommend the top 3 frustration areas with the biggest opportunity gaps.

Include source URLs for all reviews/discussions referenced.

After you get results:

1. Read real reviews yourself. Don’t just trust the AI’s summary. Pick the top 3 competitors and read 5-10 recent reviews for each. You’re looking for emotional language, specific stories, and frustrations that get mentioned repeatedly. The AI might miss the intensity of feeling that tells you this is a real pain point.

2. Look for economic evidence. The best frustrations to solve are ones people already pay to address; whether through tools, workarounds, or time investment. When you see complaints like “spend 10 hours a week on this” or “costs us $5K per month,” you’ve found something people will pay to fix. Ignore frustrations that are merely annoying but don’t cost anything meaningful.

3. Distinguish pain from preference. “Would be nice to have” is not the same as “will pay money to solve.” You’re looking for frustrations that have consequences: missed deadlines, lost revenue, career impact, or genuine emotional toll. Those are the ones worth building for.

Define Your Wedge

Based on the frustration gaps you identified, craft a sharp differentiation statement that directly addresses the biggest unmet needs.

AI Positioning Prompt:

Act as a positioning strategist.

TOP FRUSTRATION GAPS I IDENTIFIED:

- [PASTE YOUR TOP 3 FRUSTRATION AREAS FROM PREVIOUS RESEARCH]
- 
- 

COMPETITOR LANDSCAPE:

- [PASTE KEY DIRECT/INDIRECT COMPETITORS]
- 
- 

MY SOLUTION APPROACH:[DESCRIBE YOUR BASIC IDEA IN 2-3 SENTENCES]

Create 3-5 different positioning statements using this format:

“While [current solutions] force customers to [main frustration], we [your approach] so they can [specific benefit] in [timeframe/measurement].”

For each positioning option:

- Which frustration gap does it target
- Why is this angle defensible (what makes it hard to copy)
- Commercial potential (will people pay for this?)
- Risks or weaknesses

Then recommend the STRONGEST positioning statement and explain:

- Why this wedge matters most to users
- What evidence supports this (from frustration research)
- How to validate this positioning with real customers
- One potential weakness to watch

Make each statement specific and measurable. Avoid generic benefits like “easier” or “better.”

Your Wedge Statement:

Take the AI’s recommended positioning and refine it into one crystal-clear sentence:

“While [current solutions] force customers to [main frustration], we [your approach] so they can [specific benefit] in [timeframe/measurement].”

Examples:

✅ “While project management tools force teams to spend hours in status meetings, we automatically extract progress from their existing work so they can skip 80% of status calls.”

✅ “While expense management requires manual receipt entry, we extract data directly from credit card transactions so finance teams can close books 5 days faster.”

❌ “We make project management easier and more efficient for teams.” (Too vague! easier how? Efficient at what?)

Checkpoint

STOP IF:

• You can’t explain your wedge in one clear sentence

• Every major competitor already solves the problem well (HIGH frustration + HIGH solution coverage)

• The “do nothing” option is genuinely acceptable for most customers

• Nobody mentions spending money or time on this problem

YOU SHOULD HAVE:

• One crisp wedge statement (the sentence above)

• Three pieces of evidence supporting it (user quotes, market gaps, competitor weaknesses)

• Clear understanding of which frustration you’re targeting

IF YOU’RE STUCK: Your idea might be solving a problem people don’t care enough about to pay for. Consider pivoting to a different frustration from your research, or validate whether this problem is worth solving at all.

Now that you know what makes you unique, you need to identify the specific people who truly care about that difference. Not everyone, just the ones who’ll pay you first.

Step 2: Define Your Target

“Everyone” is not a market. It’s the fastest way to burn through your runway, talking to people who will never buy.

You need a focused starting point: a small, specific group of people who feel the pain intensely and have the budget to solve it.

Why This Matters

The best first customers are people who:

• Deal with your problem daily (not monthly)

• Have budget authority or strong influence over spending

• Are already spending money on partial solutions

• Can implement new tools without 6-month procurement cycles

Find these people first. Expand later.

Identify Your Beachhead Segments

A “beachhead” is your focused starting point: the specific group you’ll win first before expanding to other markets.

Think about who feels your problem most acutely and can act on a solution quickly.

Copy-Paste AI Prompt:

Based on this wedge: [PASTE YOUR WEDGE STATEMENT FROM STEP 1], identify 4-5 potential beachhead segments who would pay for a solution.

For each segment, provide:

SEGMENT PROFILE:

- Specific job titles/roles (be precise: not “managers” but “Engineering Managers at Series A startups”)
- Company size and industry (if B2B) or demographics (if B2C)
- Why this problem is expensive/painful for them specifically
- Estimated budget they might have for solutions
- How easy they are to reach (LinkedIn groups, Slack communities, subreddits, events)
- Decision-making process (do they control budget or need approval?)

SCORING (rate 1-10 for each):

- Pain Intensity: How much does this problem cost them in time/money/stress?
- Budget Availability: Can they pay $X/month without major approvals?
- Your Access: Can you reach 100 of them in the next 2 weeks?
- Speed to Close: Can they start a pilot in days/weeks vs months?

TOTAL SCORE: [Sum of all four scores]

Rank segments by total score and recommend the best beachhead with specific reasoning.

After you get results:

1. Prove you can reach them. Open LinkedIn (for B2B) or relevant social platforms (for B2C) and search for people matching your segment description. Set a timer for 10 minutes. If you can’t find at least 50-100 people who match your profile in that time, your segment is either too narrow or poorly defined. Refine and try again.

2. Verify they have budget. For B2B, search “[job title] typical software budget” or look up salary ranges and company spending patterns. For B2C, check if people in this demographic actually spend money on solutions in your category. Look for evidence in reviews, forum discussions, or competing products’ customer testimonials. If they consistently choose free options or complain about pricing, they might not be your ideal first customers.

3. Choose one and commit. Yes, just one segment. The temptation will be to hedge your bets and keep multiple options open. Resist it. You need laser focus to validate quickly. Pick the segment with the highest total score and go deep. You can expand to adjacent segments after you’ve proven the model with your first group.

Create Your Ideal Customer Profile

Get specific about exactly who you’re targeting in your chosen segment. This profile will guide every outreach message, interview question, and product decision.

Copy-Paste AI Prompt (B2B):

Create a detailed Ideal Customer Profile for: [PASTE YOUR CHOSEN SEGMENT FROM PREVIOUS STEP]

Include:

PROFESSIONAL CONTEXT:

- Exact job titles (3-5 variations they might use)
- Seniority level and typical reporting structure
- Company size (employee count and revenue)
- Industries where this role is most common
- Years of experience typical for this role

CURRENT STATE:

- Tools they currently use for [your problem area]
- Typical workflow and daily tasks related to your problem
- Team size they manage (if applicable)
- Budget they typically control

PAIN & MOTIVATION:

- Top 3 KPIs they’re measured on
- How your problem impacts those KPIs
- Career consequences if the problem persists
- What “success” looks like for them

BUYING BEHAVIOR:

- Typical procurement process (self-serve vs approval needed)
- Budget ranges for tools in your category
- Who else is involved in purchase decisions
- How they evaluate new solutions (trial, demo, references?)

Where to find them:

- LinkedIn search strings
- Slack/Discord communities
- Subreddits
- Industry events or conferences
- Professional associations

Format as a one-page profile I can reference during outreach.

Copy-Paste AI Prompt (B2C):

Create a detailed Ideal Customer Profile for: [PASTE YOUR CHOSEN SEGMENT FROM PREVIOUS STEP]

Include:

DEMOGRAPHICS:

- Age range
- Gender (if relevant to the problem)
- Location/geography
- Income level or spending capacity
- Education or professional background (if relevant)

PSYCHOGRAPHICS:

- Values and priorities
- Lifestyle and daily routines
- What motivates them
- What frustrates them
- How they spend free time

BEHAVIORAL PATTERNS:

- How often they experience your problem
- Current solutions they use (even if inadequate)
- Where they discover new products (TikTok, Reddit, friends, ads?)
- Purchase decision process (impulse vs research-heavy)
- Price sensitivity and typical spending in this category

GOALS & FRUSTRATIONS:

- What they’re trying to achieve
- Obstacles in their way
- Emotional impact of the problem
- What “success” would look like

TRIGGERS & TIMING:

- What events push them to seek solutions (life changes, seasons, specific frustrations)
- Best time to reach them (time of day, day of week, time of year)

Where to find them:

- Online communities (Reddit, Facebook groups, Discord servers)
- Social media platforms they use most
- Physical locations or events
- Influencers or creators they follow

Format as a one-page profile I can reference during outreach.

After you get results:

1. Make it real with actual people. Find 3-5 specific individuals on LinkedIn or social media who perfectly match your ICP. Don’t just note their names, read their recent posts, see what they share, notice how they talk about problems in your domain. This transforms your ICP from an abstract persona into something concrete. When you write outreach messages later, you’ll write to these specific people, not to a theoretical construct.

Map the Job They’re Hiring You to Do

Before you move to Step 3, you need to understand something critical: people don’t buy products, they “hire” them to make progress in their lives.

When someone adopts your solution, they’re not buying features. They’re hiring you to help them get from their current frustrating situation to a better outcome. Understanding this “job” helps you in three ways:

1. You’ll know what to build. The job reveals what actually matters vs what sounds cool.

2. You’ll price appropriately. You’re not charging for software, you’re charging for the progress you enable.

3. You’ll position against alternatives correctly. You’re competing with everything customers currently “hire” to do this job, not just direct competitors.

This framework is called Jobs-to-be-Done (JTBD), and it’s how successful products discover what customers actually need.

Copy-Paste AI Prompt:

Act as a Jobs-to-be-Done expert. For this customer segment: [PASTE YOUR TARGET CUSTOMER FROM YOUR WEDGE], map out the job they’re trying to do related to [PASTE YOUR PROBLEM AREA].

Break down:

1. JOB STATEMENT: “When [situation], I want to [motivation], so I can [expected outcome]”

2. JOB STEPS (what they do today):

- Define the problem/need
- Search for solutions
- Evaluate options
- Implement chosen solution
- Monitor results
- Adjust/optimize

3. PAIN POINTS at each step:

- Where do they get stuck?
- What takes too much time?
- What’s expensive or risky?
- What requires expertise they don’t have?

4. SUCCESS METRICS (how they measure job completion):

- Time savings
- Cost reduction
- Quality improvement
- Risk mitigation

Identify the 3 most underserved outcomes where our solution could have the biggest impact.

Checkpoint

STOP IF:

• Your ICP doesn’t have budget authority or influence over spending

• The problem happens monthly or less frequently

• You can’t find at least 50 people matching this profile in 15 minutes of searching

YOU SHOULD HAVE:

• One-page ICP profile (either B2B or B2C version)

• List of 3-5 real people who match this profile

• Confirmed channels where you can reach 100+ similar people

• Validated budget range they can spend

Step 3: Get Paid

This is the only step that actually validates whether you have a business. Steps 1-2 were smart research, but Step 3 is existential.

Everything up until now has been preparation. This is where you find out if you have a business or just an interesting idea.

Why This Is Different

Most founders collect feedback: “Would you use this?” “Does this sound interesting?” “What would you pay?”

This is worthless. People lie, especially when being polite. They’ll say “yes” to vague future questions to end the conversation.

The only validation that matters is: Will someone pay you money right now to solve their problem?

Not:

• ❌ “I’d definitely use this when it’s ready”

• ❌ “This sounds really interesting”

• ❌ “I’d probably pay around $50/month”

But:

• ✅ “Here’s $200. Solve this problem for me in the next 2 weeks.”

That’s what we’re building toward.

The 5-Day Validation Sprint

• Day 1-2: Reach out to 20 people in your ICP

• Day 3-4: Conduct 5 problem interviews

• Day 5: Pitch paid pilots to 3 qualified prospects

Let’s break down each day.

Day 1-2: Outreach That Gets Responses

Your goal: Get 5 people to agree to 15-minute calls about their problem.

Finding People

For B2B:

• LinkedIn: Use exact job titles from your ICP. Filter by company size/industry. Send connection requests with personalized notes.

• Communities: Join Slack workspaces, Discord servers, or subreddits where your ICP hangs out. Participate first, then reach out.

• Company websites: Find the right role at target companies and email them directly (use Hunter.io or similar to find emails).

• Warm intros: Ask your network for introductions. “Do you know any [job title] at [company size] companies?”

For B2C:

• Reddit/Facebook groups: Search for communities discussing your problem area. Look for people actively complaining about the issue.

• Discord/Slack communities: Hobby or interest-based communities where your ICP congregates.

• Local events: Meetups, classes, or gatherings where you’ll find your demographic.

• Social media: Search hashtags or keywords related to your problem. Find people posting about the frustration.

• Friends of friends: Ask your network for warm intros to people matching your demographic.

Pro tip: You need to reach out to 20 people to get 5 interviews. That’s normal. Don’t get discouraged if most people don’t respond.

The Outreach Message

Copy-Paste Outreach Script (Customize for your ICP):

Subject: Quick question about [THEIR MAIN CHALLENGE]

Hi [NAME],

I noticed you’re [SPECIFIC ROLE] at [COMPANY/CONTEXT]. I’m researching how [ROLE TYPE] like yourself handle [SPECIFIC PROBLEM AREA].

Most [ROLE TYPE] I talk to spend [TIME/EFFORT] on [CURRENT APPROACH] but still struggle with [MAIN FRUSTRATION FROM YOUR RESEARCH].

Would you be open to a quick 15-minute call this week? I’m not selling anything, just trying to understand if this is as much of a headache as it seems.

I can share what I’m learning from others in similar roles.

Best,

[YOUR NAME]

P.S. Happy to work around your schedule. I know how busy [THEIR ROLE] gets, especially during [RELEVANT SEASON/EVENT].

Why this works:

• Shows you understand their world (specific role, specific challenge)

• Makes a small ask (15 minutes, not 30+)

• Promises value exchange (share what you’re learning)

• The P.S. adds personality and shows you get their context

Example:

Subject: Quick question about expense reporting

Hi Jessica,

I noticed you’re a Finance Manager at a 50-person startup. I’m researching how finance teams handle monthly expense reporting.

Most finance managers I talk to spend 5-10 hours per month chasing receipts and fixing categorization errors, but still end up with incomplete data.

Would you be open to a quick 15-minute call this week? I’m not selling anything, just trying to understand if this is as much of a headache as it seems.

I can share what I’m learning from other finance teams.

Best,

Marcus

P.S. Happy to work around your schedule. I know how busy month-end close gets.

Follow-Up Script (Send 3 days later if no response)

Hi [NAME],

Following up on my note about [PROBLEM AREA].

I’ve now talked to [NUMBER] other [ROLE TYPE] and discovered that [ONE SPECIFIC INSIGHT FROM INTERVIEWS].

Still interested in a quick call? Takes 15 minutes and I think you’d find the benchmarking data useful.

[CALENDAR LINK] if that’s easier.

Thanks,

[YOUR NAME]

Why follow up: 50% of responses come from the follow-up. People are busy. One reminder is professional, not pushy.

Day 3-4: Problem Interview Script

You’ve got 5 people scheduled. Now you need to actually talk to them.

Your Goal

Understand their current approach, frustrations, and what they’d pay to solve it. Don’t pitch your solution yet. This is pure discovery.

You’re listening for:

1. How painful is this problem really?

2. What are they currently doing about it?

3. How much does it cost them (time, money, stress)?

4. Would they pay to solve it?

The Interview Structure (15-20 minutes)

Opening (2 minutes)

“Thanks for making time. I’m researching [PROBLEM AREA] and how people like you handle it today. I’m not selling anything, just trying to understand the landscape. Sound good?”

Then: “Before we dive in, tell me a bit about your role and what a typical day looks like.”

Why this works: Gets them talking, establishes rapport, gives you context.

Current State (5 minutes)

Ask:

• “Walk me through how you currently handle [PROBLEM]. What’s your process?”

• “What tools or systems do you use?”

• “Who else is involved? Do you do this alone or with a team?”

• “How much time does this take per week/month?”

Listen for: Specific workflows, tools they pay for, time invested. If they’re already spending money or significant time, the problem is real.

Pain Points (5 minutes)

Ask:

• “What’s most frustrating about your current approach?”

• “When you mentioned [FRUSTRATION THEY JUST DESCRIBED], can you tell me about the last time that happened?”

• “What happens when [PROBLEM] isn’t handled well? What’s the impact?”

• “Have you looked for better solutions? What did you find?”

Listen for: Emotional language (”hate,” “dread,” “nightmare”), concrete costs (”we lost a client because...”), frequency (”this happens every week”).

Get specific stories: “Tell me about the last time this went wrong” is better than “Does this ever go wrong?” Stories reveal true pain.

Economics (3-4 minutes)

This is where most founders get timid. Don’t be. You need to understand if they’ll pay.

Ask:

• “What do you currently spend on [PROBLEM AREA]?” (Include tools, people’s time, other costs)

• “If someone could [SOLVE THEIR BIGGEST FRUSTRATION], what would that be worth to you?”

• “Help me understand your budget situation. How do you typically pay for tools like this?”

If they seem unsure about cost, help them calculate:

“Let’s add this up together:

• You spend about [X hours] per week on this, right?

• If your time is worth roughly [estimate based on role], that’s about $[X] per month just in your time

• Plus you’re paying for [TOOL] at $[Y] per month

• And when it goes wrong, you mentioned [CONSEQUENCE]

So you’re probably spending around $[TOTAL] per month on this problem. Does that sound about right?”

This isn’t pushy, you’re helping them see the true cost. Most people underestimate what problems cost them.

Decision Process (2-3 minutes)

Ask:

• “Who else would be involved in evaluating new solutions for this?”

• “How do you typically try new tools? Do you need formal approval or can you just start using something?”

• “What would need to be true for you to pilot something new?”

Listen for: How fast they can move. “I can just use my credit card” is very different from “I need to get three VP approvals.”

Closing (1 minute)

“This has been super helpful. I’m working on something that might help with [THEIR BIGGEST PAIN POINT]. Would you be interested in hearing about it once I have something concrete to show?”

If yes: “Great, I’ll follow up in the next few days.”

Don’t pitch yet. You’re setting up the paid pilot conversation.

Real Interview Example

Here’s what a good exchange looks like:

You: “Walk me through how you currently handle expense reporting.”

Them: “Ugh, it’s a mess. Everyone submits receipts in Slack, email, or our expense tool, whatever they remember. I spend probably 6-7 hours at month-end chasing people down and fixing errors.”

You: “When you say ‘fixing errors,’ what does that typically look like?”

Them: “People upload the receipt but don’t categorize it, or they put it in the wrong month, or they claim mileage without documentation. I have to go back and forth with each person. It’s tedious.”

You: “Tell me about the last time this caused a real problem.”

Them: “Last month we had an investor audit and couldn’t close the books on time because three people hadn’t submitted receipts from a conference. Made us look disorganized. Our CFO was not happy.”

You: “What did that delay cost you?”

Them: “Hard to say exactly, but we delayed the board meeting by a week. And I spent probably an extra 10 hours tracking everything down. Plus the stress. I looked bad.”

Notice:

• You asked follow-up questions to get specific

• You got a real story with consequences

• You uncovered emotional cost (stress, looking bad) and time cost (10 extra hours)

• You didn’t pitch anything, just listened

Day 5: The Commitment Validation

After 5 interviews, you should have 2-3 people who:

• Have the problem daily or weekly (not monthly)

• Currently spend money trying to solve it (tools, time, workarounds)

• Have decision-making authority (or strong influence)

• Showed genuine interest when you asked about hearing your solution

These are your paid pilot prospects.

The Paid Pilot Pitch

Here’s the critical insight: You don’t need to build software to validate your business. You just need to solve the problem.

Offer to do it manually first.

Copy-Paste Paid Pilot Script:

Hi [NAME],

Great talking to you yesterday about [THEIR SPECIFIC PAIN].

I have an idea that might help immediately. Instead of waiting months to build software, what if I could solve this problem for you manually in the next 2 weeks?

Here’s what I’m thinking:

I’ll [SPECIFIC TASK YOU’LL DO MANUALLY] for your next [PROJECT/CYCLE/MONTH]. You pay me $[SMALL AMOUNT] for the service. I’ll document everything so you can see the exact process.

This gives you immediate relief from [THEIR PAIN], and if it works, we can talk about automating it with software later.

Even if you decide not to automate, you’ll have a proven process that works.

Interested in trying this for $[AMOUNT]?

Best,

[YOUR NAME]

P.S. I can start as soon as you’re ready, even this week.

B2B Example:

Hi Jessica,

Great talking about the expense reporting chaos yesterday.

I have an idea: What if I handled all expense collection and categorization for you for the next 2 weeks?

Here’s what I’ll do:

- Set up a simple system where your team submits receipts

- I’ll categorize everything and flag any issues

- You get a clean report ready for your books, no chasing required

You pay me $200 for the 2-week trial. I’ll document the entire process so you can see exactly how it works.

This gives you back those 6-7 hours you’re spending now, and if it works well, we can talk about automating it.

Interested?

Best,

Oliver

B2C Example:

Hi Alex,

Really appreciated hearing about your frustration with meal planning yesterday.

Quick idea: What if I created a personalized meal plan for you each week for the next month?

Here’s how it works:

- Tell me your dietary preferences and what you have in your pantry

- I’ll send you a full week’s meal plan every Sunday with recipes and a shopping list

- All meals under 30 minutes, using ingredients you can find anywhere

$40 for the full month (4 weekly plans). If you love it, I’ll eventually build a tool that does this automatically.

Best,

Oliver

How to price the pilot:

Don’t overthink this. You’re testing willingness to pay, not setting your final price.

For B2B:

• Research what they currently spend. In your interviews, you should have discovered their current costs (tools, time, opportunity cost). Price your pilot at 25-50% of their monthly cost.

• Example: If they spend $400/month in time and tools, charge $100-200 for a 2-week pilot.

For B2C:

• Look at comparable products. What do similar solutions charge? Price your pilot at 40-60% of the monthly equivalent.

• Example: If competing meal planning apps charge $15/month, charge $30-40 for a 4-week pilot ($7.50-10/week).

Next Steps After Validation

Got someone to pay you? Congratulations! You’ve done what most founders never do. You’ve validated demand with actual money, not promises.

Now what?

Scale from 1 to 10 paying customers. Use the same outreach and interview process to find more people with the exact same problem. Don’t expand to new segments yet! Go deep in your wedge.

Build your MVP based on actual job requirements. You know exactly what job customers are hiring you to do and what outcomes they measure. Build the minimum solution that delivers those outcomes reliably.

Expand research only when needed for decisions. Now that you have paying customers, you can ask them about pricing, features, channels, and expansion markets. Let customer needs drive your research agenda, not academic curiosity.

Document everything for future team members. Your validation research becomes the foundation for product development, marketing messaging, and sales training. Keep detailed notes on customer language, objections, and success criteria.

The Bottom Line

Most startup advice assumes you have unlimited time and money to “get it right.” You don’t.

You have a limited runway and limited patience from potential customers. Use both wisely.

Research that doesn’t lead to paying customers is just expensive procrastination. Focus on the minimum viable validation that answers one question: will people pay me to solve their problem?

Everything else can wait until you have that answer.

The fastest way to validate your startup idea isn’t perfect research. It’s finding someone who will pay you to solve their problem. Start there, and let paying customers guide everything else.

Now stop reading and start reaching out. Your first paying customer is waiting.