Planning to Do a SaaS Startup? Don’t Forget the 20 Interview Rule.

Here’s a fun (and slightly uncomfortable) truth: the fastest way to waste 6 months building a SaaS product is to build it before you’ve talked to enough real humans who might pay for it.

Founders love building. Customers love… not being bothered. The gap between those two loves is where your startup will either find product-market fitor become a cautionary LinkedIn post that starts with “Humbled to announce…”

That’s where the 20 Interview Rule comes in. Before you commit serious time, money, or sanity to building your SaaS, you do at least 20 high-quality customer interviews. Not surveys. Not “a few chats.” Not asking your friends (who love you and therefore cannot be trusted). Actual conversations with the people who have the problem, live with the problem, and already spend time or money trying to solve it.

This article breaks down what the 20 Interview Rule is, why it works, how to do it without being weird, and how to turn those interviews into a sharper idea, a better MVP, and a go-to-market plan that doesn’t rely on hope and caffeine.

What Exactly Is the 20 Interview Rule?

The 20 Interview Rule is a simple pre-build discipline:

• Complete 20 customer discovery interviews with your target buyers/users before you build your SaaS product beyond a prototype or lightweight demo.
• Use those interviews to validate (or destroy) your assumptions about pain, urgency, current alternatives, decision-making, and willingness to pay.
• Look for patterns, not compliments.

Why 20? Because it’s big enough to reveal repeating themes (and expose your favorite delusions), but still small enough to finish before your motivation gets replaced by “maybe I should just buy a laundromat.”

Why 20 Interviews Beats 20,000 Lines of Code

1) You’re not validating the productyou’re validating the problem

Early-stage SaaS success usually hinges on one boring, beautiful thing: a painful problem that a specific group of people will pay to solve. Interviews help you confirm the problem is real, frequent, expensive, and emotionally annoying enough to trigger action.

2) Patterns need volume

After 3 interviews, you have stories. After 10, you start seeing themes. After 20, you can say things like “This pain shows up in nearly every team with X workflow” without lying to yourself.

3) You learn the customer’s language (which is also your future SEO)

People tell you how they describe their work, their constraints, and their nightmares. Those exact phrases become your:

• homepage headline
• Google/Bing keywords
• cold email copy
• demo narrative
• pricing page “aha” bullets

Founders often guess at messaging. Interviews hand you messagingalready field-tested in the customer’s own words.

4) It prevents “feature buffet syndrome”

If you ask customers “What features do you want?” they’ll give you a wish list longer than your runway. Interviews done properly focus on past behavior: what they did last week, what broke, what they tried, and what it cost. That’s where real requirements hide.

The Golden Rule of Interviews: Don’t Pitch

If your interview sounds like a TED Talk with a Q&A at the end, you’re doing it wrong.

You are not selling. You are investigating.

In early SaaS discovery, the goal is to understand:

• Context: What job are they trying to get done?
• Trigger: What makes the problem urgent?
• Current solutions: Spreadsheets, Zapier duct tape, internal tools, competitors, “we just suffer,” etc.
• Cost of pain: Time, money, risk, stress, missed revenue, compliance issues
• Buying dynamics: Who decides? Who blocks? Who signs?

If you must talk about your solution, keep it to a 10-second concept line and return to questions. Think: detective, not magician.

How to Run the 20 Interviews Like a Pro (Without Becoming a Nuisance)

Step 1: Define a tight ICP (Ideal Customer Profile)

“Small businesses” is not an ICP. It’s a cry for help. Your early interviews should focus on a narrow segment where:

• the problem is frequent
• the buyer has budget authority
• the workflow is similar across companies
• the pain has measurable consequences

Example ICP (good): “Revenue operations managers at B2B SaaS companies (50–300 employees) struggling to unify lead routing across Salesforce and HubSpot.”

Example ICP (bad): “Anyone who uses spreadsheets.” (So… Earth.)

Step 2: Recruit interviewees (yes, you can do this without ads)

Here are practical channels founders use to find the right people:

• LinkedIn: targeted outreach to specific job titles
• industry Slack/Discord communities
• relevant subreddits and forums (carefully, respectfully)
• your network (but not your “supportive best friend” network)
• cold emails to companies that match your ICP
• customer lists from adjacent products/partners (if you have access)

Incentives: For busy professionals, a $25–$100 gift card can raise response rates. For founders doing B2B, “I’ll share a summary of industry benchmarks I’m collecting” can also workif you actually do it.

Step 3: Use an interview script that forces truth

Great interview questions are usually:

• open-ended (“Walk me through…”)
• specific (“Tell me about the last time…”)
• behavior-based (past actions, not future opinions)

Try this structure:

Part A: Warm context (5 minutes)

• “What’s your role, and what does success look like this quarter?”
• “What tools are essential to your workflow?”

Part B: The pain safari (20 minutes)

• “Tell me about the last time you ran into [problem area].”
• “What happened next?”
• “How often does this come up?”
• “What does it cost youtime, money, risk, reputation?”
• “What have you tried to fix it? What worked, what didn’t?”

Part C: Buying reality (10 minutes)

• “If you wanted to solve this, what’s the process?”
• “Who else needs to be involved?”
• “What budget bucket would this come from?”
• “What would make this a ‘must-fix-now’?”

Part D: Light concept check (optional, 5 minutes)

• “If there were a tool that did [outcome], what would you worry about?”
• “What would you need to believe to try it?”

Note: If they say “That sounds cool!” treat it as decorative. The real signal is: “We tried to solve that last month and spent $8k and two engineers on it.”

What Counts as a “Good” Interview?

A good customer discovery interview has these traits:

• 30–45 minutes, focused and respectful
• the customer talks at least 70% of the time
• you collect specific stories about real events
• you leave with clear notes about triggers, constraints, and alternatives

A bad interview is mostly you explaining your vision while the other person politely wonders what they’re having for lunch.

How to Analyze Interviews Without Drowning in Notes

1) Tag the evidence

After each interview, capture these in a structured doc:

• Pain severity: mild annoyance vs. business-threatening
• Frequency: daily/weekly/monthly
• Current workaround: tools, manual process, internal build
• Cost of pain: time, money, churn, compliance, delays
• Buyer map: user, champion, economic buyer, blocker
• Language: exact phrases they use

2) Do an “insight tally,” not a vibes recap

Create a simple grid of recurring themes and count mentions. You’re looking for:

• top 3–5 recurring pains
• common triggers (audits, onboarding spikes, pipeline reviews, quarter-end chaos)
• consistent willingness to pay indicators (budget exists, prior spend, urgency)

3) Separate “nice-to-have” from “they’re already spending money”

The most valuable interview moment is when someone reveals they already pay for a workaround (or spend a lot of labor). That’s your pricing gravity. In B2B SaaS, willingness to pay often shows up as:

• existing vendor contracts
• consultant spend
• internal engineering time
• risk exposure (fines, audits, security incidents)

Turning 20 Interviews Into Your SaaS MVP Plan

After 20 interviews, you should be able to write (without sweating) the following:

A clear problem statement

Example: “Mid-market RevOps teams lose qualified leads because routing rules across tools are inconsistent, causing delayed follow-ups and missed pipeline.”

A narrow first use case

Not “all of sales ops.” Pick the smallest workflow that creates a measurable win. SaaS startups win by being painfully specific before they become broadly useful.

A realistic MVP scope

Your MVP is not “the full product but uglier.” It’s the minimum that lets customers achieve the key outcome. If interviews show the pain is mostly about visibility, an MVP might be a dashboard and alertsnot a full workflow engine.

Early pricing and packaging hypotheses

Interviews reveal what they value (speed, risk reduction, compliance, fewer headcount hours). That helps you test pricing models early, whether it’s per-seat, usage-based, or tied to outcomes.

Common Interview Mistakes (That Quietly Murder SaaS Startups)

Mistake 1: Interviewing the wrong persona

The user feels the pain. The buyer pays. Sometimes they’re the same person. Often they’re not. A strong interview set includes a mix of:

• end users
• team leads/managers
• economic buyers (budget owners)
• gatekeepers (security, IT, procurement)

Mistake 2: Asking hypothetical questions

“Would you use this?” is a trap. People will say yes because they are nice. Ask about real past behavior instead.

Mistake 3: Ignoring the competition that isn’t a product

Your biggest competitor is usually “do nothing,” plus a spreadsheet that has survived seven re-orgs out of pure spite.

Mistake 4: Stopping early because you heard one exciting story

One intense pain story can be realand still not be a market. The 20 Interview Rule protects you from building a company around a single loud edge case.

A Simple 20-Interview Sprint Plan (2 Weeks)

Days 1–2: Setup

• Define ICP and top hypotheses
• Write a 10-question interview guide
• Draft outreach messages

Days 3–10: Interviews (2–3 per day)

• Schedule 30–45 minutes each
• Record (with permission) or take structured notes
• Do a 10-minute debrief right after each call

Days 11–14: Synthesis

• Affinity map insights
• Choose one beachhead problem
• Define MVP outcome + success metric
• Write positioning using customer language

If you can’t finish 20 interviews in two weeks, it’s not because the universe is cruel. It’s because your recruiting funnel needs loveor your ICP is too vague to target.

Realistic Example: Turning Interviews Into a SaaS Wedge

Imagine you want to build “an AI tool for customer support.” That’s adorable. Also, that’s like saying you want to “build an internet.”

Now apply the 20 Interview Rule and you learn:

• Support leaders don’t want “AI.” They want fewer escalations and faster resolution.
• The biggest pain isn’t writing repliesit’s finding the right answer across messy internal docs.
• Security and compliance are deal-breakers.
• They already pay for a helpdesk and knowledge base, but search is unreliable.

Your MVP changes from “AI support agent” to something sharper:

• a knowledge retrieval layer that plugs into existing tools
• confidence scoring + citations inside the agent reply
• an admin panel for approved sources and redaction rules

That’s a wedge. It’s sellable. And it came from interviews, not vibes.

Conclusion: The 20 Interview Rule Is Cheaper Than Regret

Planning a SaaS startup isn’t just about picking a tech stack and a logo color that “feels premium.” It’s about reducing risk with evidence. The 20 Interview Rule forces you to earn your idea with reality.

Do the interviews. Hear the patterns. Let the problem shape the product. Then build the smallest thing that delivers the biggest relief.

Because the only thing worse than launching to crickets is realizing the crickets are your target market.

Experience Notes: What 20 Interviews Actually Feels Like (And Why You Should Still Do It)

I’ve watched founders approach customer interviews the way people approach dentist appointments: postponing, overthinking, and hoping the problem disappears if they don’t look at it directly. Spoiler: it doesn’t. The problem just grows teeth.

The first few interviews are usually awkward. You’ll be tempted to “help” the conversation by explaining your idea. Resist that urge. Early interviews are like spotting wildlife: the moment you sprint into the forest yelling, “DO YOU WANT AN MVP?” everything disappears. The best interviews happen when you’re calm, curious, and genuinely trying to understand the person’s world.

In one early-stage SaaS project I observed, the founder did five interviews and got five enthusiastic reactions. He started building immediatelybecause five people said it was “interesting.” But when he finally pushed for detailsbudget, process, urgencythose same people revealed they were already “kind of handling it” with a spreadsheet and a weekly ritual of mild suffering. Translation: not enough pain to buy. By interview twelve, a different pattern emerged: a narrower segment (same industry, different job role) had the pain daily and had already tried two paid tools that failed. That’s when the founder’s idea stopped being “a cool app” and started being “a must-fix workflow.” The remaining interviews didn’t just confirm the painthey taught him exactly how to position the product, which integrations were non-negotiable, and which features were distractions wearing tiny clown shoes.

Another very real lesson: the “best” interviewees aren’t always the friendliest. Sometimes the gold comes from the person who sounds skeptical. Skeptical people force clarity. They’ll tell you why procurement will block you, why your security story won’t fly, or why your pricing model makes no sense for how their budget is structured. If you’re building B2B SaaS, those objections are not negativity. They are free consultingdelivered with the emotional warmth of a tax audit.

By interview fifteen, you’ll likely hit the “pattern fatigue” phase, where it feels like everyone is repeating the same three pains. That’s a good sign. It means you’re finally seeing the market clearly. This is also the moment when you can start testing lightweight artifacts: a one-page concept doc, a clickable prototype, or a mock dashboard. The key is to use those artifacts as prompts“How would this fit into your process?”not as a pitch deck.

By interview twenty, something weird happens: you stop guessing. You can describe the customer’s day, their constraints, and their internal politics with uncomfortable accuracy. You can write copy that sounds like it came from them. You can prioritize features with confidence, because you know which pains are frequent and expensive. And if the interviews don’t show real pain or willingness to pay? That’s not failure. That’s you saving months of work. The interview rule isn’t a hurdle. It’s a refund policy for your time.

If you’re planning a SaaS startup, treat interviews like product development. Schedule them. Track them. Improve your script. Debrief immediately. The point isn’t to “talk to users” as a nice-to-have. The point is to build a business on reality instead of optimism.