How to Run Customer Interviews for Your SaaS Product: A Practical Guide for 2026

I wasted six months building a feature nobody wanted. Our team assumed we understood the problem because we had survey data, NPS scores, and a Slack channel full of feature requests. We had numbers. What we did not have was a single conversation with a customer about why they were struggling.

When I finally sat down with eight users over two weeks, the picture flipped completely. The feature we built solved the wrong problem. The real pain point was three clicks upstream in a workflow we had never watched anyone use. Eight conversations. That is all it took to see what months of quantitative data had missed.

Customer interviews are the most underused tool in a SaaS product team's toolkit. They take less time than most teams think, cost almost nothing, and produce the kind of insight that surveys and analytics cannot touch. Yet most product teams skip them entirely or do them so poorly that the results are useless.

This guide covers how to plan, conduct, and act on customer interviews, from choosing who to talk to all the way through turning conversation notes into product decisions.

Why Customer Interviews Matter More Than Most Teams Realize

Here is why customer interviews deserve a permanent spot in your product process. According to CB Insights, 35% of startups fail because there is no market need for their product (Rydoo, "Why 90% of Startups Fail," October 2024, https://www.rydoo.com/cfo-corner/why-startups-fail/). That is not a technology problem or a funding problem. It is a listening problem. Those teams built something without confirming that real people actually needed it.

Surveys tell you what people say they do. Analytics tell you what people actually do. Interviews tell you why. And the "why" is where product decisions live.

I have watched product managers stare at a dashboard showing a 40% drop-off on step three of an onboarding flow. They brainstorm fixes for hours. They redesign the page. They add tooltips. None of it works. Then someone interviews five users who dropped off, and the answer is obvious: they did not understand why they needed to connect their calendar. The problem was not the design. It was the messaging. No amount of analytics would have revealed that.

The User Interviews State of User Research 2024 report found that 81% of research teams now conduct a mix of both discovery and evaluative research projects, and 44% run continuous research programs (User Interviews, "The State of User Research Report 2024," June 2024, https://www.userinterviews.com/state-of-user-research-2024-report). That shift toward ongoing conversation with users, rather than occasional check-ins, reflects a growing recognition that the best products come from teams who talk to customers regularly.

What Interviews Reveal That Other Methods Miss

Let us break it down. Interviews give you access to context that no other research method provides.

Emotional drivers. A survey can tell you that users want "better reporting." An interview reveals that the VP of Sales is embarrassed in board meetings because she cannot pull accurate numbers fast enough. That emotional context changes how you design the solution.

Workarounds. Users build elaborate workarounds for problems your product should solve. They export to spreadsheets, copy-paste between tools, and create manual processes. You will never see these workarounds in your analytics because they happen outside your product. Interviews surface them.

Language. Customers describe their problems in their own words, and those words are gold for your marketing copy, your onboarding flows, and your help documentation. When a customer says "I need to stop drowning in feature requests from my Discord" that is a better headline than anything a copywriter would invent from scratch.

Unasked questions. The most valuable thing in an interview is often something the user mentions casually, something you would never have thought to ask about in a survey. Those surprise moments reshape your understanding of the problem space.

How Many Customer Interviews Do You Actually Need?

One of the biggest objections I hear from product teams is that interviews take too long and they need to talk to hundreds of people for the results to mean anything. That is wrong on both counts.

Jakob Nielsen's research at the Nielsen Norman Group demonstrated that testing with just 5 users uncovers approximately 85% of usability issues in an interface (Nielsen Norman Group, "How Many Participants for a UX Interview?," November 2024, https://www.nngroup.com/articles/interview-sample-size/). While interviews are different from usability tests, the principle of diminishing returns applies. After 5 to 8 interviews focused on the same topic, you start hearing the same themes repeated. Researchers call this "saturation," and it typically happens faster than people expect.

Steve Blank, the creator of the customer development methodology, recommends that startup teams conduct 10 to 15 customer interviews per week during the discovery phase (Steve Blank, "Customer Development," https://steveblank.com/category/customer-development/). That pace sounds aggressive, but Blank's point is that customer conversation should be a habit, not an event.

For established SaaS products, I recommend this cadence:

• For a specific feature or problem: 8 to 12 interviews with users who experience the problem
• For ongoing discovery: 2 to 4 interviews per week as a continuous practice
• For churn investigation: 5 to 10 interviews with recently churned users
• For pre-launch validation: 10 to 15 interviews with target users before building

The SaaStr "20 Interview Rule" puts it simply: before you commit engineering resources to a new product or major feature, talk to at least 20 potential customers (SaaStr, "Planning to Do a SaaS Startup? Don't Forget the 20 Interview Rule," November 2022, https://www.saastr.com/planning-to-do-a-saas-start-up-dont-forget-the-20-interview-rule/). If you cannot find 20 people willing to talk about the problem you are solving, that itself is a signal worth paying attention to.

Who Should You Interview?

Picking the right people to talk to is half the battle. Interview the wrong users and you will get data that leads you in circles. Here is how to choose wisely.

Match Your Interview Participants to Your Research Goal

Different questions require different audiences:

Research Goal	Who to Interview	Why
Understand a workflow problem	Active users who perform that workflow weekly	They know the pain points from direct experience
Reduce churn	Users who canceled in the last 30 to 60 days	Their reasons for leaving are fresh and specific
Validate a new feature idea	Users who requested something similar, plus non-users who fit your target profile	Requesters confirm demand; non-users reveal if the feature attracts new customers
Improve onboarding	Users who signed up in the last 2 weeks (both activated and not activated)	Recent memory of the experience, with contrast between success and failure
Explore a new market	Potential users who have never used your product	Fresh perspective without product bias

Avoid the "Fans Only" Trap

I see this mistake constantly. Teams send an email to their most engaged users asking for interview volunteers. They get 15 enthusiastic responses from people who love the product. Every interview confirms that the product is great. The team walks away feeling validated.

The problem is obvious: they only talked to fans. The users who are struggling, the ones who are about to churn, the ones who signed up and never came back, those are the people with the most valuable feedback. And they are the hardest to reach.

Make a deliberate effort to include:

• Users who signed up but never activated
• Users whose engagement has dropped in the last 30 days
• Users who contacted support with complaints
• Users who downgraded their plan
• Users from segments you are trying to grow into

If you track feature requests through a system like RoadmapAI, you already have a list of users who care enough to submit feedback. Those users make excellent interview candidates because they are invested enough to spend time talking, and their requests give you a starting point for the conversation.

How to Prepare for a Customer Interview

Walking into a customer interview without preparation wastes everyone's time. Fifteen minutes of prep transforms a meandering conversation into a focused research session.

Write a Discussion Guide, Not a Script

A discussion guide is a loose framework of topics and questions that keeps the conversation on track without making it feel like a survey. Here is why the distinction matters: a script forces you through a fixed sequence of questions regardless of where the conversation goes. A discussion guide gives you freedom to follow interesting threads while ensuring you cover the topics that matter.

A good discussion guide includes:

Opening questions (2 to 3 minutes): Warm up with background questions. "Tell me about your role. What does a typical week look like?" These questions are not filler. They establish context that makes everything else the person says more meaningful.

Core topic questions (15 to 20 minutes): These are the questions tied to your research goal. Frame them around behaviors and experiences, not opinions. "Walk me through the last time you tried to [specific task]" beats "Do you like our [feature]?" every time.

Deeper probes: Prepare follow-up questions for the themes you expect to hear. "You mentioned that took a long time. Can you show me what that looks like?" or "What did you do before you found a workaround?"

Closing questions (3 to 5 minutes): "Is there anything I should have asked about but didn't?" This question has given me some of the best insights I have ever received in interviews. People mention things they were holding back because they were not sure if it was relevant.

Research the Person Before the Call

Spend five minutes looking at the person's account data before the interview. Check their signup date, plan tier, feature usage, support tickets, and any feature requests they have submitted. This lets you ask informed questions instead of generic ones.

"I noticed you submitted a request for bulk export last month. Can you walk me through why that matters for your workflow?" That question shows you did your homework and gets the conversation to a useful place fast.

If your team uses a feature request tracking system, you can pull up a user's history before the call. That context turns a cold conversation into a warm one.

How to Conduct the Interview

The interview itself is where most teams go wrong. Asking the right questions is a skill, and it works differently than normal conversation. Here are the techniques that produce real insights.

Ask Open-Ended Questions About Behavior, Not Opinions

This is the single most important interviewing skill. Compare these two questions:

Bad: "Would you use a feature that automatically categorizes your feedback?"

Good: "Walk me through what happens when you get a new piece of feedback from a customer. What do you do with it?"

The first question leads the witness. Of course they will say yes. Who would say no to something that sounds helpful? But their "yes" tells you nothing about whether they would actually use it, how it fits their workflow, or whether the problem is even big enough to solve.

The second question reveals the actual workflow, the pain points, the workarounds, and the emotions. It gives you data you can act on.

Here are my go-to interview questions that consistently produce useful answers:

• "Tell me about the last time you [did the thing you are researching]. Walk me through it step by step."
• "What was the hardest part of that?"
• "What did you try before you found your current approach?"
• "If you could wave a magic wand and change one thing about how you handle [topic], what would it be?"
• "How are you solving this problem today?"
• "What happens when this goes wrong?"

Use the "Five Whys" to Get Past Surface Answers

When a user gives you a surface-level answer, do not accept it and move on. Dig deeper by asking "why" or "tell me more about that" repeatedly until you reach the root cause.

Here is an example from a real interview I conducted:

Me: "What is the biggest frustration with managing your product roadmap?"
User: "Keeping it updated."
Me: "What makes it hard to keep updated?"
User: "I get feedback from so many places. Slack, email, support tickets, our Discord."
Me: "And what happens when the feedback comes in from all those places?"
User: "Honestly? Most of it gets lost. I copy some into a spreadsheet but I miss a lot."
Me: "What is the impact when feedback gets lost?"
User: "We end up building things based on whoever yells the loudest instead of what most users actually need."

The real problem is not "keeping the roadmap updated." It is that feedback gets lost across channels, leading to bad prioritization decisions. That is a completely different (and much more actionable) insight than the first answer suggested.

Be Comfortable With Silence

When you ask a question and the person pauses, resist the urge to fill the silence. Count to five in your head. Often, the user is thinking through their answer, and the most thoughtful responses come after a pause. If you jump in to rephrase or offer options, you rob them of that thinking time and bias the response.

Silence is your friend in interviews. Get comfortable with it.

Take Notes Without Losing Eye Contact

I recommend recording interviews (with permission) so you can focus on the conversation rather than scribbling notes. A simple "Do you mind if I record this so I can focus on our conversation instead of taking notes?" works for almost everyone.

If recording is not an option, have a second person take notes while you run the conversation. This two-person approach, where one person asks questions and the other captures everything, produces much better results than trying to do both yourself.

Mistakes That Ruin Customer Interviews

I have conducted hundreds of customer interviews and made every one of these mistakes at least once. Learn from my failures.

Mistake 1: Leading the Witness

"Don't you think it would be great if we added a dashboard for tracking feature requests?" That question contains the answer you want to hear. The user nods along because disagreeing feels confrontational. And you walk away convinced you should build a dashboard, when the real need might be something entirely different.

Keep questions neutral. Instead of "Would X be useful?" ask "How do you currently handle X?" Let the user's experience guide the conversation, not your hypothesis.

Mistake 2: Talking More Than Listening

The ideal ratio in a customer interview is 80/20. The customer talks 80% of the time. You talk 20%. If you catch yourself explaining your product, pitching features, or defending design decisions, you have flipped the ratio and the interview is no longer producing useful data.

Your job in an interview is to ask questions and listen. That is it. Save the explanations for after the research is done.

Mistake 3: Asking About the Future Instead of the Past

"Would you pay for a tool that does X?" is one of the worst questions you can ask. People are terrible at predicting their future behavior. They say yes to everything that sounds good in theory and then never use it in practice.

Instead, ask about what they have already done. "Have you ever paid for a tool to help with X? Which one? What did you like and dislike about it?" Past behavior is the best predictor of future behavior. Someone who has already spent money solving this problem will spend money on your solution. Someone who just thinks it sounds nice probably will not.

Mistake 4: Only Interviewing When Something Is Wrong

Many teams only do customer interviews during a crisis. Churn spikes, so they scramble to understand why. A launch fails, so they talk to users after the damage is done. This reactive approach means you are always learning too late.

The best product teams interview continuously. Two to four conversations per week, every week, builds a running understanding of your users that makes every product decision sharper. You catch problems before they become crises. You spot opportunities before competitors do.

Mistake 5: Not Closing the Loop

Users give you their time and their honest feedback. If you never follow up to tell them what you learned or what you are building as a result, they feel ignored. And they will not agree to another interview.

Closing the feedback loop means circling back to interview participants when you act on their input. A short email, "Based on our conversation and similar feedback from others, we are building X. It will ship next month," turns an interview participant into a loyal advocate.

How to Turn Interview Insights Into Product Decisions

Raw interview notes are useless if they sit in a document nobody reads. The real work happens after the interviews, when you synthesize what you heard into patterns that guide product decisions.

Step 1: Debrief Immediately After Each Interview

Within 30 minutes of finishing an interview, write down your top three takeaways. What surprised you? What confirmed something you suspected? What contradicted your assumptions? Memory fades fast, and the nuances that make interview data valuable are the first things you forget.

Step 2: Look for Patterns Across Interviews

After completing a batch of interviews (5 to 8 on the same topic), lay out your notes and look for themes. I use a simple method: write each distinct insight on a sticky note or card, then group similar ones together. The largest groups represent the strongest patterns.

Be honest about what you find. If the data contradicts your pet hypothesis, let the hypothesis go. The whole point of interviews is to replace assumptions with evidence.

Step 3: Connect Interview Data to Other Feedback Sources

Interview insights become more powerful when you combine them with quantitative data. If five interview participants mentioned confusion about pricing tiers, check your analytics. Is there a high drop-off rate on the pricing page? Do support tickets about pricing spike at certain times?

When qualitative insights (from interviews) align with quantitative signals (from analytics and support data), you have a strong case for action. RoadmapAI captures feature requests from community conversations and organizes them by theme, giving you a quantitative layer on top of your qualitative interview findings. If interviews reveal that users struggle with feedback organization, and your feature request data shows 200 people asking for better categorization, that alignment makes the prioritization decision clear.

Step 4: Share Findings With Your Team

Interview insights should not live in the researcher's head or in a 30-page report nobody reads. Create a one-page summary for each research batch that includes:

• Research question: What were you trying to learn?
• Who you talked to: Number of interviews, user segments, relevant demographics
• Top 3 to 5 findings: The strongest patterns, each supported by 2 to 3 direct quotes
• Recommended actions: What should the team do with this information?
• Open questions: What needs more research?

Share this summary in your team's Slack channel, present it at your next product meeting, and attach it to relevant items on your product roadmap. The more visible interview insights are, the more they influence decisions.

Step 5: Feed Insights Into Your Prioritization Process

Interview findings should directly influence what you build next. When you use a feature prioritization framework, interview data adds depth that request counts alone cannot provide. A feature requested by 50 users is less compelling if interviews reveal they would not actually change their workflow to use it. A feature requested by 10 users becomes urgent if interviews show it is the reason those users are evaluating competitors.

The combination of interview insights with structured feedback data from tools like a feature voting board gives your team both the depth (from interviews) and the breadth (from aggregated requests) to make confident product decisions.

Building a Continuous Interview Practice

The real payoff from customer interviews comes when they become a habit, not a one-time project. Here is how to build a sustainable interview practice that fits into a busy product team's schedule.

Set a Weekly Interview Quota

I recommend starting with two interviews per week. Block 30 minutes for each interview plus 15 minutes for the debrief. That is less than two hours per week, which is a fraction of the time most product managers spend in internal meetings that produce far less insight.

Once two interviews per week feels natural, increase to three or four. The teams that maintain a steady interview practice build an intuitive understanding of their users that makes every product decision faster and more accurate.

Create a Recruitment Pipeline

The biggest bottleneck for regular interviews is finding people to talk to. Build a system that keeps your pipeline full:

• In-app recruitment: Add a subtle prompt in your product: "Help shape our roadmap. Chat with our product team for 20 minutes." Target it to specific user segments based on your current research questions.
• Post-support follow-up: After resolving a support ticket, ask: "Would you be open to a 20-minute call to help us improve this experience?" Users who just had a frustrating experience are often eager to share more context.
• Feature request follow-up: When users submit feature requests through your feedback system, invite them to discuss it in more detail. They have already demonstrated interest in improving the product.
• Onboarding invitations: During the first week after signup, invite new users to share their experience. Their fresh eyes catch things long-time users have learned to work around.

Rotate Interviewers Across the Team

Customer interviews should not be one person's job. When designers, engineers, and product managers all participate in interviews, the entire team develops empathy for users. An engineer who hears a customer describe a painfully slow export process will prioritize performance differently than one who only sees a ticket in the backlog.

Set up a rotation where different team members join interviews each week. Pair experienced interviewers with newer ones so skills transfer naturally. The more people on your team who hear directly from users, the more user-centered your product decisions become.

Customer Interviews for Specific SaaS Scenarios

Interviews for Reducing Churn

Churn interviews are some of the most valuable conversations you will ever have. A user who just canceled knows exactly why they left, and that information is worth its weight in gold for your churn reduction strategy.

Reach out within 48 hours of cancellation. The longer you wait, the less detail they remember and the less they care. Keep the ask simple: "I noticed you canceled. Would you be open to a 15-minute call? I want to understand what we could have done better. No sales pitch, just learning."

Questions that work for churn interviews:

• "What originally brought you to [product]? What problem were you trying to solve?"
• "At what point did you start considering leaving?"
• "Was there a specific moment or event that triggered the decision?"
• "What are you using now instead?"
• "If one thing had been different, would you have stayed?"

Interviews for Validating New Features

Before committing engineering time to a new feature, interview users who are experiencing the problem it would solve. The goal is not to pitch the feature. The goal is to understand the problem deeply enough to build the right solution.

Start with the problem, not the solution. "How do you currently handle [the thing this feature would address]?" If they describe a painful process full of workarounds, you have validation that the problem is real. If they shrug and say it is fine, maybe the feature is not worth building.

Only after you understand the problem should you introduce the concept. "We are considering building something that would [describe the outcome, not the feature]. How would that fit into your current workflow?" Watch their reaction. Genuine excitement looks different from polite agreement.

Interviews for Improving Onboarding

New user interviews reveal the gap between how your team thinks onboarding works and how users actually experience it. Follow the approach described in the SaaS onboarding best practices guide, and supplement it with direct conversations.

Interview users within their first week. Ask them to walk you through their experience from signup to their current state. Where did they get confused? Where did they feel confident? What almost made them give up? The answers reshape your onboarding more effectively than any heatmap or funnel analysis.

Tools and Logistics for Running Interviews at Scale

You do not need fancy tools to run great customer interviews. But a few practical choices make the process smoother.

Scheduling: Use Calendly or a similar tool with a dedicated booking page for research interviews. Include available time slots and let participants pick what works for them. This eliminates the back-and-forth scheduling emails that waste time and reduce participation rates.

Recording: Zoom, Google Meet, or any video call tool with built-in recording works. Always ask permission first. Most people say yes.

Note-taking: Dedicated note-taking tools for research (like Dovetail or Notion) help you tag and organize insights across multiple interviews. If you are just starting out, a shared Google Doc with a consistent template works fine.

Incentives: For B2B SaaS, a $25 to $50 gift card for a 30-minute interview is standard. Some users will talk for free because they care about the product. But offering a small thank-you increases participation rates and shows you value their time. For users who prefer not to accept incentives, offer to donate to a charity of their choice.

Feedback integration: Connect your interview findings to your broader product feedback strategy. When interview insights confirm patterns you see in feature requests and support data, the case for action becomes impossible to ignore.

Frequently Asked Questions

How long should a customer interview last?

Aim for 25 to 35 minutes. Shorter interviews feel rushed and do not give enough time for deep exploration. Longer interviews cause fatigue for both the interviewer and the participant. Tell participants upfront that the call will take 30 minutes and respect that boundary. If the conversation is going well and the participant is willing, you can extend by 10 minutes, but always ask first.

How many customer interviews do I need for reliable insights?

For a focused research question, 5 to 8 interviews typically reveal the major patterns. Nielsen Norman Group research shows that 5 participants uncover about 85% of usability issues. For broader discovery research or validating a new product direction, aim for 15 to 20 interviews across different user segments. You will know you have done enough when new interviews stop revealing new themes.

What if customers do not want to be interviewed?

Low participation usually means the ask is unclear or the friction is too high. Make the invitation specific ("20-minute call about your onboarding experience"), explain why their input matters ("Your feedback directly shapes what we build next"), and reduce scheduling friction by sharing a booking link with open time slots. Offering a $25 to $50 incentive also increases response rates by 2 to 3 times for B2B users.

Should product managers or researchers conduct interviews?

Both. Dedicated researchers bring interviewing expertise and reduce bias. Product managers gain direct empathy with users that informs daily decisions. The best approach is for researchers to lead interviews with PMs observing and occasionally asking follow-up questions. If your team does not have a dedicated researcher, train PMs on interviewing techniques and pair them with a teammate who takes notes.

How do I avoid bias in customer interviews?

Ask open-ended questions about past behavior rather than hypothetical future scenarios. Do not describe your planned solution before asking about the problem. Let silences happen instead of filling them with suggestions. Interview users from different segments, not just your biggest fans. Review your discussion guide with a colleague to catch leading questions before the interview happens. Recording interviews helps you review your own technique and spot patterns of bias over time.

How do customer interviews connect to product roadmap decisions?

Interview insights add qualitative depth to your feature prioritization process. When you combine what users tell you in interviews with aggregated data from feature requests and support tickets, you get a complete picture of demand and urgency. Tools like RoadmapAI capture feature requests from community channels automatically, giving you the quantitative layer. Interviews give you the "why" behind those requests, which helps you design better solutions and prioritize more accurately.

Sources

• Rydoo, "Why 90% of Startups Fail," October 2024, https://www.rydoo.com/cfo-corner/why-startups-fail/
• User Interviews, "The State of User Research Report 2024," June 2024, https://www.userinterviews.com/state-of-user-research-2024-report
• Nielsen Norman Group, "How Many Participants for a UX Interview?," November 2024, https://www.nngroup.com/articles/interview-sample-size/
• Steve Blank, "Customer Development," https://steveblank.com/category/customer-development/
• SaaStr, "Planning to Do a SaaS Startup? Don't Forget the 20 Interview Rule," November 2022, https://www.saastr.com/planning-to-do-a-saas-start-up-dont-forget-the-20-interview-rule/