Customer Discovery 101

At the heart of any successful startup is a deep understanding of its customers. Customer discovery is the process of learning about your target users – their problems, needs, and behaviors – so you can build something they truly want. In plain terms, it’s about getting out there and talking to people (or observing them) before you build anything. This guide will walk you through several popular customer discovery methods, each with practical tips to help you get started. The tone here is relaxed and conversational – imagine a mate in his 30s walking you through the basics over a coffee. By the end, you’ll have a better sense of where to begin and what to expect from each method, so you can choose the approach that fits you best.

Customer Interviews

A one-on-one customer interview can be as casual as a friendly chat – here a founder listens to a user’s experience while taking notes.

Talking directly with customers is often the first thing people think of for discovery – and for good reason. Customer interviews let you dive deep into someone’s thoughts and experiences. It’s just you (and maybe a co-founder) having a chat with a potential user about their problems and how they currently solve them. If you’re new to this, you might be wondering: How do I find people to talk to? How many interviews do I need? And how do I make sure I’m hearing the truth, not just what I want to hear? Don’t worry – the following tips will help you tackle those questions:

- Find participants in your target audience: Start with people who resemble your ideal customer. This could mean posting a call for volunteers in relevant online communities, reaching out through your personal network, or even approaching folks in real life where your audience hangs out (for example, dog park visitors if your idea is for dog owners). You might be surprised – many people are willing to talk about their frustrations if you just ask sincerely. If you’re struggling, consider incentives like a gift card or a free coffee, but often genuine curiosity and a polite request are enough.

- Aim for a handful of interviews to start: You don’t need to interview dozens of people to get valuable insights. In fact, even 3-5 interviews can surface major themes, and many seasoned product teams find that after about 8-10 interviews, they start hearing the same points on a given topic. So begin with a small number (say 6) and see if patterns emerge. Quality matters more than quantity here – the goal is to learn, not to hit a huge sample size. Later, you can always do more interviews or switch up methods to validate what you heard.

- Keep it casual and avoid biases: Think of an interview as a guided conversation, not an interrogation. Start with open-ended questions (“Can you walk me through how you do X currently?”) and let the customer do most of the talking. Avoid leading questions that push them toward confirming your assumptions. For example, instead of “Don’t you find it annoying when…?”, ask “How do you feel when…?”. The idea is to hear their genuine thoughts, not to get them to say what you hope is true. Maintaining a curious, I’m here to learn mindset helps a lot – you’ll get more honest input if the person senses you’re not there to pitch or argue.

- Dig deeper with follow-ups: One of the best phrases in an interviewer’s toolkit is “Why?” (asked in a friendly, non-judgmental way). If a customer mentions a problem – say, “I hate how long it takes to do X” – gently ask, “Why is that?” or “What makes it so frustrating?”. Follow-up questions help uncover the root causes of their pain points and reveal what truly matters to them. Keep asking “why” (or “tell me more about that”) until you hit either a fundamental reason or you risk annoying them. Often, the real gold lies a few layers down. Just be careful to stay neutral; you’re not questioning them, you’re clarifying their experience. And if something they say is unclear, don’t pretend to understand – politely ask them to elaborate, so you don’t misinterpret anything.

- Create a comfortable atmosphere: People open up when they feel at ease. Start the conversation with some small talk or a genuine compliment to break the ice. Let them know there are no right or wrong answers and that their honesty is super valuable – even if it’s negative, it’s helpful. If you can, interview in a context that’s convenient and non-threatening for them (their office, a cafe, a Zoom call at a time that suits). As they talk, show that you’re listening actively – nod, maintain eye contact, say “mm-hmm” – but try not to interject your own opinions. Sometimes a short silence or a thoughtful pause from you will prompt them to continue and share even more. Remember, listen more than you speak. You’re here to learn about their world.

(And a quick practical tip: ask for permission to record the interview, or at least take detailed notes. Recording (audio or video) can free you from scribbling furiously and allows you to review later for nuggets you might have missed. Just be sure to ask first and respect their comfort – most people won’t mind, especially if you explain it helps you focus on the conversation.)

Surveys

Not everyone has time for a chat, and sometimes you want feedback from a larger group. Surveys are a ok-ish discovery method to gather input at scale. Essentially, you’ll be sending out a questionnaire – could be a Google Form or a Typeform – to get quantitative data or quick hits of opinion from many people at once. Surveys are less personal than interviews, but they’re fast and cost-effective. If you’re wondering how to create a survey that people actually respond to (and that gives useful insights), consider these pointers:

- Keep it short and focused: These days, people’s attention spans for surveys are really short. Survey fatigue is real, and if someone opens yours and sees 50 questions, they’ll likely drop it. Aim for a concise survey that takes 5 minutes or less to complete. Stick to the key questions that tie back to what you really need to learn (your hypothesis or uncertainty). It helps to start by clearly defining for yourself what decision or assumption each question is meant to inform. If it doesn’t map to a goal, consider cutting it. A good survey might have, say, 5-10 questions max. You can always do another survey later for other topics.

- Use simple, unbiased wording: Write questions that are easy to understand and don’t push people toward a certain answer. For example, instead of asking “How much do you love the current solution for X?” (which assumes they do love it), ask “How do you feel about the current solution for X?” and offer a balanced scale. Where possible, use close-ended questions (multiple choice, rating scales, yes/no) to get structured data you can analyze. If you use rating scales (e.g. 1 to 5 for “How difficult is…?”), keep the scale consistent throughout the survey to avoid confusion. It’s also wise to include an “Other / Additional comments” open text field at the end in case respondents have insights you didn’t think to ask – but make this optional.

- Pilot test your survey: Before blasting your survey out to 100 people, test it on a friend or two (ideally someone who fits your target user profile). Watch them take it (or have them describe their thought process) and see if any questions confuse them or if the response options don’t cover their situation. A quick pilot can catch ambiguous wording or technical glitches. It’s much better to find out before you send it wide that question 3 made no sense to respondents.

- Distribute smartly (and consider incentives): To get responses, send your survey through channels where your potential customers are likely to see it. This could be an email list, a relevant subreddit or Facebook group (check the rules first!), Slack communities, or even Twitter/LinkedIn if that’s where your crowd hangs out. When posting, include a short blurb about why you’re doing the survey and how their feedback will help – people like to know there’s a human and a purpose behind the form. If responses are trickling in slowly, you might nudge folks with a small incentive. It could be “Complete this survey for a chance to win a £20 Amazon voucher” or “We’ll send you a summary of the findings” (the latter works if the topic itself is interesting to them). Incentives aren’t always necessary, but they can boost participation if you’re asking a lot. Just make sure any prize isn’t skewing who responds (for instance, an iPad giveaway might attract random prize-hunters who aren’t your target users).

- Analyze with care: Once the results are in, look for patterns in the data that inform your original questions. If 80% of respondents say they perform a certain task on mobile, that’s a strong signal. Be careful not to over-index on the numbers if your sample size is small, though. For example, if 5 people answer and 4 of them say A and 1 says B, it’s tempting to declare 80% prefer A – but five people is too few to be sure. Use small survey results directionally, not as absolute fact. And always cross-check with any qualitative insights you have (from interviews or elsewhere). The survey can tell you what or how many; combine it with other methods to understand why.

(Oh, and one more thing: respect people’s time and privacy. If you promised a summary or offered a prize, follow through. And never share individual responses publicly – aggregate data only – unless you’ve gotten explicit permission. Building goodwill with potential customers now means they’ll be more likely to help you again later!)

Focus Groups

Focus groups bring several people together at once to discuss your idea, product, or just the problem area in general. Think of it like hosting a small roundtable chat with members of your target audience. This method can be super insightful because participants might bounce ideas off each other, sparking thoughts they might not have had in a one-on-one interview. However, as a beginner, running a focus group can feel a bit like juggling – you’re managing multiple people and group dynamics. Common newbie questions include: How many people do I invite? How do I make sure everyone gets a say (and one loudmouth doesn’t dominate)? And what do I even ask them? Let’s break it down:

A focus group session in action – a small group gathered around discussing a prototype on screen. A good mix of participants and a relaxed setting can generate lively, honest discussions.

- Keep the group small and diverse: In focus groups, less is more. Aim for around 5 to 8 participants in a session – enough to generate discussion, but not so many that voices get drowned out. If you invite, say, 12 people, it tends to split into side conversations or some folks won’t speak up at all. With about 6 people, you can usually hear each person’s perspective. Also, try to include a mix within your target audience – for example, different ages, backgrounds, or slightly different use cases – so you don’t get a one-sided view. Everyone in the group should share the common problem or context you’re exploring, but variation in perspectives is healthy for discussion. And practically speaking, expect a couple of no-shows; if you need 6, you might invite 8 to account for this.

- Be prepared, but stay flexible: Successful focus groups feel conversational but have a guiding structure behind them. Beforehand, prepare a discussion guide – a list of topics or open-ended questions you want to cover. This could include things like “What’s the biggest challenge you face with X?” or an activity like reviewing a concept or prototype. Start with broader questions, then drill down into specifics. However, once the group gets talking, be ready to go with the flow. If they veer into a relevant tangent where everyone is engaged and you’re learning new stuff, that’s fine! The guide is there to steer back if the discussion completely derails or hits a lull. As the moderator, your job is to keep the conversation on track and on topic, but not to micromanage every moment.

- Moderate to include everyone: In any group, some people are naturally more talkative while others are shy. Early on, make it clear that you want to hear from each person. You can set this expectation by directly inviting quieter participants to share (“Sarah, what’s your experience been like?”) and politely curbing the over-enthusiastic talkers (“Thanks John, let’s hear what others think too”). Sometimes a gentle tactic is to go “around the room” for a particular question so each person speaks in turn. Encourage a respectful atmosphere where differing opinions are welcome – you might even say “It’s okay if you disagree with each other; we’re here to learn all viewpoints.” Also, watch for any consensus that forms too quickly; group settings can lead to peer pressure or groupthink. One trick: have everyone quietly jot down their individual answer to a critical question before discussing, then share – this way, each person’s true feelings come out, not just what the first outspoken person said.

- Logistics and comfort matter: Make it easy for people to participate. Choose a convenient time (outside work hours if they’re working professionals, for example) and a comfortable venue. This could be a casual meeting room, a coffee shop private area, or even a virtual video group call if your audience is spread out. If in person, providing a few snacks or drinks can create a warm, inviting environment (and who doesn’t open up over some pizza or biscuits and tea?). If virtual, ensure the video platform is one everyone can use without hassle and consider features like screen sharing if you plan to show any materials. At the start, do quick introductions to break the ice – maybe a fun fact round – to get people familiar with each other. When folks feel at ease in the setting and with the group, they’ll speak more freely, which is exactly what you want.

(One more note: like interviews, recording a focus group (audio/video) is helpful since it’s impossible to take detailed notes when multiple people talk. Just be sure to get consent from everyone before you hit record, and assure them it’s just for your team’s use to remember what was said. You might even offer to share the transcript or a summary of the discussion with participants afterward, so they feel part of the process.)

Usability Testing

You might be thinking, “Usability testing? I don’t even have a product yet!” That’s fair – not every startup founder will do usability tests in the very early discovery phase. However, if you do have a prototype, a landing page, or even just a demo of an idea, putting it in front of users and watching them interact can reveal tons of insights. Usability testing is essentially observing someone as they try to use a product or perform a task, then noting what works and what doesn’t. It’s more about how they use something than what they need (so it’s slightly different from an open-ended interview), but it’s incredibly useful to discover pain points. For a beginner, key questions include: How do I run a usability test? How many people should I test with? And what am I looking for? Here’s how to get started:

- Define what you want to learn: Before you invite anyone to test, be clear on your objectives. Are you trying to find out if people understand how to navigate your app’s sign-up process? Or maybe whether the solution you have in mind actually solves their problem in a convenient way? Having a focus will help you create tasks for the user to perform. For example, “Find and add a product to your cart” could be a task if you’re testing an e-commerce prototype. Write down a handful of tasks or scenarios that cover the core things you want to observe. Keep them realistic (“Imagine you want to do X, show me how you’d go about it here”) and avoid giving away how you would do it – let them show you their way, or where they get stuck.

- Recruit the right testers: Just like interviews, you want participants who resemble your target users. They don’t have to be the exact perfect customer profile at this stage, but the closer the better. If your product is for accountants, grabbing a random college student to test might not yield relevant feedback. Reach out to the same channels as you would for interviews or surveys. Sometimes friends and family can help as testers if they fit the demographic – just ensure they feel free to be honest (people close to you might hold back criticism). It’s often worth stating up front, “You won’t hurt my feelings – if something is confusing, I need* to know.”* Also, test one person at a time, not in a group, so you can focus on their individual experience.

- Be a neutral observer during the test: This is critical. When the person is using your product or prototype, fight every urge to coach or correct them in the moment. If they’re clicking the wrong thing or looking in the wrong place, that’s valuable data! It means your design or concept isn’t as intuitive as you thought. Let them flounder a bit (within reason) and see if they eventually find the solution or if they give up. Pay attention to any signs of frustration (sighs, long pauses, repeated clicks). Encourage them to “think out loud” – ask them to narrate what they’re trying to do and what they’re thinking as they do it. For example, they might say “I’m looking for the settings… I guess it might be under this menu… nope, not there.” Those comments are user experience gold. Resist jumping in with, “Oh, it’s over there”. Instead, ask, “What did you expect to happen?” or “What are you looking for?” to understand their mental model. Only intervene if they’re completely stuck and you need them to move on to get through your test scenario – and even then, note that as a big red flag.

- Take notes of issues and positives: As you watch, jot down not just the problems, but also what worked well. Maybe everyone breezed through the sign-up but stumbled on the next step – that tells you sign-up is fine, but something later isn’t. After each test session, while it’s fresh, quickly summarize the key observations: e.g., “3 out of 5 users couldn’t find the Save button without help”. This will help you spot patterns once you’ve done a few tests. You might also record the session (screen recording and audio, or video if in-person) – with permission – so you can replay later and catch details you missed. It can be painful to re-watch someone struggling with your baby, but it’s the path to improvement.

- You don’t need a crowd of testers: Good news – usability testing doesn’t require a huge sample. Testing with just 6 people uncovers the majority of common issues, according to so called UX research experts. Beyond the first handful of users, you see diminishing returns. So start with 6 (one at a time) and see what you learn. If you discover serious problems, fix those first, then test again with a new set of users. This iterative approach means you’re always testing a slightly improved version and learning new things. If your product has very distinct types of users, you might do 5 of each type to cover the differences. But in general, testing a small number in each round is faster and more effective than one big test with 20 people at once. It’s also less overwhelming to process the feedback.

(Side note: Even if you don’t have a prototype of your own yet, you can do a form of usability testing on competitors’ products or on the way users currently solve the problem. Watching someone go through the motions with an existing tool or workaround can highlight pain points (aka opportunities for you). For example, if your idea is to simplify budgeting, have a user show you how they budget today using Excel or an app – you’ll quickly see what’s clunky for them. This is a sneaky way to do discovery and usability research in one go.)*

Observation & Ethnographic Research

Sometimes, the best way to learn about your customer is to watch them in their natural habitat. Observation-based research (often called ethnography when it’s more formal) involves seeing what people do, rather than just listening to what they say. This method is all about immersing yourself in the user’s environment – whether that’s in a store, at their workplace, or online – and quietly noticing behaviors, workflows, and problems. For instance, if you’re building a tool for cafe owners, you might spend a day at a café observing how the staff works and where they encounter hiccups. The big benefit here is you can uncover unmet needs or pain points that users themselves might not articulate, because it’s just part of their routine. If you’re new to this, you might be unsure how to conduct observations or what to look for. Here are some tips to get you started:

- Plan where and what to observe: First, identify a situation that’s relevant to the problem you’re interested in. It could be a location (like a supermarket if your idea is grocery-related, or a hospital waiting room if it’s healthcare) or a process (like observing someone cooking dinner if you’re designing a kitchen gadget). Go where the action is. Sometimes this means literally going out into the field; other times, you might do it virtually (e.g. observing user sessions via screen-sharing software, or even just watching how people behave in online forums or communities). Have a rough idea of what you’re looking to learn, but remain open-minded. For example, you might focus on “How does a shop manager close up at end of day?” and observe all the steps and tools they use.

- Be as unobtrusive as possible: When observing, try not to interfere with the natural flow. Ideally, the people you’re observing should forget that you’re there (or not even know you’re observing, if you’re in a public space where that’s appropriate). Obviously, ethics and permission are important – if you’re in a private setting or shadowing someone specific, get their consent and maybe the organisation’s approval if needed. You can frame it as “Do you mind if I tag along and watch? I’m trying to learn about how XYZ is done in real life.” Many people are actually flattered by this interest. Once you’re on site, blend into the background. Take notes discreetly (carry a little notepad or use a notes app on your phone – you might even look like just another person texting). If someone asks who you are, be honest that you’re researching to build a better solution for this kind of task.

- Note details and context: As you watch, pay attention not just to what the person does, but also to how and why. Are they improvising any workarounds? (E.g., paper sticky notes on a screen – a sign their software doesn’t support something they need). Do they show visible frustration at any point? Is there an obvious time sink or a bottleneck where things pile up? Also note the environment – noise, interruptions, tools they keep close at hand. All these contextual clues can spark ideas for features or improvements. For example, noticing that a barista has to write down mobile orders from one device onto another could hint at integration issues. You might sketch a little journey of the steps you observe. If possible, take photos of the environment or artifacts (not people, unless you have permission) – like a photo of that wall of sticky notes or the messy cable setup they deal with. These serve as visual notes for later analysis.

- Ask spur-of-the-moment questions (sparingly): During pure observation, you generally don’t want to interrupt. But if you’ve built some rapport and the timing feels right (say there’s a natural break or the person is doing something and not minding a bit of chat), you can ask brief questions to clarify what’s going on. For instance, “I noticed you did X just now. Could you tell me why?” Keep it light and curious. Don’t turn it into a full interview on the spot – remember, your primary role in this method is to watch and learn. If they start explaining a lot, certainly listen, but you might also say “This is great – I’d love to ask you more about that later” and then indeed follow up afterwards in an interview format. That way, you keep the observation focused and can dive deeper once the observed session is over.

- Combine observation with other methods: Often, observation is a starting point and you’ll pair it with interviews. For example, you might shadow a nurse during a shift (observation), then interview her after to ask about some of the things you saw. This combo is powerful – you can refer to specific moments (“I noticed you had to switch between three different systems there; is that typical?”) to get rich insight. Observing real behaviour can reveal needs or pain points that users themselves might not mention. People sometimes take inefficient routines for granted (“that’s just how it’s always been”), but your fresh eyes can spot those inefficiencies as opportunities. By the end of a good observation session, you’ll likely have a list of “aha!” moments and new questions you want to explore further.

Introducing the GRAMS Framework

As you dive into customer discovery, especially interviews, you’ll quickly realise that asking the right questions is an art. This is where the GRAMS framework comes in handy – it’s like a cheat sheet for structuring your customer conversations to ensure you cover all the important bases. GRAMS stands for Goals, Reality, Alternatives, Meaning, and Solutions. It’s a framework originally developed to help researchers and product folks dig deeper in interviews, and it can guide even a first-timer to uncover golden insights. Let’s break down what each part means in practice:

- Goals: What is the person ultimately trying to achieve? In other words, what’s their end goal or job-to-be-done? You want to understand the outcome they care about. For example, if you’re talking to someone about their fitness routine, their goal might be “stay healthy and feel energetic,” not just “go to the gym.” When you know their goal, you see the big picture of what they value.

- Reality: How are they achieving that goal today? This is all about their current process or workaround. Ask them to walk you through how they do things now, step by step. Sticking with the fitness example, what’s their weekly routine? Do they use any apps or just wing it? Who or what do they rely on? Uncovering the reality of their routine – including the tools, time spent, and challenges – gives you context and often highlights pain points. You’re basically mapping out their status quo.

- Alternatives: What alternatives have they tried (or considered)? People often switch solutions or have tried different methods to reach their goal. Maybe your fitness enthusiast used to attend group classes, but now uses a workout app, or is considering hiring a personal trainer. By discussing alternatives, you learn why certain options didn’t work for them or what they feel might be missing. This can reveal gaps in the market and help you position your idea. It also tells you who/what your future competition really is.

- Meaning: Why does this matter to them? This digs into the personal significance or underlying motivation. It’s a bit abstract, but incredibly powerful. Essentially, you’re exploring the deeper “why” behind their goal and behaviors. In our example, why do they want to stay healthy? Perhaps it’s to be a good role model for their kids, or because they had a health scare. Understanding the meaning can connect you to their emotions and values. It’s the difference between a functional need and an emotional driver. Often, asking something like “What’s at stake if this problem isn’t solved?” or “How does it make you feel when you can’t achieve [X]?” opens people up about the meaning. As one "expert" puts it, it’s not about what they drive, but what drives them (That'll be me :). These insights inspire messaging and features that resonate on a deeper level.

- Solutions: What do they wish for? or Which pain points do they most want to solve? This is where you gather what the person thinks would help them. Note, it’s not necessarily where you pull out your solution (be careful not to turn the interview into a pitch unless you’re specifically testing a concept). Instead, encourage them to describe in an ideal world how their life could be better. “If you had a magic wand, what would you change about [current process]?” You’ll hear their priorities – maybe they say “I wish my workouts could adapt to my schedule automatically” or “I need someone to just keep me accountable.” These are hints at features or value propositions that would matter to them. When they do mention specific pain points or desires, ask which are most urgent or important. This helps you prioritise. By understanding desired solutions from the customer perspective, you ensure that what you build addresses something they genuinely care about.

Using the GRAMS framework as a guide, you can structure your interviews to hit all these notes. It prevents that awful feeling of ending a conversation and realizing you forgot to ask something crucial. It’s also a way to avoid bias – because you’re systematically covering the user’s world (Goals/Reality), not just pitching your idea, and you’re digging into why things matter (Meaning) before jumping to how to fix it. Many founders find that GRAMS brings a nice flow: start broad with their goals, get the gritty details of their reality, explore past solutions, then get reflective about meaning, and finally talk about ideal solutions.

Even if you don’t follow it strictly, keeping these five elements in mind will make your discovery conversations more productive. You’ll walk away with a richer understanding of your customer – not just what they do, but why they do it and what they wish could be better. And that’s exactly the insight you need to build a product that truly resonates.

---

In summary, customer discovery is all about staying curious and learning from real people. Whether you’re interviewing someone over Zoom, sending out a survey, running a focus group in a coffee shop, watching a user try out your prototype, or quietly observing a day in their life, each method gives you a different perspective. There’s no one “right” way to start – pick the method that you’re most comfortable with and that fits what you need to learn first. You can always mix and match techniques as you go. The key is to listen and observe more than you talk, embrace whatever you learn (even if it’s not what you expected), and keep iterating on your understanding of the customer. Treat this guide as a menu of discovery options and dive in. The fact that you’re taking the time to do this at all already sets you on the path to building something people will love. Happy discovering!

PS. If you want to quickly master the art of the customer/stakeholder interview, check out our courses at https://whatdrivesthem.com