Over $45B is spent by companies each year on market research and more than 40% of companies have a dedicated team for user research. Despite this effort, a quick Google search of ‘product flops’ says it all: watermelon-flavored Oreos, purple EZ-Squirt ketchup, and even Google Glass.

So why is it so hard to understand consumers and what they want? Can’t we just ask? Well, yes, but asking is only the first step.

Customers are (not) always right. What we, as customers, say we want is often different than what we actually want. Truly understanding customers requires listening, but also careful observation and product testing.

Before joining Lightspeed, I led several user research projects for startups including Thumbtack, BetterUp, Fundbox, and a pre-launch Indian beauty brand. Time and time again, I noticed that customers would say one thing, but mean another. When they said they wanted more choice, they actually meant they wanted fewer, yet more curated choices. When they said they wanted to be inspired, they first wanted to be educated.

Over time, I’ve curated a set of techniques to help me better understand customers — in aggregate, a way to triangulate what consumers really need. In the same way a founder can triangulate total addressable market or projected growth rate, I believe a founder can also triangulate customer need. Listening, observation, and product testing techniques form the foundation of this approach.

Below I’ve included some example techniques for each — listen, observe, test. Think of this list as a menu. For any research project, you can select the most appropriate techniques based on what you’re trying to learn and level of effort required.

I’ve also included some real-life anecdotes below. The photos are from my experience helping my friend, Divya Vishwanath, launch her Indian beauty brand. Thanks to Divya for allowing me to share these photos with you.

—

Listen

“If I had asked people what they wanted, they would have said faster horses.”

(Henry Ford)

Listening is where most companies start. And it’s unfortunately where most end. While listening is a good place to start, it will almost always lead you down the wrong path if you take what the customer says at face value. Instead of just focusing on the ‘what,’ focus also on the ‘why.’ Try to understand your customer’s motivation and aspiration. Build a hypothesis that can serve as a starting point for further observation and testing.

For the Indian beauty brand, we spent over 30 hours listening to our customers — at the mall, at local Indian stores, and even in their homes. One thing we heard over and over again is that people didn’t want turmeric-based products. If we had stopped there, we would’ve missed a key insight — that the concern was about staining vs. the actual ingredient. By asking ‘why’ over and over, we realized people actually loved turmeric’s anti-inflammatory and acne-fighting properties, they just didn’t want any staining. The first product we ever trialed used non-staining turmeric, a key ingredient in Divya’s grandmother’s recipes. However, because of our extensive customer interviews, we focused a lot more on customer education and messaging than we otherwise would’ve, given existing misconceptions.

There are several ways to listen to your customer. I prefer to start with the informal, like mall intercepts, to understand who the customer is and what they care about. From there, I develop a longer set of questions for more formal sit-down interviews.

Mall intercept

Mall intercepts are bite-size customer interviews. They are helpful when first getting to know your customer. They’re also helpful when asking about shopping preferences and routines, given customers are already primed by the environment. You should ask 2–3 precise questions and move-on. Here you’re optimizing for volume and breadth vs. depth. Use insights gleaned here to inform a more robust set of questions for longer interviews.

Customer coffee

Take your customer (or hypothetical customer) out to coffee for a 30–60 minute chat. Prepare an initial set of interview questions grouped by topic, but use this as a starting point. Know that some of the best information can be gleaned from a customer going ‘off script.’ If the topic starts to veer, let that happen and eventually get back on track. Ask why … ask why again … and ask why once more (IDEO’s “5 whys”).

I learned a ton about effective customer interviewing in Stanford’s Startup Garage class. The Stanford d.school has a concise and effective list of customer interviewing tips here.

Observe

“When you meet chimps you meet individual personalities. When a baby chimp looks at you it’s just like a human baby. We have a responsibility to them.”

(Jane Goodall)

Go out into the wild and spend time with your customers. Observe their routine and try it out for yourself. See what it’s like to run a small business or be a beauty influencer. In the words of Jane Goodall, see what responsibility you have to them.

When making product updates at Thumbtack, I often met with our service professionals in-person to observe them using the product in-situ. On one afternoon, I drove to San Francisco’s Mission District to meet a Thumbtack caterer in her home. She was in the process of prepping for a dinner party that evening. As I observed her messaging with customers, I noticed that she checked her wall calendar several times.

Recognizing the inconvenience, we talked about why she used the wall calendar vs. something like Google Calendar and what she would want to see if Thumbtack were to enable scheduling. The information gleaned from that one afternoon was critical to informing the scheduling features we later built into the product.

There are several different ways to observe, with varying degrees of intrusion. The techniques below have worked well for me in the past.

Follow-along

Follow-along with your customer in their daily routine. Ask to join for the commute to work or a trip to the shopping mall. See what it’s like to be them. What challenges do they face and how do they overcome these challenges? Be a friend and confidant as much as possible. It’s better to blend in, make mental notes, and take some pictures. Journal about your experience once you leave for the day. Capture as much detail as you can, focusing on emotions and motivations.

Walk the talk

Perhaps obvious, but worth saying — you should become the customer as much as possible: use your product and use competitors’ products. You can learn a lot about competitors from looking at their websites, but can learn even more from shopping at their stores, interacting with their sales reps, and going through the end-to-end process (including payments, reviews, etc.). At Thumbtack every employee is given a budget to hire local services pros for personal needs. Some of the best product feedback I got was from my own colleagues who had recently hired through Thumbtack.

Recorded session

There are several websites like UserTesting.com that allow you to observe customers engaging with apps, websites, and other online media through a recorded session — e.g., live audio and screen share. It’s helpful to record customers using competitors’ sites and products to understand existing routines. Getting a sense of what works well and what doesn’t work well can help you determine how to differentiate.

Test

“If we worked on the assumption that what is accepted as true really is true, then there would be little hope for advance.”

(The Wright Brothers)

Once you’ve gathered information from listening and observing, refine your design requirements and build a prototype to test in the wild. At the earlier stages of a company, it’s okay for the prototype to be low-fidelity — e.g., a sketch or wire-frame.

At Thumbtack I would draw up wire-frames in my sketchbook to bring to 1:1 interviews with our pros. After some refinements I would bring print-outs of actual designs from the design team. Having a physical object or representation for the customer to react to is critical for actionable feedback.

When giving a prototype to a customer to test, don’t over-explain. Ask the customer to tell you what they think it is and how they would use it. The best products are intuitive — your customer should be able to ‘get it’ without much prompting.

Role play

Have the customer walk you through how they would use the product (or actually let them use it). Prompt the customer to voice-over what they like and what they don’t like as they go through the experience. Ask the customer to elicit the emotions they are experiencing — are they excited? confused? Map their emotions in the format of a ‘customer journey’ (I usually draw out a line — like a timeline — and make notes of the highs/lows). Identify at what points customers are likely to get frustrated and subsequently design interventions to provoke delight.

I Like, I Wish, What If …

This questioning framework is helpful to solicit honest feedback and creative input from customers. As the title suggests, you ask customers to tell you what they like about the product, what features they wish the product had, and get them to co-create and ideate with you by offering up ‘what-ifs.’ It’s a technique I learned from Stanford’s d.school. You can find more about this technique here.

Focus group

It’s useful to get consumers together in a focus group to test the product, as it’s helpful to know when they agree with each other vs. when they disagree. Focus groups can elicit debate on features you perhaps did not think were very important. However, focus groups require a lot of preparation and can be expensive, so use them to test and get feedback on higher-fidelity prototypes.

—

Understanding the customer early can reduce the number of times you have to iterate later-on. As such, user research is a critical capability area to embed in your startup from the outset.

Remember that the customer is (not) always right. It takes careful listening, observation, and testing to create a product with true staying power.