How to run customer interviews that work in B2B
B2B marketing broken? you probably aren’t talking to your customers.
B2B is about building a case for an organization to buy your solution in a likely saturated market in 6 months. It's not easy and much harder to do if you don't even know how to run interviews to find what people value about your product or service.
Most marketers agree that talking to people who value the product (users, buyers, customers) can inform all your efforts. With user insights, you can write better copy that turns into clicks, design better creative more attuned to the category and problem, and find pain points, competitors, and benefits that can transform into keywords and ads.
So why is there so much friction when trying to run a few interviews?
Why do we want to hack ourselves out of talking to our customers?
It may be because most marketers don't know how to speak with customers to extract valuable information. We can talk, sure, but fail to listen and translate stories into impact.
Let's change that.
Running efficient interviews in 4 steps
Step 0: Ensure you have buy-in from C-Level / VP-level execs.
Complaining that research is an expensive and slow process that doesn't translate into actions is trending in B2B marketing. But more often than not, the inaction and the lack of use of research projects stem from the lack of commitment to implementing and updating products and processes based on the findings.
Having no buy-in from decision-makers and doing research rarely yields any positive results. So, before you even start, make a case for the need to have conversations with your customers to improve your understanding of the problems you are solving. Sometimes, during this due diligence process, you may find assets or previous investigations that can provide more clarity (you may not even need to run interviews). At other times, you will find there are still some gaps you can fill by talking to customers candidly about their experience with your product.
Whatever the case, getting top executives to join you in this exploratory journey to understand problems more accurately is also part of the job. Doing research is as much about the quality of the information as the delivery and the agency and persuasion you can embody.
Your research is only as valuable as it is actionable.
Note: If you don't have buy-in, try something else and skip the interviews. Save yourself the pain down the line.
Step 1: Understand the objectives.
Once you have decision-makers support, you must ensure that the interview process's objectives are clear for you and your organization. You may face multiple conversations along these lines:
"Can we include an NPS question to..."
"Can we also ask about blockers that..."
"Since we're doing the rebrand, could we test that..."
Given the perception of running interviews as a "slow" type of activity rather than an agile process, many departments and decision-makers will try to get you to include their priorities in the process. Politely decline. Your job is not to please everyone and interview someone to tell you their entire history with your product or category; your job is to understand a foundational problem (often a singular objective).
Identify what that primary objective is, and learn to focus. Once you do, include a couple of additional objectives you are confident can also be tackled and make sense to include in the conversation.
Note: You can suggest running other interview processes to solve different problems later. Training your organization around cherishing customer feedback via interviews is also your role as a researcher (a weird role) in a world of techies in love with numbers, not stories.
Step 2: Ask the “right” questions.
Many people in the technology field believe they can answer questions simply by asking directly. For example, if you wanted to know the problems that led them to find your product, you would ask something like this: What were the top problems you were facing that led you to consider (Insert product)?
Question - Answer.
Much like a machine. Input - Output.
While this is technically correct, you need to build rapport and trust with the person on the other end to unlock the highest-value information. That means you start the conversation at a category rather than a problem level.
For example, suppose you worked for a cybersecurity SaaS. In that case, you may not want to start asking about "what cybersecurity problems you were facing when you bought our product," but rather something more general.
Your initial questionnaire may look something like this:
What does the internet mean to you?
What impact does the internet have in your life?
Discuss work, personal, and social dimensions.
How did it get better? and worse?
How safe is your data online?
What is cybersecurity?
When did you first learn about cybersecurity?
How has it changed in the last couple of decades?
What problems were you facing when you bought our product?
Having a more genuine conversation with the person you are interviewing is one of the top secrets to unlocking real, actionable insight.
We are not machines. We need to warm up and feel like we are in a safe space before we open up. You may not need an hour-long conversation to get there, but honoring that we are connecting with another person makes a difference. It shows.
Note: Pay a lot of attention to the stories people share. They may be already answering your questions before you even ask them. While you can revisit and go a bit deeper, it can also be frustrating for them to repeat themselves.
Note 2: Leading questions are usually not great. Use them only as a last resort for particular questions if the conversation is going nowhere.
Pro-tip: During the interview, act as if you know nothing about the space. You should be interested in learning how the interviewee sees the category and the problem (not confirm what you read on Reddit or in imaginary Personas). It's not about you think, or what you think they should answer. It's about why they think and act the way they do.
Pro-tip 2: You can find multiple "standardized" questionnaires in Google to run perfect interviews for various purposes. Look at those, but continually adapt your work to the objectives and needs. Research playbooks are an excellent place to start, but copying them might not yield the best results. More on this in step 4.
Step 3: Share the questionnaire internally. Align stakeholders.
This step is self-explanatory yet often forgotten. Remember that your research is only as valuable as it is actionable, meaning those who can take action must vouch for you.
You can do this in two ways.
Share the questionnaire openly with them, including an initial section explaining the objectives, purpose, outcomes, and general reasoning behind the questions.
Set up a meeting with all stakeholders and walk them through everything we mentioned above but in a real-time call or meeting.
The main objective is to get them to sign off on your work. This will increase your chances of implementation once you finish and make it easier for them to support you in the implementation phase.
Note: Never share the questionnaire without supporting information/documentation. Remember, your job is to align them and get decision-makers to vouch for your project and its actionability (not to brainstorm new questions or updates).
Pro-tip: This is a moment to show true expertise and manage up. Own it. Stand your ground if you must, but also be open to updating sections if the quality of the suggestions exceeds the quality of your current questionnaire. You will also get extra points if you can tie the interview process to a KPI that matters to the organization, so do your best to do so.
Step 4: Think and ask outside the box. Search for the unpredictable.
Anyone with experience in social sciences would tell you that, while predictable, humans are much more complex than we imagine. Even if you have the "right" questions, they might lead nowhere. In other cases, they may only provide standard knowledge-based information, rendering your research obsolete unless you push yourself to think beyond your comfort zone.
Don't be afraid to be a bit more creative. You may find emerging patterns from the conversations or particular stories that help you gain insight into the problem. You may also find that some of the best research findings may happen outside the "right" questions asked.
As a general rule of thumb, do your best to be genuinely interested in learning more about the experiences of those you are interviewing. Take your time, read and consume the content they consume before you meet them, try to learn about their world, and more importantly, discover what they think and do (not confirm what you already know).
The job is to find behavioral patterns and stories, not just actions. You're searching for stories and meaning (not just product features or positioning fine-tuning).
Finally, remember that just because you're in B2B doesn't mean you're interviewing a business. You're interviewing a real person who must watch out for the business's best interests. Act accordingly. Keep it human.
Final Pro-Tip: Customer, User, and Buyer interviews differ slightly in a B2B Context. Ensure you are targeting the right people and building a questionnaire that matches their relationship with your product.
—--
Did you enjoy this piece? Sign up for our newsletter to learn how to have better conversations with your customers and discover ways to improve your demand generation strategies that your competition might miss.