The Art of the Customer Development Conversation

By now all Lean Startup and Customer Development practitioners should know that if you're not getting out of the building, you're not doing Customer Development. Each of the following, while often a necessary and beneficial activity, does not constitute Customer Development:

  • surveys
  • automated customer feedback mechanisms (e.g., uservoice, get satisfaction, et. al.)
  • embedded product use analytics
  • marketing analytics
  • feature request mechanisms
  • sales calls
  • product demo
  • usability testing
  • focus groups

Each has its place in the Customer Development process, but without live Customer Development Conversations, you are likely compromising your ability to learn your way to Product-Market fit or startup scaling.  What you seek to learn evolves over time, as do the tactics you employ, but every step of the way should be grounded in real time conversations.

Generally speaking:

Pre-Problem-Solution Fit, you concentrate on learning as much as you can about the problem, who are the real customers (user? buyer? boss?), and possible solutions.

Pre-Minimum Viable Product, you concentrate of learning, developing and testing the minimum features and functionality required o solve the problem to a degree the customer will buy.

Pre-Product-Market Fit, you concentrate on learning about funnels, testing messaging and positioning, and likely iterating on product and market segment in search of P-M fit.

The conversation itself is difficult for many. The key to effective conversations is in developing the conversation around your objectives for the discussion.  Establish 3 or 4 "must learns" for each conversation.  Depending on what you're trying to learn, you can use "open-ended" questions and direct questions.  Use your conversations to calibrate your other tactics.  Are you asking the right questions in your surveys?  Do responses gibe?  Are product demos necessary from the buyer's perspective?   Are users using the product the way they say they want to?

I thought I'd share some recent experiences my clients have had with customer development conversations to help illustrate conversation tactics.

1. Talk problem first, not solution.

Client: "I can't stop talking about the solution.  It's how the conversation starts."

Clearly, whenever you meet someone for the first time, you need to introduce the space your product is in.  You must provide a context for the conversation.   You need to be able to state what you're about succinctly, so you can move on to the purpose of the meeting.  Otherwise, you spend the whole time trying to explain what you're doing.  This is why the elevator pitch is so important.   In a conversation, you don't necessarily repeat the pitch verbatim, but you use components of it to steer the conversation.  So, for example, if you are testing your problem assumptions, use a classic sales cold call method of "flipping" the conversation from talking about yourself (solution) to the customer (problem):

Hi, I'm Brant Cooper from Market By Numbers, and I teach Founders Customer Development principles to help get their startups off the ground; (intro that provides context)

I often hear from startup CEOs like yourself, (The Flip)

That they have trouble figuring out how to prove their business model and how to prioritize what they should be working on; (problem hypotheses)

Do you experience this at all? (Open ended question)

Follow up questions depend on the answer.  If answer is no, you might say: "That's great, it sounds like you have things under control. "What are the primary issues you face as CEO?" or "Do you hear other CEOs that talk about these issues?"

2. Except when that doesn't work.

Client: "I couldn't get her to describe her pain.  She kept wanting to know how she could help me!"

The contact was a senior person in the exact type of company needed to confirm a key hypothesis.  My solution would help her company's relationship with its clients, but I couldn't get her to talk about that.  Instead, she talked about problems her clients had and the opportunities available to solve them.  I asked if she was a sales person and it turns out she was.    I said it sounds like she's a good sales person, who understands the needs of her clients, and a good contact, but not his customer. I suggested my client ask her to refer him to her company's product managers who would be more likely to be interested in his ideas.  "More contacts to speak with" should always be a core objective.

If a problem statement doesn't resonate, either your problem statement is not articulated correctly, is flat out wrong, or you are not talking to the your customer.

3. This is not feature mongering.

Client: "Conversation inevitably turns to features discussion."

If you are talking about features, see 1. above.  If your contact or customer is talking features, slow it down by by asking why.  Use the same 5-whys approach Eric Ries talks about to discover a problem in your development process to figure out what is driving the feature request.  Continue to ask why the customer wants specific functionality until they are able to tell you the actual problem they're trying to solve.  This helps you consider a couple of things.  First, is the feature a high priority to the customer?  Second, are you building the right Minimum Viable Product?  Obviously, one data point is not enough to draw conclusions, but it contributes to the analysis.

4. How to test messaging.

Client: "Can't I just A/B test?"

It's fine to use split testing to test messaging, but it's most effective when you're optimizing your funnel. If you do it too early, you actually don't know if you are optimizing for the right product and market.  Also, you can find out that one message works better than another, but you can't learn why.  In a live conversation, not only can you ask why a message doesn't work, but you can test several iterations in the same conversation.   When testing messages live, I like to use the "pause method."

5. Ask them.

Client: "Everyone says I need Facebook, Twitter, a blog.  How do I know?"

If there are two things the world doesn't need more of, they are Internet Marketers and Social Media Gurus. It's easy to get caught up in the excitement and clearly Internet and social media marketing can be effective.  But their effectiveness depends less on which expert you hire than whether your customer and their buying process intersect with your online tactics.  I spoke with a CEO recently who had two B2B deals in pilot for his new startup.  He had personally orchestrated the sales process through 1-1 relationship building in a carefully honed market segment. Yet his marketing plan called for Internet and social media marketing that didn't jive with his experience.  He immediately described a shotgun approach to figure out what the right online channels would be.  While it might be the case that online marketing would be successful and even necessary to scale, at that time all evidence pointed to offline methods.  Using the "Flipping the Funnel" idea, I recommended he Ask his existing customers!  Here's a question you can ask straight out to your contacts:  Do you belong to social networks?  Do influence your buying decisions? What blogs do you read, etc.

I hope you find this helpful.  What issues do you encounter during your Customer Development Conversations?

10 COMMENTS ON THIS POST To “The Art of the Customer Development Conversation”

  1. Tristan Kromer September 2, 2010 at 11:37 am

    Excellent post,

    I’d be really interested to hear your thoughts on how to integrate email exchanges, forums (such as uservoice) *into* customer development talks in a meaningful way.

  2. Edwin Oh September 3, 2010 at 11:24 am


    Great PRACTICAL post (just like your book)! For some of my clients, the hardest part is just figuring out how to get the conversation started in the right direction which is why I’m a huge fan of getting them to develop and practice an elevator pitch.

  3. Rob Fitzpatrick September 4, 2010 at 7:12 am

    Really excellent post, thanks. One problem point that I fall for depressingly often is starting to sell too soon, which tends to end the discovery conversation.

    Another that I haven’t seen much discussion about is how to navigate complicated customers where you’ll ultimately need approval from 3-5 parties while dealing with motivated saboteurs. It makes for a tough discovery process with a lot of false positives and hidden dangers. Would really appreciate hearing your thoughts on it if you’re in the mood to keep on typing 😉

  4. Colin H September 6, 2010 at 12:31 pm

    Great post, asking customers and prospects questions seems simple enough in theory, but for some reason seems so intimidating for startups.
    We’ve been working through our customer discovery processes and found that asking a couple key questions often gets the conversation started and focused on their pain points. For example for our target market we ask something like “How much time and money have you invested or expect to invest in starting/ managing your online payments?” or “If you got paid daily rather than monthly for your sales, how would you use that additional available cash-flow to grow your business?”

    • brantcooper September 13, 2010 at 11:27 am

      Great questions to frame the conversation! Thanks, Colin.

  5. Cindy September 26, 2010 at 8:42 pm

    For the person who leaves the comment/UserVoice vote/email, you can ask them for clarifiction:

    “Thank for requesting feature X. Just to be sure I understand what you’re looking for, could you tell me what you expect feature X would allow you to do? Could I give you a quick call at one of these times…?”

    For other people, you can use the great “other customers have said” tactic. This is a great way to get honest feedback — customers are often a bit hesitant to disagree directly with you the interviewer, but don’t have that same hesitation about some unnamed third party. “Some of our other customers told us they had problem X – have you experienced problem X as well?” and so on.

Leave a Reply

Your email address will not be published. Required fields are marked *

Your Name *
Your Email *