FAQ: Customer Development for Product Managers

What is customer development?

Customer development is the opposite of product development, or “if we build it, they will come” thinking.   We all know that’s not the case – often you build it, and no one cares.

The idea is to validate that you have a market, with problems, that they’re willing to pay to solve — and then build something that solves those problems.   The emphasis is on learning and discovery before you write a 50-page spec or spend months writing code. Read Eric Ries’ post  What Is Customer Development for a lot more detail.

“This doesn’t sound like anything new,” you may be thinking, and you’re pretty much right.   There are plenty of examples of successful companies who used interviews and cheap prototypes to validate ideas before committing to building them.

What has changed is the availability of free/cheap tools and direct access to customers through social media, making it more practical for an individual product manager to do this on their own, on a skunkworks basis even.

But asking the customer what they want never works –

I know.  Customers tend to ask for “me-too” features, or they ask for something and then after you build it, say “I know we asked for this, but it isn’t what we really wanted.”

Customer development isn’t asking customers what they want – it’s seeking to understand what they need, how they work, where their pain points and highest priorities are.  Customers may not be able to articulate what they want, but they can’t hide what they need.

What will I learn?

The short answer: how people are really getting a task done, who is doing what, and why it sucks.

The more detailed answer: What you should be learning from customer development interviews

How is this different from usability testing?

Usability testing teaches you whether or not people are capable of using your product; it doesn’t tell you how likely they are to actually buy it.

In usability testing, you learn about the product you’ve already built.   If your product hypothesis was wrong, it’s kind of too late to do much about it.  In customer development, you learn about your customer before you build the product.  That way, if your product hypothesis is proven wrong, you can change course quickly without having wasted resources and time.

How long will it take?

I don’t know.

Sorry, but that’s the truth.  I can’t guarantee that this methodology will get you useful answers in two days or two weeks.  I can guarantee that you’ll start learning useful new tidbits from your first customer interview, but it may not be something actionable.

At KISSmetrics, I quickly learned that the reason so many companies loved our Survey.io customer development survey is that “I don’t have to write anything – you tell what I should be asking and how to ask it”.

People don’t like to write – in fact, they’ll put off a task if it requires them to think and write something.  That was a good insight, but not enough (on its own) to drive any new product decisions.  We needed more of those insights before they added up to something we could base a product on.

For this reason, customer development needs to be a process — not a project.  Always be listening, always be asking why, and always be testing hypotheses, and it will pay off.  You just never know when.

How do I get started?

If you already have existing customers, reach out to one today and schedule some time to talk with them about how they get their jobs done.  Ask about how they use your products, but also take a step back and ask about the general task that your product is supposed to help them with (i.e. if your product is a bug tracker, ask them about QA testing in general).

You don’t have a hypothesis to test (yet), so this is just an open-ended conversation.  Let the customer talk.   The Evolution of the Customer Development Interview has good tips.

You may be at a company who has existing products and customers, but you don’t have access to them.  As insane as this is, it’s not uncommon to try and lock the product managers up in a little room to wrangle up requirements somehow.   The fastest way to get access is probably to bargain directly with sales.

Ask to join them on a sales visit or conference call.  Offer to help with demos.  If you have to promise to be a “fly on the wall” in order to make them feel comfortable, do it.    In my experience, customers love having a product manager in the meeting – they feel like you’re an ally who isn’t just trying to sell them — and as a result, they’re often MORE likely to buy.  (Tell your salesfolk that.)

How can I find customers before I’ve even built a product?

How were you planning on finding them after you’ve built a product?

Seriously, this is the million-dollar question, and the answer is not “we’ll build it and they’ll come” or even “we’ll advertise a LOT” — unless you’re Apple, Microsoft, Google, or Facebook, you don’t have the dollars or distribution to make those work.

So you have to get creative.   I’ve previously written about how to find prospective customers using methods that take less than $20 and a couple days.

How do I convince my boss that this is a good use of my time?

Unless your boss is already a devotee of Toyota Lean Manufacturing processes or Four Steps to the Epiphany, start out skunkworks style. Remember, bosses like solutions, not problems.  Don’t say “we need to be doing X”, figure out how it can be done and prove the benefits.

Carve out some time, do some customer interviews.  Work with a designer to throw together a fake splash marketing page for a new product and get feedback.  Once you’ve learned something useful, use that to get buy-in for making this a permanent part of your process.

How does this change when I have an established brand and customers?

You may need to create somewhat more polished splash pages/demos in order to not hurt your credibility.

You may also feel more comfortable testing concepts without connection to your company name.  Using another domain name is a good option and worth the time to set up if you’re going to be running experiments frequently.  But you can go even simpler and use tools like shared Google docs (create a new Google account that isn’t linked to your name) or Skitch to post screenshots privately.

Where can I learn how to do this?


Blog posts:

People to follow on Twitter:

23 thoughts on “FAQ: Customer Development for Product Managers

  1. @abecrystal: The main differences are the “when” and “who”, which are interrelated.

    User research has typically not been a part of companies' arsenals until they're already pretty big (big enough to hire UX/IxD types). So it often ends up somewhat marginalized, instead of being a primary driver of business and product decisions.

    Also the “who” – I used to be an interaction designer, and I used these exact same techniques. But the trend is towards these techniques moving more into the product manager or entrepreneurs' hands.

    This is good for 2 reasons – by moving “upstream”, this research IS more likely to drive business/product decisions, and, by “hearing it straight from the horse's mouth”, it becomes harder for the product manager to ignore unpopular findings.

    (I have seen, on many occasions, user researchers come back with findings that contradicted the product manager or executive sponsor's vision – and the response was to conveniently ignore the findings.)

    Oh: and the other difference – a lot of user researcher folks have a more academic background, so they tend to be less comfortable with the small sample sizes, scrappy experimentation, and user deception (i.e. advertising a product that doesn't exist).

  2. Thanks for clarifying! All makes sense to me — the only area where I might differ is that UX is definitely trending toward more “scrappy,” agile approaches, see e.g. http://www.ugleah.com/ux-team-of-one/ .

    Enjoy your blog, glad you are bringing the user-centered perspective to a larger audience!

  3. Concept testing – at least in my experience – is more of saying “here's my solution for widget making; do you like it?” vs. seeking to understand how the customer makes widgets, what their priorities and problems are.

    You're learning about your product, not about your customer. Now, concept testing is used in later phases of customer development, definitely.

    And you can do the reverse as well – use concept testing as a jumping-off point to seek to understand the customer's needs. But unfortunately, many people show subconscious bias towards their solution, which blocks them from the “a-ha!” moments that lead to a really awesome feature/product breakthrough.

  4. In any organization, project management software is used for different purposes like communication, collaboration and overall management and tracking of projects. If you need to schedule multiple tasks and events at work then project management software can be very valuable.
    cursus timemanagement

  5. When I am working on a project, blog post or anything else, I usually just ignore email, RSS feed and TweetDeck. The one thing I have done is shut off all notifications from all of them. I don't know if I have a new email until check. Nor does TweetDeck ding me when I get a new tweet.cursus timemanagement

  6. Does the customer development model apply to non-tech indutries? How would you apply the customer dev model to a residential painting company, and would it give you an edge over competitors in that market?

  7. DanielG, I’d say it matters MORE but it’s much harder to get the word out.

    Take your residential painting company example – you’d talk to people who had recently had their houses painted and ask about the good and bad and if they would recommend their painter.

    You’re going to hear the most vehement responses about bad ‘personal’ things like the painters left equipment where my toddler could get into it or they spilled turpentine in my flower bed and killed all the tulips, or good ‘personal’ things like they were willing to take their breaks during my baby’s naptime. Those are the reasons why people recommend or don’t recommend service providers — once you know those triggers, you make that part of your advertising.

  8. Pingback: Quora ~
  9. what are the questions i need to be asking in an interview where i’m trying to test the problem hypothesis ?

  10. Hiten (@hnshah) directed me to this post. The only thing I want to add is that this article works really well (and the “Customer Development Interviews How-to: Finding People” article) with start-ups in the idea phase, where I’m at. I was going to create entire content marketing strategies to reel people into taking my on-site surveys for lead generation. Without calling me insane, he asked me how many customers I’ve talked to. I replied with 4 names of friends and family members.

    That’s when he sent me here.

    This should be the #1 priority before design, development, marketing, writing, etc. Getting data is easier today than it has ever been, so no excuses…

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>