Step one in building a new company is the YC adage: “Build something people want.”

But how do you go about figuring out what people want? The obvious answer we all give (and receive) in the startup community is to talk to people. “Customer development” has become its own skill and body of knowledge, and there are some crucial nuances to understand up front before beginning your customer dev discussions. In other words: Beware of how you go about doing this.

Although it’s unclear whether Henry Ford actually said this, I always keep in mind that saying attributed to him that:

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

It’s become ubiquitous in our community, but that doesn’t make it any less critical. With that in the back of our minds, let’s dive into a few of those crucial first-step nuances as you look to gain traction and hold successful customer development conversations.

The early problems you’ll encounter

The problem with most cust dev is twofold. The first is that people tend to think in terms of existing constraints. So when asked what one wants, one is more likely to think of something that is a derivative of what already exists, when something completely different may actually be preferred. The second is that it’s very easy to ask for feedback in the wrong way. As someone in the venture business, I see a lot of concepts for new products, and it’s super easy to casually ask folks, “So, what do you think about a service that does X?” But although asking this may give me some helpful information, almost by accident, it’s actually a bad question that almost always yields pretty shallow responses. It’s too easy to lead the witness in the way you ask a question, or get a reaction that tells you very little about a customer’s core needs or problems.

So: You need to break people out of their current reality/constraints. And you need to avoid leading the witness.

That begs the question: What is a better way to go about this?

I’ve been thinking about this a fair bit recently and have been reading up a bunch on the topic. I was also able to help out with a course taught by Tom Eisenmann at HBS recently during some early modules focused on this. Below are a few broad thoughts, as well as some recommended further reading at the end.

Surveys suck. Talk and listen to customers instead

The gut reaction for many people is to approach customer research through some sort of survey methodology. This is typically a really bad idea. Surveys kind of lull you into a false sense of security. They are pretty easy to create, give you what seems like quantitative data, and are very familiar to most people. But for a true seed-stage company or for folks just thinking about companies to start or products to build, I think surveys are generally terrible.

What should founders do instead? Talk to people… and listen. Talk to potential customers, people around the same ecosystem you are targeting, users of similar products/brands, people you think will hate your product, people you think will love your product, etc. Talk to lots of different potential users and try to do so in a deep way. There are good guidelines around this too, but I actually think that even if you do it poorly, spending a lot of time talking to people in an ecosystem to understand their world inside out ends up being pretty effective.

So what good are surveys? I think surveys are good if they are very tuned towards answering a specific question. I think they are also good for keeping a pulse on things over time (especially if your approach is pretty consistent). That’s why NPS surveys are successful, but most useful when employed over time. Same goes for brand awareness surveys. However, as an isolated data point, I think they are much less helpful.

(For how to actually execute valuable surveys, read this from SaaS startup Price Intelligently.)

Know what you are trying to figure out, and plan accordingly (generative vs. evaluative feedback)

As I said, just investing time to talk in depth with customers and potential customers can get you a long way. But it’s helpful to have some sort of a plan around this kind of research. One framework I found pretty helpful recently is from Tristan Kromer.

One of the key ideas here is that one should use different methodologies for different needs. And one distinction in particular is whether you are seeking generative feedback or evaluative feedback.

I find that often, founders gravitate quickly towards trying to get evaluative feedback. Usually, this is because they are pretty convinced that their view of the world is consistent with the view of users. They are very anxious to start testing hypotheses around specific products concepts, features, or solutions. But often, this is way too premature, and one would be better served by seeking generative feedback to understand customers and their needs better.

For example, I often find that founders do a very poor job of using competitors when seeking feedback. If one goes in with an evaluative mindset, discussions around competitors will quickly focus on a) whether any competitors get the job done or b) the features that are missing from competitor offerings. From there, confirmation bias sets in, as a founder may be dying to conclude that they don’t have direct competitors, or that their competitors are fundamentally flawed for a specific reasons.

But with a generative approach, talking about competitors can yield much better insight. First, you might find your competitive set actually expand as you focus on how customers get the job done today through a combination of tools or processes. Second, you can use your competitors’ products to probe deeper into core needs or customer reactions without having to build your own product. You may find that there are things your competitors do that actually resonate much more than expected, or that customers have an underlying problem that isn’t initially obvious, or realize that you are focusing on the wrong type of customer altogether.

Make listening to customers a superpower

I do a lot of reference calls on founders and product folks. I hear a wide variety of strengths and weaknesses, but one thing I almost never hear is something like “person X is really good at understanding customers and figuring out what to do with that”. But when I think about it, this is one of the super-powers I’d find most valuable in a founder or early team member. In particular, you want to be someone who is really good at eliminating bias, driving down into core needs and problems, and connecting these insights to actual product design and prioritization decisions.

A lot of great startup advice is often shared on the internet, but I find that this area is under-represented relative to how important it is. It’s particularly important for our target audience of seed stage founders and product designers, so to keep the dialog going, below are a few links that I hope prove to be a helpful start:

https://sites.google.com/site/hbspm101/home/2015-16-sessions/the-mrd-customer-discovery (great collection of links from Tom Eisenmann’s PM 101 course at HBS)

https://blog.kissmetrics.com/26-customer-development-resources/

https://www.quora.com/What-are-your-favorite-methods-for-doing-problem-interviews-during-Customer-Discovery

https://medium.com/user-research/never-ask-what-they-want-3-better-questions-to-ask-in-user-interviews-aeddd2a2101e

Welcome

http://www.usersknow.com/blog/

These should be a decent fist step, but please share other resources you value in the comments section below!