> It's the modern variation of Ford's apocryphal "if I had asked people what they wanted, they would have said faster horses."

This x100 times. After building several B2B products I've found that the most successful features are the ones the customers weren't expecting and didn't ask for...

It's ok to do customer development in order to track what your users want and at least cover their expectations, but you'll never create amazing products with this mentality...

Relevant: "Obvious to you. Amazing to others."

https://sivers.org/obvious

"You'll never create amazing products with this mentality..."

I think that this kinda misses the point. If you're only building the things they explicitly ask for, then yes, you're right. However, customer development isn't mean to track what your users want and cover their expectations; it's to uncover the biggest pain points/real problems to come up with amazing solutions for - what's actually _worth_ pursuing and potentially building? How exactly those solutions manifest themselves often times does not come directly from customers, but the amazing ones are just as often inspired by insights gleaned from customer development.

The article you shared actually backs up my point exactly. Example: what could be obvious to one of your customers can end up being amazing to you and your other customers. In practice, I've found this dynamic play out time and time again.

The Atari Lynx wasn't massive because it needed to be; the creators listened to their customers:

"In all the focus group testing, and we did a lot of it with consumers, we had a bunch of different models that we showed them," Mical recalled in an interview with 1UP.com. "[We asked] "which one do you like? Which one would you like to have it be?" We showed them big ones; we showed them little ones. We showed them gigantic ones; we showed them little tiny ones. They loved the big ones. They all told us, 'Make it big. Make it big. This one feels like it's substantial and I'm really getting my money's worth.' They all told us to make it big, so we made it big. And when it came out on the market, they all said, 'Why is this damn thing so big?' It'd drive me nuts, because the original Lynx was mostly air space inside. We put it in, because that's what they told us they wanted."

http://www.usgamer.net/articles/too-good-for-its-day-ataris-...

Focus groups are absolutely the worst way of getting product insights. Unless they are extremely well run to counterbalance a huge number of cognitive biases and random influences within the group, don't do them. They can be worse than useless.

Yep, and in addition to not knowing what they want (at least sometimes), customers also don't generally have a detailed understanding of the costs associated with building new features. It's easy to say "I want X" if you're not the one who has to build it, integrate it into an existing platform, support it, etc.

Also, "customers" are not a monolithic group. Some will try to pull you in one direction, and others will want you to go the opposite way. You'll go crazy trying to please all of them. Just saying "listen to your customers; they'll tell you what to build" is a massive oversimplification.

At the end of the day, customer feedback can be very valuable, but only as a starting point. You have to internally decide which of the things, if any, you actually want to build.

But if you wrote a blog post that said: "listen to your customers and consider building what they ask for", it would sound obvious (because it is) and wouldn't drive any clicks.

> There can be counterexamples to that advice.

Welton's law of startup advice: for each trite bit of advice, there is an equal and opposite bit of trite advice.