[ScottBurson]

While I think this is a great piece — especially the part about listening — and it's something we engineers all need to take to heart, I think its applicability varies depending on the nature of the product and the market.

If you're trying to serve a previously completely unmet need, as many startups are, then this advice is spot on and you should print it out and pin it to your wall. But there are different cases. Sometimes you're trying to bring an incremental advance to a well-established market. In that case there may well already be established marketing and distribution channels, and the value of your product may be readily evident to users of the existing ones. This is especially true in cases where the primary barriers to improvement are technological. To take an extreme example, people working on fusion power don't need to validate the market at all; when they finally build a working plant, they'll just have to plug it into the grid and get paid. Some of the things we build have this character, at least to some extent.

Another exception has to be made for truly visionary products. Sometimes, as Steve Jobs famously pointed out, people don't know they want something until they see it. The Macintosh might be the best example; computer users in 1983 were not generally thinking that they needed a GUI-based OS. This is a dangerous one, though, because we all want to think we're visionaries. If you're going to go down this path, do it with your eyes open, and realize your vision may be wrong — or, more frustratingly, right but too early (e.g. [0]).

[0] https://en.wikipedia.org/wiki/VPL_Research