I've learned so much from this summer job.
I've learned that house owners in our neighborhood tend to own dogs.
I've learned that many times dogs are much smarter than their owners... (the owners would many times leave the dogs running loose, despite having asked us to deliver)
I've learned to run fast. Very fast. Or at least faster than the dogs. Or at least faster than the chained dogs :-)
And I've learned that knowing your users is great, and lets you have a better product or service.
But I had 60 users. That's it.
When you're building a product with scale, with thousands - millions - of users, can you really know your users?
Working at a Wix for more than 4 years, with over a 100,000,000 (thats one hundred million) users - well, you can't know them all.
No matter what.
And you can't survey them (well you can - but we don't trust surveys - and that's a subject for a whole different blog post).
So what can you do?
Talk to your users. As many as you can.
True - you will never interview/talk with as many as you could survey - but at least you'll get two things that are much more accurate:
You'll feel the real passion of the user, for better or worse... They won't be answering theoretical questions about what feature they would possibly use (as often happens in surveys) but instead complain about what doesn't work or is missing - or what they love and why - and you'll hear their tone and voice.
You'll learn aspects of their business you would never understand from a survey. It's not about "Would you use a shipping feature in a store - and what provider do you want" but instead - "Holiday season was great - and horrible - at the same time. I sold so many products in my online store - which was great. But then I had to carry all the packages to the Post Office, and stand in line for hours. If only you had a shipping feature in your store..."
So the short answer is:
You can - you must - know your users.
You most probably can only know a small portion of your users.
But be sure to know them. Listen to them. Talk to them.
Then, go and do a winning product...