Imagine its 1977.
The new Apple ][ computers have come out. Nerds are buying them as fast as they can be made. Other geeks are starting businesses around them, trying to build useful programs.
They go to your local accountant and generic MBA man and ask him the golden question:
"What would you buy for a computer?"
Shocked, the business man doesn't know what to say. So he mumbles an answer. It's utopic, its non-sensical. It pulls from the only thing he can think of with computers:
"I'd want a computer that I could talk to, does my taxes, and cleans my dishes."
The confused nerd thinks in despair: "computers can't do that yet!"
Yet the would-be customer doesn't realize is that Visicalc is just two years away. And when it came out, computers would explode with the first ever "killer app."
The lesson is that people really don't know what they want until you put it in front of them. They can't imagine how certain products would be useful until it exists.
You need to remember when talking to potential customers is that they won't know what they really want. Questions like "what would be useful to you?" is not helpful, not in most cases anyway (there's always the exception that proves the point).
But they do understand their own problems.
You as the budding product builder know what's possible on a computer and have the creativity to make something practical. Listen to your domain expert, your potential customer. See what they say about the problems they face.