You're right, I don't know how it works because I've never seen it work for the type of information he needs to do his job well. It's a red herring. Waste of time except to perhaps help him identify with whom to spend time.
For example, to whom will the survey be sent? You can't know how to target without getting the on-the-ground details. He needs to see how the people are dressed, how they talk, their reaction to what the sales folks are saying. It's okay to be biased if every salesperson is biased...you need to get into the head of the potential user/customer...just can't be done with check boxes, write a note, answer my question. Half the questions aren't known for the first three engagements.
Later, when you know your target market, when you know the key questions that qualify, then you can survey to figure out whether a certain untapped segment might be worth focusing upon.
Can surveys tell management why their 10x estimates is achievable or laughable? For that you need stories, context.
In the time it takes to create a "good" survey (that even then doesn't really have the right questions), he could have had face-to-face meetings with 10 people who bought and 10 people who considered buying it but didn't and be a lot closer to figuring out the business opportunity and what it takes to develop it.
Then he can scale his data collection with a survey.
I once was part of a big presentation to the CEO of AMD about a new product and had to reconcile our numbers we built from the ground up with Forester Research, the lords of storage equipment research, and their much larger numbers. I was told that our number would be harder to swallow considering Forester's reputation if we couldn't close the gap.
We had gone over our numbers very carefully, even originally doubling them as an error factor on what we might have missed.
Finally the night before the presentation, I called the research firm and reviewed our methodology, etc. They said it sounded good, we probably hadn't missed anything, and then I was asked, "What number did we give you?" I told him.
"Oh, we just revised our number two days ago down to XX" I was greatly relieved because although the numbers did not match, it was now close enough that it was not material.
"What changed? Why did you shrink your number?" Answer: "Manufacturing shortfall" "Oh, what's Manufacturing shortfall?" "Well, the way we built this number was to call all of the suppliers and ask them about their plans. They recently reduced their plans." !!!
The lesson I learned was 'supply-side' market research is not very useful for anything but commodity products.
Now, I realize that's just one subset of the market research universe, but these guys were the standard bearers. That combined with my own first-hand experience taking surveys and questions that don't even scratch the surface to understand my purchase decision making, tell me working with market research and surveys is just not a good return-on-effort until you are measuring very specific things or shifts of feelings.
So, I would never, ever, bet my job or company on market research. Whereas I've looked VCs and BoDs in the eye and shifted my company's focus based on face-to-face engagement with prospective customers.
Yes, I do feel strongly about it.