It might be plausible. If the sessions were brief, the researcher could probably conduct 16 interviews per day. At that rate, it would only take 114 working days dedicated to that task per year to hit the 40,000 mark. That's roughly equivalent to a part-time job dedicated to interviewing subjects, which doesn't seem terribly unreasonable, especially when the author's states that a significant portion of his work is dedicated to just that: "In my job I listen to (and look at) a lot of people."
