We need more ethnographers in the (enterprise) software industry.
We would produce better software (by which I mean software that achieves its intended benefits more) if we started our projects off with an understanding of how people really work in their day to day lives.
Instead we start off with interviews and workshops in which we gather a view of what managers say their staff do. Or rather, what they say they think their staff do. Which is often several steps removed from what people really do.
The only way to really understand what people do is to spend time with those people. If you were to take this kind of approach in enterprise software development you would spend a year or even 18 months figuring out how people really work, and only then would you start designing new software. But your software would be better.
On one cynical level I wonder whether this happens anyway. The first version of an enterprise system is implemented based on what key individuals have said they think the organisation needs to do. It fails to deliver its benefits. Then it is reworked and reworked over the next 18 months.
If so it would definitely be better to do some ethnography first, before implementing your new system.