I can’t say how much I’ve waited for this, I even wanted to write a post like that myself (but I’m too lazy/busy to blog).

So thanks, Tarmo, for writing that.

You pretty much summarized 80% of what agile dev, crystal clear and the rest of the current in-vogue methods (that work!) are all about.

Some things I might have added/weighted differently:

“focus group”
No. Really. No.

“Focus group” has been completely destroyed by marketing-droids who never took a single course in qualitative research methods. More than that, it is so often misunderstood by sofware developers to mean that “listen to your user” means “user is a designer”. They are not. Often they do not know what they need, can’t put it to words so that you understand it or just have downright bad ideas about how to implement it. This is not to say, don’t ever follow your users’ ideas: if the idea is good, proven, implementable, fits into schedule, etc. then by all means do it. If not, try to capture the essence of what is important.

Of course, I don’t want to say that don’t have structured meetings with users and clients. But just don’t use the f-word and especially the pseudo-methodology in has been watered down into.

Also, I want to stress the importance of informal communications esp. with end users, who almost always clog up and say nothing of real importance in structured meetings, especially if any of supervisors are present.

Which brings me to my next point (from agile dev):

Communicate with your clients and users
This is so much more important that having a contract with them. And by client I mean both the end users and the persons paying for the software. Understand them, influence them, ask them. Keep each other posted and solve problems honestly together. This pays off as better quality software, higher user satisfaction and usually more on schedule projects.

A contract specifying what to do, is not communication for obvious reasons, as you already pointed out in 13.

User-centric is too thin a slice for me, personally. Consider the wider human implications for an overall improved experience (i.e. human centered design). For best results in efficiency/errors/learning/outcome, consider activity centered design.

Good stuff, although somewhat theory driven at times.