[tootie]

The number one reason agile projects fail in my experience has absolutely nothing to do with planning session, sprint cadence or estimation. It's because the client was not properly prepared to accept iterative delivery or play their part as product owner. I see a lot of team organize themselves around a well-groomed backlog and set their priorities only to have clients come in and ask for deadlines and fixed scopes and all the other stuff that is anathema to agile. If you client is able to set priorities effectively and allow a slower, quality-driven model then everything else becomes just so much easier.

[beat]

I just read a marvelous book called Handmade, by a fine furniture woodworker, and something he says over and over is "Go slow to go fast". The sales pitch to the business for a well-controlled agile process is that it maximizes productivity. Shifting priorities and poor planning undermine the productivity of the development team.

[gridspy]

Yes, this is also worded as "Slow is smooth, smooth is fast" - used for instance when following emergency checklists.

It's a good mantra when under serious pressure.

[vitro]

I remember one movie scene where old sniper teaches young one "To be slow is to be precise, to be precise is to be fast"

[e12e]

"slow is smooth, and smooth is fast"

Ed: see also: "practice makes permanent".

[wool_gather]

Ironically, such clients also seem to expect that whatever additions/changes they dream up should be able to be folded in to the plan willy-nilly. Whereas if they accepted an iterative process that would come naturally, without constant re-negotiation or ill will.

[tootie]

They can! That's exactly the point of agile and is how you get them onboard. When clients want a feature that is weird or complicated or whatever, the answer is never "no" the answer is "sure, now tell us where it fits in the priority list".

[wool_gather]

Totally, I get that. They don't seem to. ;)