Thread by @swardley

An agile organisation is one that uses appropriate methods according to the context whether project management (i.e. XP to Lean to Six Sigma) or purchasing or team structures.

An organisation which uses "Agile" everywhere will never be an agile organisation.

X : What's your advice?
Me : On agile organisations?
X : Yes.
Me : Depends upon how much I like you.
X : Explain.

Me : If I like you then I'd say ... understand your context, use appropriate methods, go serverless, go finops and think principles.
X : If you don't?
Me : Write stories, go agile, go containers, go devops and re-organise to a spotify model.
X : Bimodal?
Me : I'm not cruel.

X : Go on, be cruel.
Me : You should try my patented IT strategy development flow ... it has all the answers.
X : You patented that?
Me : You should buy some vBlocks. Have you considered a SAN?

X : What if you really dislike someone?
Me : If you buy enough SANs then you can build your very own data lake. I'll even throw in a few scoops of innovation. Would you like ketchup with that?

X : What's a vBlock?
Me : Lol. You've made my day. They were once all the rage for a bunch of people who wanted the future to be just like the past. Go on, ask me "What's a data centre" because that would make my year.

X : By that definition then six sigma is agile?
Me : An agile organisation will use appropriate methods according to the context. It will outsource that which needs to be outsourced. It will build in-house with XP that which is novel. Yes, six sigma is part of the picture.

X : That is brutal.
Me : No. This is brutal.

X : When do you come up with this?
Me : When did I learn that "agile" didn't work everywhere? About 2003-2004. I couldn't explain why until 2005 which is when I had created my maps. Then it clicked, I finally understood what I was doing wrong in my pursuit of a one size fits all.

X : Why don't you run a company doing this?
Me : Oh, I've done that company thing but I don't like "leading" ... it's totally exhausting. I keep on accidentally creating things but I try to get others to lead it. I only get involved if I have to, usually because of some crisis.

Leading is an incredible responsbility, it's a sacrifice you make of yourself to the great good of the collective (your family, club, company, whatever) ... the dedication you have to put into finding the path, a future for everyone, playing the game, building the culture ...

... you have to become the servant to the collective, the pillar on which you enable others to shine ... it's a tough thing to do, it ages you visibly such is the toll. Don't take it on lightly. You are responsible for the fates of others ...

... unless you're going to be one of those down the golf course, long lunches, long speeches, hire a consultant to come up wth a strategy based on some generic meme and then bark at your people over execution. That's a doddle but that's not leadership, that's being a parasite.

X : The hardest part?
Me : It's all difficult. From $0 to $5M in annual revenue (the getting it started) to the turnaround (from loss making to profit making) to finding a new future (overcoming existing inertia, herding cats into a new path etc). All create unique challenges.

X : What's your problem with DevOps?
Me : No problem with DevOps. It's a great set of practices but like it or not, it has become tied to containers whilst the battle has moved further up the stack. Tribes gotta tribe. Or ... from the wonderful @forrestbrazeal

The best bit of all of this, the bit that gets me howling ... AWS got a 7 year head start in IaaS, an almost unimaginable concept of competitor incompetence. In serverless they basically got a 7 year head start again. Twice? Ok, MSFT is in there and Google ... well.

I listen to companies tell me about their great plans for the future ... they're going to go cloud, adopt devops, go agile, go "spotify" ... it'll take them at least 7 years, so that's 2028 if they start now ... 2028 to become more 2012 and discover the world moved on long ago.

X : You don't have to re-architect with containers.
Me : Alas, the big fib. The real value comes NOT from the technology but the change of practices and you absolutely have to re-architect to take advantage of this. You can't have the future which is "just like the past" ...

... there is no lift and shift to the cloud which gives you all the benefits without re-architecting. There never was. That was just a myth, a holy grail created by some canny individuals back in 2008-2010 to keep flogging the past ...

... it used to be so funny listening to CIOs (around 2008 - 2010) say they would wait until "the cloud matured" and was ready for the enterprise. Many really believed in the whole lift and shift myth or that building their own clouds was more than a transitional step.

X : I do like your overlay of teams.
Me : Thanks. You can overlay a lot more i.e. attitude with pioneer - settler - town planner.
X : Is this new?
Me : Using maps to determine teams - no. That map however is relatively new, it's a tidied up version from 2014-ish.

X : The project was run that way?
Me : In this case no. I used the map to explain how they should organise and why their planned contract structure / outsourcing would lead to excessive cost overruns.
X : What happened?
Me : They ignored it and embraced excessive cost overruns.

X : How often does that happen?
Me : Which bit?
X : The ignoring?
Me : How often do I spend a few hours mapping out a contract, pointing out what's wrong before they start, they ignore it and then go blow hundreds of millions in totally avoidable cost overruns? Too often.

X : You map out contracts?
Me : Rarely now. I refuse especially if the org is run by story tellers ... ditto strategy. It's a waste of my time and too depressing. Life should be fun. There's only so many times you can tell execs that marching troops off a cliff isn't a good idea.

X : How long does it take to produce a map like this?
Me : That one was me plus one person from the project team and an afternoon. 600 page spec into one page which I used to examine the contract structure and demonstrate it would fail before they had even signed the contracts.

X : So what did the contracts look like?
Me : In mapping form, roughly like this - each blob being a contract lot. A predictable disaster before they had even signed the paperwork but then that's what story telling and zero situational awareness gets you.