An Experiment in Scotch

It’s a dirty lit­tle secret but but there really is a big glar­ing hole in the whole agile soft­ware move­ment in large Amer­i­can busi­nesses and this guy nails it. The busi­nesses built around the soft­ware that most agile teams write aren’t actu­ally agile at all and that’s impor­tant. The agile teams I’ve worked on have been small and iso­lated to the team and pos­si­bly a cou­ple of sup­port­ing man­agers. The enter­prise as a whole isn’t agile and because it’s not, agile typ­i­cally fails to deliver what it promises. As that arti­cle says, you have to “embrace con­tin­u­ous inte­gra­tion of the enter­prise” for agile to suc­ceed and Amer­i­can busi­ness isn’t ready for that. A 2 month QA cycle at the end of the project isn’t agile. One month of prepar­ing for deploy­ment isn’t agile. At its heart, agile is suc­cess­ful when the entire oper­a­tion is agile. Toy­ota didn’t rev­o­lu­tion­ize car manufacturing through Lean by apply­ing it to hybrid engine design. It was the total company.

Don’t get me wrong, I’m still a big fan of agile. But in my mea­ger expe­ri­ence with it, if the enter­prise isn’t agile, the team isn’t really agile either. They’re just pre­tend­ing. There’s plenty of ben­e­fits to that pre­tend­ing but the real gains from agile aren’t being rec­og­nized. In the end, the pre­tend­ing saboto­ges the entire process. If you can’t ship at the end of the sprint, you aren’t really “done”. If you can’t hand the prod­uct to a cus­tomer and say “Happy work­ing”, you aren’t really “done”. Con­tin­u­ous inte­gra­tion of the enter­prise really is impor­tant and it’s so dif­fi­cult to do that iner­tia takes over. With that kind of weight sit­ting on top of a team, the agile process is just a facade. It pro­vides some pro­tec­tion to the team but it doesn’t pro­vide the ben­e­fits it could. Maybe agile is just still way ahead of its time.