This blog is written by Marcus Catt, a principle consultant for Infuse with over 40 years worldwide experience working within the I.T sector.
I could have written this article pretty much anywhere I have been at...
What businesses should do to really make it easy to fail at AGILE:
- Hire people to lead your tech and product initiatives who would really be better off selling PPE to the government
- Dont express your needs well so Tech go and build the wrong thing for you
- Always expect a fully functional enterprise grade solution to be the first thing you see from Tech. Never be happy to explore an experimental model.
- Keep changing you mind.... like every week is a really good rate of change...Agile welcomes change after all.
- Prefer illusion and fiction over fact and realism.... sometimes NO actually was the best and most positive answer.
- Keep everything a secret so it’s hard to help you.... Tech love it when you don’t tell them you have promised the client a 3ms response time.
- Don’t bother to understand your technology at all. In this world where everything is driven by Tech you don’t need to know do you?
- Don’t prioritise work and/or cost reduction exercises based on real value or benefit - use Penis size, that’s a much better measure of value.
What Tech can do to help AGILE fail for the business:
- Hire people with only one skill (A Business Analyst; A Developer; A Quality Analyst specialist etc). Agile requires a multi skilled workforce to succeed. I am sorry, but, it does! Those who say it doesnt have an investment in that thought train that clouds their mind.
- Build a classic waterfall approach using AGILE language.... If delivering to production is not part of your DoD for the work and you don’t deploy every iteration to production.... you are waterfall and kidding yourselves.
- Adopt tools/techniques/ways of working based on LinkedIn and/or fashion. Yeah, Spotify use this, Google do it like this....so we should too, obviously
- Focus on the messaging .... Let’s not let fact get in the way of a good story. The business loves the positive message, until they wake up, smell the coffee, and get angry.
- Treat quality as optional.... It doesn’t matter where you set the bar. It does matter that you set it somewhere and try with all your might to get over it. and once you have done that, move it higher.
- Hide behind the process model. We did it wrong because the process was wrong.... Yeah right!
- Create internal conflict and disquiet by refusing to listen when staff say no! If the team say no, they will be right. They will start to go into a silent miserable holding pattern and when it suites them, they will leave. good luck coming back from that one.
- Confuse empowerment and delegation with abdication and blame passing.... If you are truly empowering people, then you have to be patient and allow that it will go wrong at times. And realise that that’s part of the fun.
- Promote based on optics over substance.... Trust the guy who says no, explains why and then gives you a different solution model even though it is not what you want to hear.
- Cheat the game.... if you cut corners (leave out none functional and good infrastructure work, pay lip service to your GDPR and Security work) it will bite you, every-time!
- Last on my list. Develop cr*p software. Why is it last? because it’s probably the only one you can fix easy.
Consider this:
- The dominant advantage AGILE offers over other delivery paradigm's is that it is FACT based and provides rapid feedback on your planning accuracy. This enables better outcomes over the medium term and reduces the amount of risk being carried. If you compromise on that by adopting poor practice based on local cultural norms, then perhaps AGILE is not for you
- I have two lists above, one for Business and one for Tech. If it was working well there will not be two lists because there would only be one team and only one list.
- If you look at the lists above and more than a couple of points resonate with your current lot in life, then you know you are mucking up AGILE. Stop it please. Do something else.