This post is based on the IpX ConX19 Keynote that I gave together with Martin Haket, remember when we could still do in-person conferences? And it is the prequel to the posts We need a cross-platform interface for impact analysis! and CM2: the cross-platform standard for impact analysis! Thanks to Max Gravel from IpX who was so kind to let me use his images. James Newman triggered me with his comment on my post ‘Would you go on a roller coaster if you knew that #ConfigurationManagement at its supplier was not in order?’ :
“Agree, but the problem here, just as it is in business, is that the majority of people don’t even think about checking on the CM (or maintenance or quality) controls. They “assume” all is in order and someone else has done that check, so they trust (blindly) that all is ok.”
Did you ever feel like you could choke the living daylights out of somebody because they messed up AGAIN implementing a change without checking all possible impact and now you can work overtime and on the weekends away from your family?
To fix a problem you did not cause, while the person that caused it is having a great time with family at Disney!
So performing good impact analysis on changes is key to make good decisions (rejecting a CR is also a good decision if the business case is not providing enough value for the company, the purpose of a change request is not to get it approved, it is to make a good decision). However good impact analysis requires people to communicate, reach out to people from other disciplines/functions, ask questions and verify assumptions. While in a small company this is typically not an issue, because everyone knows each other by face and name, in large companies a lot of people tend to stay in their comfort zone.
Imagine you have to talk to somebody from another department… bhhhhhhh…
No, it is easier to just focus on what you do best and not talk to anyone, that is not sitting next to you and assume somebody else will make sure alignments and checks take place. John is going to review it anyway, so all is good.
Ignorance is Bliss! Hallelujah!
But this behavior can lead to bad decisions and significant cost for a business. I heard a story about an eager engineer who thought of a way to reduce 20% of material cost. This was in a mass production environment. There was no alignment with downstream functions. And when the factory got confronted with the change, they had to change the factory layout. Because the process needed to change to still be able to drill a hole. Even after more than 10 years, the ROI has not been achieved for this change.
Fair to say, impact analysis was not done correctly. Blinded by the prospect of 20% reduction, ignorance hit, not purposefully and unintentionally, but still ignorant for the potential impact in other areas. It is rather a systemic problem because I have to believe that people always want to do their best. People do not want to cause significant loss to the company or in the worst case be responsible for a change that ends up killing people. The system allowed them to become ignorant.
But why is alignment, communication with other people so difficult in these cases?
Small vs Big Companies
Often in Small companies communication across functions is not an issue, because Joe the Engineer knows Jane the production line manager. Whereas in large companies the lines between departments/functions are like the iron curtain. From knowing everyone and talking to everyone and direct feedback on actions (cause/effect) to anonymous, hiding in the masses staying in your comfort zone. In a small company, your network is maybe 100% of all employees and diverse, whereas your network in a large company might just be 1% of all employees and a lot less diverse.
Product complexity is another aspect for some large companies, as typically products are more complex than in small companies, impossible to comprehend by a single person causing people to reduce the context of the change, resulting in assumptions that others will take care of the rest. Which is never the case. Leading to incomplete impact analysis and bad decision-making.
Another problem is that in a siloed company, automation is also often silo-centric. So instead of looking end-2-end, a lot of automation stays within the silo which creates at best sub-optimization while reinforcing a siloed culture.
So from small companies (cause and effect are transparent), where lines are blurred to large companies where silos exist and automation are silo-centric. In large companies, people can hide in the masses where anonymity is the breeding ground for ignorance.
But how do you perform a high-quality impact analysis and what do we expect of the output of impact analysis?
The previous posts on the Impact Matrix describe how the CM2 Impact Matrix can act as a standard for communication across functions and even across the supply chain and help facilitate a good quality impact analysis.