Tuesday 16 July 2013

A Short Rant About Rogue CMMI Consultants


If you've been waiting patiently for the next part of my seven deadly sins posts, I do apologize. The post has been written for some time, but the subject matter could have been misconstrued by the folk I was under contract with and I didn't want to get caught up in a needless conflict. That contract is now over so I will post in due course when I feel that the time is right. If I'm brave enough, I might also reflect on some of the issues that have made the past twelve months even more challenging than usual!

In the meantime, a Twitter post caught my eye last week which sent me reeling. Actually, it wasn't so much the post, but the link behind it that caused my kernel panic - which got worse as I read more articles by the same indvidual on his website. Again, to avoid unnecessary unpleasantness, I'm not going to directly reference the person or provide a link to the website in question, because, at the end of the day, that person is entitled to their point of view, just as much as I am.

The post that sparked this off was about "creating a Quality Management System designed for a company going for CMMI Maturity Level 3". (I'm sure regular readers can already see where this is going!)

My first reaction was to respond by asking - "why would you want to do that?" - to which I have never received a response! My second, was to follow the link and read the underlying article and some associated posts. It turned out that the author has a list of CMMI and ISO certifications as long as your arm, and is a specialist CMMI consultant. And every post on the blog was directly or indirectly about getting through a CMMI Appraisal (or ISO certification).



Not once did I read about adding value to the customer or the underlying business. The overall impression was that a 'CMMI Implementation" will lead to success in your software development organisation, regardless of the context of a wider set of business requirements. There was no mention about the principles and concepts underpinning process improvement. A document centric approach was the focus - perhaps unsurprisingly as the company could provide an off-the-shelf QMS to meet any specific CMMI Maturity Level requirement.

Some of us have been around long enough to understand that this kind of attitude and approach is precisely why CMMI (and other models and standards) have such polarized reactions to their usefulness, and why in so many cases, improvement initiatives fail so dismally.

If your focus is on achieving an arbitrary maturity level through appraisal or pinning a certificate or badge on the wall without having clearly defined business objectives, it is highly unlikely that you will manage anything more than a short lived success. You will not achieve much in the way of a return on investment and you will probably unfavourably prejudice future improvement opportunities.

What I find most sad about this whole episode is that a consultant can hide behind their own set of certifications and peddle such nonsense to an unwary, often ill-informed, lemming like audience as an absolute truth. In the past it was easy to lay some of the blame at the hands of the SEI - after all CMMI was their cash cow. It's easy to turn a blind eye to what is going on in the real world when the cheques keep appearing in the bank. Let's hope the CMMI Institute has better values, morals and principles regarding the messages they wish to convey. Certainly some of the people involved in the new venture believe whole heartedly about how and why CMMI can be used as part of a suite of tools and techniques to improve a business, and many have written passionately about its short comings in the past.

Even then I doubt they can exert much control over rogue lead appraisers, or trainers, who have paid their money and done the necessary work to achieve their qualifications, any more than they can exert any control over people like me, who only have years of hard earned experience and a set of notebooks full of horror stories of how it can all go so badly wrong.