Loading ...
Sorry, an error occurred while loading the content.

158850Re: [XP] ROI of "Engineering Practices"

Expand Messages
  • John Roth
    Dec 4, 2013
      Hi Adam,

      If he's asking you, I suspect it means that he doesn't have a good story that ties in to his upper management's goals. I like the comments I've seen about finding out what they are.

      On the basis of a complete lack of information, I'd make a couple of suggestions about what those goals could be.

      1. Stability. The XP engineering practices underlie project management practices that allow predictable productivity.

      2. Defect reduction. The XP practices tend to lead to cleaner software releases.

      3. More frequent releases, which reduces the inventory cost of written software that hasn't been deployed. In other words, getting the R faster for the I.

      Are those goals that would make sense to your client's upper management? If so, does your client have enough experience to say this approach will meet them?

      John Roth



      On 12/4/13 9:22 AM, Adam Sroka wrote:
       
      I have a client at a large enterprise who is trying to justify to his management the expense of adopting TDD/ATDD/Pair Programming, etc. -- basically, the XP practices. To do so he needs coaches in a variety of locations which will cost a considerable amount of money. In order to present the justification to his management this client has asked for information regarding the return on investment for adopting these practices.

      I have seen this come up a few different times, and I'm not really sure how to answer it. My take is that we are talking about safety here. So, we need to weigh the guaranteed expenditure of bringing in a bunch of coaches plus the likely additional cost of near term productivity losses against the less quantifiable risk of loss due to improperly tested software and/or poor communication. 

      I am not an actuary. This seems like a tough nut to crack, because while there could be measurable gains in customer satisfaction and time to market it is difficult to say when/if those will be realized. Even when they are they still probably represent a fraction of the actual advantage, because the majority is risk mitigation. 

      It's like having an old beat-up car and trying to justify buying a brand new one. The old one burns a lot of gas and a bit of oil, parts for it are expensive, and it's a total death trap. But, I own it, I haven't crashed it yet, and it's still running. I can quantify savings in terms of fuel economy and long term upkeep, but that's not enough to justify the sticker price. It's much harder to quantify what I am gaining in terms of airbags and sensors and stuff that may save my life. I intuitively know that I want those things and should pay for them, but I don't know how to assign a value to that. 

      Thoughts? Suggestions? 

      Best,
      Adam


    • Show all 34 messages in this topic