Friday, March 26, 2010

One small step for IT, one giant step for Data Quality.

I had a vision. My vision was to get buy-in from IT resources to start data management activities at the start of every IT project. Tough thing to achieve when performance is strictly related to delivery, not quality. I could speak [rant..whatever they wanted to call it :P] to them about benefits until I was blue in the face, but until I could describe what the benefits were in terms of project delivery risk management, they quickly went back to managing projects to the way they are used to so they could quickly start on the next one.

I knew that to really get them to listen I needed to put it in their perspective. What are the project risks they encounter every day? What goals are they trying to achieve and how did this tie into the satisfaction of their clients?

Here is what I know:

  • They are busy all the time
  • They have a lot of projects to deliver and limited resources - so no time to even think about new processes
  • They have some satisfied and some not-so satisfied clients
  • They are measured on project delivery (dates) not project quality

So how did I get their attention and show them that pro-active data management activities could help them achieve some of their goals?

I made a diagram. Everyone loves a diagram and no one has time to read text anymore. Plus, I made the diagram in such a way that it has a look similar to a Gantt chart. The goal of the diagram was to show that by not integrating data management activities into their project processes they were actually increasing the data related after project churn: Change requests, data clean-up, data reconciliation etc. These post project activities also used more not less resources (including business resources), which in turn cost money and time and led to less than satisfied clients.

When put in that perspective, most of them nodded their heads in agreement and even provided their own examples of what that after project churn looked like. I’d like to say that after a few of these presentations the corporation saw the light and made a broad announcement that from this day forward Data Quality is a top priority! I’m kidding here…! What did happen though is that the requests to provide Data Quality direction in pre-project analysis increased significantly thereafter. One small step for IT, one giant step for DQ :)

Here is the diagram. Click on it to see a larger view.



I

8 comments:

  1. Great job! I used to use this approach when working in corporate architecture, but I think the examples of data rework is much more straightforward for folks to grasp - they see this much more readily in their day-to-day activities.
    Way to go!

    ReplyDelete
  2. Great idea with the diagram. Whoever said a picture is worth a thousand words new what they were saying. Awesome of you to take on that challenge!

    Another element that adds value to the project would be "information optimization" (IO) The project can leverage old data and research, add tags, bookmarks, notes to facilitate collaboration, and find 'relevant' information specific to that project. Would love to see that factored in on your diagram. May even help you show real economic value.

    ReplyDelete
  3. My approach was to make it simple and relevant so thank you for Marty for the validation from one such as yourself!
    Vivisimo, I'm less familiar with the 'optimization' but would love to see how that would look in the diagram and certainly it would make it more comprehensive. If I sent you the file would you be willing to incorporate?
    Thanks again!

    ReplyDelete
  4. Excellent diagram Jill!

    Simple, direct, easy to understand, and difficult to argue against.

    Thanks and Best Regards,

    Jim

    ReplyDelete
  5. Jill, to add to your post, starting this week, we are now entering our time against a post-project activity for all post release data related activities. This mean that IT will soon start seeing the real resource cost of the reactive approach to data quality ... the numbers will confirm your diagram.

    Nicole.

    ReplyDelete
  6. Nicole, you have made my day! I will be interested to see how the results are communicated out.
    As a side note, I think I will make the following changes to the diagram:
    1/ add resources to the post project effort for the 'desired' project process to reflect the governance. The effort will be shown as minimal.
    2/ Add a milestone line to reflect project end date.
    Any other comments and feedback would be most appreciated.
    Thanks Nicole!

    ReplyDelete
  7. Jill,
    while I totally agree with the idea shown in the picture .. but where is the proof? We've got to have some indication that the extra three resources do indeed save four in the long run.
    Also, my old boss would say that everything is more expensive during the project while the project is going on, and the (alleged ;-)) cost savings are after the project is over -- so why should the project care?!

    Again, I do agree with the idea shown, but I would still have some problems convincing the bean counters. What's your experience?

    Thanks for the diagram
    Thorsten

    ReplyDelete
  8. Thanks for the excellent feedback Thorsten! You make a good point. Where I come from, the proof is easily quantifiable:
    1/ we have on average 5-10 post deployment Change Requests or Problem reports directly related to data
    2/ we have 1 full time resource manually resolving data issues on behalf of the business (again, after the project is over)
    3/ we have hired on average 6-7 students each summer (for the past 3 years) to resolve data related problems that are a direct result of implementing IT projects without a sound governance approach.

    Let me please say that the data related issues that need to be resolved are in no way directly the result of IT project delivery. They are due to a combination of a lack of stewardship/ accountability on the business side, and a lack of comprehensive data related risk management on the IT side.

    My goal is using this diagram and speaking to IT resources was to raise awareness within the IT project delivery community, in the hopes that there would be increased understanding of the need for data risk management as part of their delivery, and they could then identify these risks to business project sponsors.

    ReplyDelete