I wanted to write this around Halloween as the title would have been more fitting, also thought about changing it to the “Grinch Who Stole the BI Project Delivery Date”, but didn’t really like it even though it is funny, or not so funny when it happens to your project.
If you have ever been on a Business Intelligence Project or any Information Technology project for that matter it is no secret that scope creep is a killer and can send any project into a tailspin. Some of you might not agree with my next statement about who is responsible to ensure this does not happen, which is okay. I welcome any healthy debate on the subject. The person responsible is the Project Manager.
One of the many duties of a Project Manager is to ensure that the delivery date of a BI project is met. For this to happen a project plan must be crafted with milestones that collectively add towards meeting the delivery date with some buffer room because things happen. The following milestones are examples and should not be followed verbatim. One milestone (first) is to have a fully defined scope that contains business requirements from which technical specifications (second) can be created for the engineers to architect or design (third), build (fourth), then test and verify (fifth) a solution. So when the VP of Marketing etc. who is paying for the project out of his budget comes to you when you’re near the third or fourth milestone and says “you know…the marketing spend per product measure we talked about two months ago, we need to add this that and the other because its not really complete how we defined it so please do that and add these additional figures” (via email). This very situation has happened to me more than once and sometimes you are lucky to get an email and not just a phone call. This is how it usually starts and then immediately you go to the team and ask if we can do this because the VP said it needs to be done…and so on and so forth. The best way to handle this situation is to document the exact conversation of what is desired, get most of questions out of the way, then have a meeting with your project team and explain the situation. Have the team provide an honest assessment with some buffer room and then communicate back to the VP the detailed feedback from your team. I am not saying he or she will be happy if you tell them the team needs an additional three weeks, but they will understand if you communicate the facts to them and you just did your job of managing the project. Just saying yes has the potential to blowup your well oiled plan to deliver on time. Also, be creative and explain that three weeks can be added to the plan but that will delay the delivery date, or we can circle back and address this in an additional phase after the initial delivery date.
There are also some things you can do to mitigate scope creep. Keep the time between scope and rollout as short as possible with smaller more iterative subject area releases. The longer it takes for your team to deliver a solution, the more of a chance you are exposing your project to business changes that were valid six months ago but not today. I have had conversations with business folks that say it all needs to be released at the same time as it is all very important. When you discuss the possibility of the business changing if you wait six months, and that you have determined that their business measures really can be segregated into four areas that can be released separately, they will see what you mean. They want the solution earlier rather than later in most cases and in my experience it is not all (collectively) critical, rather certain pieces are more than others. Here is another area where you just managed the project successfully.
When I go into clients to assess a project that is in trouble it usually has some scope creep that happened once or multiple times. You can throw more resources at a project to increase the time to delivery but there still needs to be some re-scope or it has the potential to be another source of scope creep. Your original plan was probably devised for your planned resources (if done correctly) and now tasks have to be assigned to the new resources to make it useful. Don’t forget it is going to take the new resources some time to get up to speed. I hope this helps those out who manage BI Projects or IT Projects in general.
Happy Holidays as well. I want to write another blog entry before the Holidays get here but who am I kidding this one was supposed to be done for Halloween 🙂