Succeeding at BI Projects

i’ve had a ton of experience in this area, especially with the data vault methodology.  the methodology is comprised of best practices from many different facets.  my intent with this entry is to introduce you to some of the foundational cornerstones that i feel are important to bi project success.

issues are everywhere, and sometimes it seems no one is willing to take ownership and make the necessary changes.  let’s face it, change is hard to do!  blame is easy, and often leads to an “us versus them” mentality (like it versus business) which shouldn’t happen at all.

it needs to change their thinking, the project managers need to address some foundational concepts in small grass roots projects in order to begin to affect their own culture.  cultural change is not easy, and is fraught with hazards, but if i can do it – so can you.

the time i did it, i was in a company of 159,000 employees, 7 sectors of business, with 59 companies – and it was it’s own never never land of implementation and system maintenance.  the good-old-boys network was as strong as ever.  i managed not only to change the basic it culture (within my group), but was able to create a first-ever repeatable delivery system for on-time and in-budget successes.

i was able to help the company succeed at turning our it department in to a profit center for our business unit sponsors, and at the end of the day – we were all a team, all part of the business working with the business instead of for the business.

here’s the short of it:

  • treat your project plan as a business plan – in fact, go out and get yourself a program called: business plan pro (if you’ve never written one).  try to see if your project will actually make money, and if investors would invest in it.
  • divide your business plan in to segments, deliverable phases if you will.  these are typically called “traunches” in investment speak – as you reach milestones, hopefully risk is reduced, and reward is increased, thus more investment money is available.
  • be transparent, although one can argue exactly what this means, i would suggest (as i have done) to setup an internal it wiki, or project web-site that is available to anyone and everyone who is inside the corporate firewall.  this not only breeds trust for your group, it is a fair bit of advertising of everything (your successes and your failures).  hopefully you’ve documented good mitigation strategies along side your failures.
  • stop treating them as “business users”, and start treating them as “customers”.  your customers can be it, or business, internal or external, if you continue to use the term “business users”, you are not only limiting yourself, you are damaging the reputation of your project.
  • use proper estimation tools.  learn something called function points (yes, it was popular in the 1970’s), but can be very very valuable.  if you can show your customers that you can set estimates, meet estimates, or be within tolerance levels of your estimates, you can win more customers – because of your transparency!would you leave your car at a repair shop for “work to be done”, without having an estimate for how much it might cost, and how long it might take?  i didn’t think so…  don’t expect your customers to trust you either, without getting good at estimations.
  • it’s ok to make mistakes, just be transparent about them.  make sure that you setup a good risk analysis process, and that you identify potential risk mitigation strategies.if investors couldn’t measure the risk of the investment, do you think they’d invest money?  no, most certainly not – that’s why a lot of “good ideas” are never funded into a company…
  • have clear communication channels, don’t hide secrets
  • if you live in a world of charge-backs, then be accountable.  if you need someone elses time and they will charge to your project, then they should fill in actual hours of specific tasks that they accomplish during the day.  it may seem like a lot of overhead, but trust me, it’s the only way to gather true actuals against your estimates, and to know if you’re meeting your goals!

these are just some of the techniques i’ve used in all kinds of projects in the past.  i hope this has been helpful and/or insightful for you.   i’m currently working on new material for the foundational aspects of the data vault methodology, and releasing it to the coaching area.

some of the foundational cornerstones of the data vault methodology are as follows:

  • project management processes (pmp)
  • project management institute (pmi) beliefs
  • capability maturity model index (cmmi) process improvement
    • key process areas – kpa’s
    • key process indicators – kpi’s
    • key process measurements/metrics – kpm’s
    • library of itil – information technology infrastructure library
    • control objectives for it (cobit)
  • six-sigma (quality driven measurements in the project)
  • total quality management (tqm)
  • lean initiatives (li)
  • cycle time reduction (ctr)
  • function points & project estimation tools
  • risk analysis and risk management
  • waterfall, spiral, and agile scope controlled systems
  • process automation components

i’m also beginning a new book on the methodology side of the data vault, so now is the time for your input, let me know what you’re thinking – and if these notions/concepts make sense to you.

thank-you kindly,
dan linstedt
ps: you can learn on-line at:

Tags: , , ,

No comments yet.

Leave a Reply