Estimating IT Project Efforts

For those of you with a fiscal year that starts 1/1 you are probably ramping up your planning and budgeting efforts. When trying to determine which IT projects you will add to next year’s plan there are two constraints that one must consider: staff time and money.

Of the two of these, I find the human constraint to be the most limiting. You could double our capital budget, and we still couldn’t accomplish significantly more.

Even though we are acutely aware of this contraint, we still manage to commit to more projects than we can complete.

At the planning stage it is extremely difficult to estimate how much staff time a project will require. There isn’t enough time before budgets are due to develop detailed project plans. Universally people always underestimate the amount of effort a project will take. Even seasoned veterans make this mistake over and over. I guess we always assume that things will go to plan. the fact is, that they never do.

Vendors are little help. The time estimates they provide for projects are not based in reality in any way. One of our HIS vendors states that their eMAR and medication scanning product can be implemented in 3 – 4 months. Our experience is more like 15 months., and we have our A-team working on that project. You can’t even get your medications unit-dosed and bar-coded in 4 months.

I was discussing this tendency to underestimate effort with one of my colleagues. He suggested that we should start multiplying our best guesses by 3. I was thinking 4. We are both probably underestimating.

8 thoughts on “Estimating IT Project Efforts

  1. Will – One thing I always find interesting is that hospitals do not hold vendors to their estimates in the contracting process by including the project plan in the contract. If you commit the needed resource to the project from the hospital side, then the liability falls on the vendor for project overruns. That’s not a slam on hospitals…just something I’ve always been curious about.

    The other issue is that most vendors’ implementations are vendor supported rather than vendor driven and hospital supported. Hospitals often do not have the resource or application experience to commit to the project. Why not make the vendor do it? You might pay a little more on the front end, but resource and time savings on the backend offsets the cost.

    I am moving my company to this model and can get eMAR installed in 4 months. Whole house rollout for a 600 bed hospital in 8 months.

    The issue of resource will continue to get worse as the demand on the hospital increase.


  2. I have always been perplexed by the way budget discussions begin with “staff time and money.” Why not “vision”? Why not “what are the three to five things we must accomplish in the coming year?”

  3. Will, great post as usual. You said “the time estimates they [vendors] provide for projects are not based in reality in any way.” I couldn’t agree more, but that’s because in many cases there is no level of detail.

    Most vendors don’t spend enough time thinking about the “deployability” of their products and need to be held accountable for it.

    If we as customers are willing to see the word “estimate” as a synonym for “guess” (which is what I’ve unfortunately seen over and over again) then we deserve what we get.

    However, if we ask for signficant detail in our estimates include vendor project plans, combined with our project plans, and do dependency analysis, resource combinatorics, and social impact resolutions we can start to get meaningful estimates that are reasonable and practical.

  4. As a technical recruiter specializing in hospital emr implementations i see this constantly. Hospitals offering 3-6 month contracts for implementations that end up lasting 15 months, plus. In fact, I have come to expect this, so have the contractors. So whats the deal guys? Is ignorance bliss?

  5. I agree with Shahid…most vendors have never been involved in developing a detailed project plan for their products. Most of the don’t even own project management software.

  6. A little off topic, but following the thread . . . Most vendors should not be selected to implement their products. The unfortunate reality is that application vendors have a difficult enough time with development and marketing that by the time it gets to the sale they just push it out the door. This creates the fertile ground for growth and churn for engagement managers, PM’s and SME’s throughout the industry.

  7. Thought I would come back and reverse my last post a bit from my experience at a hospital today. A large hospital in the Midwest is in the process of replacing their 60+ Pyxis stations with vendor of choice M. I was brought in at the last minute to oversee the integration into our pharmacy system with the new stations and the robot. I warned them several times that there had not been enough testing and planning for something of this magnitude, but onward we went. Things turned out ok, but I was struck by the total lack of planning. They had 10+ people on site with no coordination. Many of the common sense things were totally overlooked…. Will, I see your point.

  8. Planning and estimation of medical facility IT projects is problematic for several reasons.
    1.)The inexperience of clinical, administrative and IT staff with the rigors of project management.
    2.)Inadequate IT skill sets in most facilities.
    3.)The poor quality of software in the health care industry in general.
    4.)A large set of legacy software designed in the 1970s.
    5.)Poor quality integration tool sets.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s