measure and estimate

Many of the issues that I hear about from our IT consulting clients can be traced back to either an inability or disinterest in scoping and estimating projects. The woodworkers maxim of “measure twice, cut once” has been replaced by “I don’t have time to build up a detailed task estimate” and “it will work itself out as I’ll be high on some estimates and low on others.” More often than not, we find that our clients have solid processes for estimating project size and complexity but don’t bother to use them.

Shorter timelines, greater use of outside service providers (both service delivery and SaaS) and more frequent shared delivery responsibilities between IT and the business are dramatically increasing the need to develop better estimating skills among IT professionals. Gartner and others are now out on the street with the proposition that “Failing Forward Fast” will create a culture of creativity with a cost of 20-28% of projects failing. Getting to a place where we can estimate LOE and forecast realizable benefits is key to succeeding as an internal or external IT service provider.

The tips below summarize some easy, straightforward options for increasing your estimating skills:

  1. “Where have I seen a project like this before?” Using your prior experience with similar projects to estimate the LOE and other factors for a new project is the best place to start. Tracking prior project results (LOE and benefits achieved) will help over time. Even if you don’t have that info handy, look back at project documents and make some guesses about the staff capacity of the team members that worked on the project as a place to start.
  2. “Smaller chunks are better.” Breaking down projects into phases and their component parts will allow you to estimate better and to get more input from others on the reasonableness of your estimates. Going to your Dev or QA team with smaller components will increase the odds of their estimates being better and will help the group learn over time where they are missing the mark.
  3. “Separate what you know from what you think.” Draw a box around the areas where you have a lower confidence level in your estimates so that you can more carefully measure progress and readjust the schedule if you see slippage in those areas. Reminding the project sponsor that the original estimates were fuzzy in a couple of key areas and providing updates during delivery will diminish the surprise factor if you run out of calendar.
  4. “Develop, review and reconcile multiple estimates.” Having multiple team members develop their own estimates often produces different results. The good news is that the ensuing discussion (i.e. an estimate review) typically produces better results. With estimates developed for smaller chunks, the additional effort is negligible and the payoff is an improvement in estimate quality.

Getting the estimating process to work well is hard; using an incremental approach to get your staff moving in the right direction will make it easier. As your estimates improve, your internal customers will be happier and it will get easier to find funding for those additional resources that you need.

Drop me an email at with your best suggestions for improving the estimating process and we’ll share those in a follow up blog.