Original Post Date: Wednesday, April 14, 2010

The first time I remember hearing the term, “data driven estimating,” was at an ISPA/SCEA conference in Phoenix about 6 years ago. I thought I had a pretty good concept of what that meant. Then, I started hearing it more frequently about 6 to 9 months ago, and I wasn’t as sure of my understanding. So, I went to the Internet and Googled, “Data Driven Estimating.”  The first 3 pages yield links to statistical, engineering, and technical estimating founded on data; e.g. packet routing, data delivery rate, radar and rain gage data, spectral density, Markov chains. A few relevant hits surfaced, but none really define what data driven cost estimating means, with one possible exception: “Lessons Learned and Results from Applying Data-Driven Cost Estimation to Industrial Data Sets,” a presentation to the 6th International Conference on the Quality of Information and Communications Technology (QUATIC 2007). What did it say about data and estimating? This: “….few estimation methods can deal with imperfect data.” Wow! That’s not real encouraging.

 

There was one other hit that I found interesting, probably because it was the closest thing I saw to my interpretation of data driven estimating. The site is: MyBusinessMusings (MBM); http://www.mybusinessmusings.com/. Here’s the part that connected with me:  “In order to become data driven, business leaders need real-time, actionable data in a usable form. That’s how you empower a significant cultural shift: provide the right information at the right time and in the right format.  Anything less doesn’t enable the data-driven organization. …. We all want to be data driven, but the cost of usable data may prohibit…..”

 

I see a connection to parametric estimating – here it goes: the data we have for cost estimating is imperfect, and the cost to make it at least readily usable by traditional accounting methods at the right time is prohibitive. Mathematical modeling of traditional accounting (parametric estimating) is the best response to the imperfection situation. That’s the way I see it.

Bruce Fad,

VP of Professional Services & Cost Research, PRICE System