Target measures answer narrow down what must be defined in business intelligence, in a sense of dashboards or Key Performance Metrics. We use the term boiling the ocean often in my professional practice to suggest that scopes of business intelligence cannot be broad-brush approaches. Target measures provide focus and scope. Common target measures are ROI, Net Sales, Efficiency Rate, Response Time, et cetera (Liu, Laguna, Wright, & He, 2014). Probability mining uses predictive modeling methods such as logistic regression, neural networks, multiple regression, and recursive partitioning to predict the probability of particular events happening based on previous data. Econometric modeling is a combination of statistical analysis and economic theory, based on the idea that neither alone can provide an accurate or satisfactory picture of an economic phenomenon.
In terms of business intelligence, these may be combined for a particular approach. Econometric models are more specialized and may not apply to a particular instance, but certainly target measures and probability mining may be. Target measures most often should be defined before probability mining happens, in order to maintain correct focus and scope.
Imagine a human resources predictive model that mines a swath of employee data from SAP to find key correlations, then uses both multiple regression and neural nets to find predictors of voluntary separation. There is a target measure involved there: the new model is applied to the current employees in order to find their percentage likelihood of leaving in the next two years, with a standard error of 6 months.
Another target measure of note is sell-through numbers for a consumer goods corporation. Sell-in shows sales to a particular distributor. However, not all distributors currently provide their sales data back to the company, which shows how much actually got out the door in the hands of consumers (sell-through). It is a target measure, as the delta between in and through is helpful for the business to know, and timely sell-through data can help the sales, marketing, and customer service departments adjust their approaches based on customer habits.
Liu, Y., Laguna, J., Wright, M., & He, H. (2014). Media mix modeling – A Monte Carlo simulation study. Journal of Marketing Analytics, 2(3), 173–186.
Pophal, L. (2014). The Technology of Contextualized Content: What’s Next on the Horizon? Retrieved from http://www.econtentmag.com/Articles/Editorial/Feature/The-Technology-of-Contextualized-Content-Whats-Next-on-the-Horizon-99029.htm
Reiss, P. C. & Wolak, F. A. (2007) Structural Econometric Modeling: Rationales and Examples from Industrial Organizations. Retrieved from https://web.stanford.edu/group/fwolak/cgi-bin/sites/default/files/files/Structural%20Econometric%20Modeling_Rationales%20and%20Examples%20From%20Industrial%20Organization_Reiss,%20Wolak.pdf