Scoring

National Analysts often develops models using primary research, customer data, and secondary sources to augment existing information from these sources. After models have been built, we will score existing prospect/customer databases, or perhaps an integrated database, with predicted values from these models. Examples may include one or more predicted customer segments; purchase propensities for individual products/services; and bundles of products/services, likelihood-to-switch, and total wallet.

Modeled values are advantageous because they provide information that is not available in raw customer or prospect databases or from secondary sources. Models extend the value of detailed custom research, typically collected on a small subset of a firm’s customers and prospects, to the entire customer/prospect universe. 

The models National Analysts constructs make maximum use of primary research and secondary sources to enhance predictability. The model development process typically includes an evaluation of multiple alternative methods; all modeling methods are tailored to fit the nature of the available data. In particular, models make proper use of sample survey research data which might require sophisticated sampling and analysis strategies. These efforts not only ensure that model predictions are as accurate as possible, but also ensure that benchmark measures of predictive performance that are calculated using the survey data are seen in the overall dataset.

Matching/Merging
Fusion
Scoring

For more information on National Analysts, feel free to e-mail us today or call (215) 496-6800.