Data-mining helps you know more about your customers using the data that is already available. All utilities have customer information systems, and these systems are designed for the primary purpose of reporting total sales and revenues at the end of every month. A close second in priority for these systems is the ability to quickly find or update one customer's account information. Neither of these priorities are particularly helpful for knowing more about subsets of customers. However, analysts with the ability to do data-mining can access the customer information system to come up with information on types of customers, their characteristics and their general energy consumption levels. This information is often useful for DSM program planning and evaluation efforts. Sample populations and sample lists of addresses, telephone numbers and e-mail addresses can also come from this work.

Vermont Geo-Targeting Baseline Study. Daniel combined customer databases for all of the different Vermont utilities to identify types and sizes of electric customers in specific geographic areas across the state. (2010)

Processed Aclara's online Home Energy Analysis. Daniel processed customer characteristics data from the Aclara online Home Energy Analysis database and merged it with Arizona Public Service's customer billing data, doing fuzzy matches on names and addresses. This was the basis for an evaluation of the impacts of the online energy analysis. (2010)

Massachusetts State-wide Commercial Customer Database. Daniel organized, cleaned and merged commercial measure data from seven Massachusetts utilities. He also processed and standardized commercial customer lists for each utility. (2010)

Conference Attendee Sample List. Daniel purchased InfoUSA data on HVAC businesses and merged it with conference attendee data to create a sample list for telephone surveys for Arizona Public Service. (2009)

ConEd Demand Response Potential Study. Daniel analyzed billing data for all five million Consolidated Edison customers, both Residential and Commercial, to create accurate counts of customers by building type and their energy consumption category. This included in-depth analysis to identify residential use apartments within the Commercial rate class. (2008)

Small Commercial and Industrial Energy Efficiency Market Penetration Study. Daniel performed data mining on program tracking system information in Access databases to support estimates of energy efficiency market penetration for the Small Commercial and Industrial sector at MidAmerican Energy. (2007)