Practical Predictive Analytics
上QQ阅读APP看书,第一时间看更新

Internal data

Your own internal data is the first and best place to start. Many projects start with examining data from business operational systems, such as sales systems. You can start by analyzing transaction data, such as looking at all orders for a customer and analyzing the number of units sold, total sales price, and so on. However, transaction data by itself does not tell you anything about the customer. You need to also enrich your basic data by augmenting it with other attributes which can be obtained by matching common elements of your database transactions (such as customer ID, product keys) with valuable data which exists in other related databases. For example, a rich source of data is demographic and attitudinal data which may be stored in company customer surveys.

In addition to customer surveys, there may be other internal repositories that you can use to derive insight. Web logs may reveal more information about customer clickstream buying patterns. Data can be captured from customer call center data, HR and other administrative systems.

Finding other data sources often involves a lot of digging, since often internal data is siloed in large organizations.