I recently gave a talk to a business group and described recency, frequency, and monetary (RFM) models. One of the questions I got at the end was to distinguish between the frequency of RFM and the frequency of business contacts and to say which was more important. I explained that there are two equally important dimensions to direct marketing frequency. There is one definition based on data of historical transactions and another depending on the amount of communications a business creates and sends out.
Some customer data systems have wonderful table structures listing accounts and the contacts that make them up: items, orders taken, and orders shipped. The wonderful table missing from some elaborate, enterprise resource planning systems is the one that will tell the company how many times they have contacted their customers since those customers
Our data is carefully input, validated, audited and we have every reason to believe it is right. But if a customer disagrees with the data, should we automatically assume she is wrong?
As database professionals we need to be prepared for discrepancies and errors as well as customers changing their minds or deliberately misstating the facts. But sometimes the problem will be with our database in how we recorded or failed to record the order and its details.
Business-to-business (b-to-b) and business-to-government (b-to-g) mailers face greater list processing challenges than consumer marketers. These challenges include: the churn of business contacts at titled positions and within companies and more address elements with which to deal. The complexity of business and government addresses can, as I will show, make the tasks of matching customer and prospect files more challenging to match and deduplicate than consumer addresses.