October
1998
THE VALUE OF A/B SPLITS
Tracking list response is a key part of a successful,
long-term direct marketing strategy. But did you know
there's a way to learn more than simply which lists
perform better than others?
Next time you place a list order, consider splitting
individual lists into sub-groups based on criteria such
as job function and company size. In direct marketing
parlance, this is known as an "A/B Split."
For example, let's say for a given list you've selected
job functions ranging from software developers to IT
managers and CIOs. Which of these groups will generate
the higher response? If they all have the same source
code, you'll never know.
When you order the list, specify that you want the file
to be split based on the criteria you specify (i.e.,
developers in Group "A" and management titles in Group
"B") and that the names should be keycoded accordingly
(so your lettershop can identify which records belong
to which group.) Then assign a unique source code to
each group. (It's rather like treating one list as if
it were actually two.)
A/B splits are useful for testing the value of job functions,
company sizes, industries, purchase authority, and more.
It's a good way to evaluate basic assumptions - for
example, your manager insists on targeting management
titles only. Does that make sense? Here's a way to find
out.
A/B splits aren't free - list managers typically charge
an extra $5 per thousand names. But the information
you gain can pay for itself many times over.