Defining and Using Segmentation, Part 1

“Segmentation” is a word that quite often means different things to different people. It’s all the rage in Web analytics. Everybody’s doing it, and all the Web analytic tools offer it. But what is it, what does it mean, and how can it be used?

In simplest terms, segmentation is the process of dividing a group into subgroups. The idea in marketing segmentation is there are some meaningful differences between the subgroups that can be useful for marketing purposes. There are two main considerations when doing segmentation: the approach you use to segment and the basis upon which you segment.

The two main approaches to segmentation are deterministic and discovery. The basis on which you segment might be along the lines of:

  • Demographics and lifestyle
  • Behavior
  • Attitudes

Deterministic Approaches

With deterministic approaches, you create segments based on some predefined or predetermined classification. It might be a relatively simple classification, like male and female, or it may be more complex, like “First time visitors with abandoned shopping carts containing yellow socks.” With deterministic approaches, you hypothesize the segment is interesting, important, or valuable, and you test that hypothesis.

Most Web analytics tools offer this approach to segmentation. To varying degrees, they provide the ability to divide or extract visitors into different groups and run reports comparing different groups against each other. In addition, you may be able to extract email lists and other details from the segments for outbound marketing purposes.

The ability to segment and analyze different visitor subgroups is increasingly important. You can’t continue to run the site as a one-size-fits-all business. Deterministic approaches are useful for trying to identify meaningful differences or to understand underlying behavior in more detail. However, you have to hunt, and you may not always hunt in the right direction. This is where discovery-based techniques can come in to play.

Discovery Approaches

By discovery-based techniques, I mean statistical and other data-mining techniques, such as cluster analysis (define) and neural networks (define). Having done a stint in the market research industry, I often think about these techniques when discussing segmentation.

Cluster analysis is a statistical technique that segments the population into subgroups that display some commonality. There are many different cluster analysis algorithms that vary in their application and complexity. The overall objective of cluster analysis remains the same, though: maximize the similarity of the members within each subgroup and maximize the differences between the subgroups. In other words, you want each member of the subgroup to look as similar to every other member as possible (all part of the same club) and for each subgroup to have distinct and meaningful differences from each other (all the clubs are different).

Cluster analysis is an iterative statistical process. Therein lies the rub. The statistical process can create distinct segments, but that doesn’t necessarily result in meaningful segments. Using these types of techniques is as much an art as a science. Just because the analysis software reaches a statistically correct result, doesn’t necessarily mean it’s a useful result. Remember, too, these techniques are dependent on the data that you start with. As the old saying goes, “Garbage in, garbage out.”

Neural networks are a more black-box kind of technique, based on the way the brain works. They use artificial intelligence algorithms, such as Kohonen networks, to find relationships or patterns in the data. With classical statistical analysis techniques such as cluster analysis, the analyst has more control over the analysis process and can more easily interpret the findings and output. Data-mining techniques such as neural networks can be more powerful but also can be more difficult to handle (bit like driving a Ferrari, I imagine).

In either case, segmenting is only half the battle. The other half is about understanding what the segmentations mean and what can be done with them. Typically, the output of a cluster analysis will tell you a person belongs to a segment. You then have to work out what it is that characterizes the individual segments and what the differences are between the various segments. This is the profiling stage.

How the segments are constructed will be based on the data that goes into the analysis. So if you use some behavioral data to create the segments, then the differences will be based on those behaviors. That’s the first place to look. However, you will also usually want to pull in other data to help explain what the segments mean. This can be demographic or attitudinal data, for example.

Segmentation can mean different things to different people, from simple classification to more complex pattern discovery approaches. Next, I’ll look at the different types of data you may want to segment on, and how they may be useful to the Internet marketer.

Till then…

Related reading

Big Data & Travel
Flat design modern vector illustration concept of website analytics search information.