The human mind likes to stick labels on things and people. We and the competetition, our team and the other team, our country and the other countries. This simplification and labelling of groups has led to many conflicts and racism. But, it is also a very human way to handle complex situations. At school we have several, non negative, groups like student, teacher, employee, etc.

In the same way we can use this idea of grouping – customer segementation – with our customers. Even without any active form of customer segmentation we still do use segmentation. The paper magazine we use for an ad-campaign has a customer-base. This in itself is segmentation. The persons who are displayed in the ad says something about the target-audience. We all use segmentation. Segmentation can prevent us from making big mistakes, but we have to take care with incorrect segmentation.

One of the most popular segmentations is the age-segmentation. Suppose you have an online video-service like Netflix. Presenting children movies to adults isn’t a big issue, the respons might not be optimal but besides that not a big issue. When you however present thrillers to the children it might become an issue. By dividing the customers into groups (child, teenager, young adult, adult, senior) you can target the movies to the right groups and avoid these problems. It might also help improving the conversion rate of online adverts for example. We must however remember that people in the same segment might not have the same needs. Not all young adults are intereseted into buying a new laptop.

Most companies perform segmentation based on demographics. Often the reason for this segmentation is the availability and it works quite well… Examples of these demographics are age, gender, income, location and education level. Neighbourhood statistics might be available and that combined with a location can give you an idea of the average person in the area where our customer lives. It might not be correct but on average it is. The danger is that we trust to much on incorrect data.
When customers do specify information, for example birthdate, we can calculate the age and use this information. When we don’t know anything about our customer – for example a visitor of our website – we can trace the ip-address to a location and use demographic data as long as we don’t have any information that is more thrustworthy.

A category that can be somewhat more difficult is behavioural segmentation. How often did the customer buy something, which products is the customer interested in? Sometimes this data is easy to get (customer watches a movie, based on interest of other customers we can make a suggestion) but we need to build a database preferences of other customers. YouTube is a good example of behavioural segmentation. When I watch some movies about motorcycles YouTube will give me more video’s about this and other preferences. These algorithms to find relevant content can become very complex and dynamic.

Besides demographics we also can use psychographics to segment customers. We can use the group belief, lifestyle, etc. to find relevant data for this customer. That means we have to analyze and profile the customer. This might be even harder and it becomes easier to make a mistake here. When someone is interested in a specific subject it does not allways mean that that person is part of that group.

Segmentation can really be a real advantage and improve how you communicate with the customer, you become more relevant to that customer and this can lead into higher conversion ratio’s. Something that has to be avoided is over-segmentation. Do we really need the segment Woman, 24yrs, Single, Lives in New York, University Degree in International Business ? This becomes really hard to maintain and is probably not effective. The segment: Woman, Young Adult, City, University is probably more specific then we need and we could probably do with less… Keep the segmentation simple and effective. It might also be not-so creepy 🙂