Dr Angela Karlsberg, head of analytics at research and insight agency McCallum Layton, explains what segmentation is, how it works and the pros and cons of different methods. She also reveals how successful segmentation can help companies increase their target audience and inform the development of new products and services.
What is segmentation?
Marketers use segmentation to identify subsets of buyers within a market who share similar needs and display similar buyer behaviour. There are many ways you can segment a group, from demographics and geography to socio-cultural factors and attitudes.
Why segment your data?
The benefits of a successful segmentation can greatly outweigh the time and costs involved in its implementation. Not only does it allow you to gain a deeper understanding of your core customers and their needs, it also helps you identify your most and least profitable customers, which will help to protect and increase revenue. The challenge is identifying the most effective segments.
How does segmentation work?
For the segmentation to be successful the segments need to be relevant and actionable in terms of their number and definition. First, you need to determine the number of segments you’re going to work with. If there are too many, it’s difficult to see what variables discriminate between them; if there are too few, you may lose valuable information about the differences between the segments.
The optimum number of segments will vary for each segmentation strategy; however a good rule of thumb is four or five segments. When deciding on the number of segments to use, marketers need to achieve a balance between the extent of discrimination and the actionability of their segments.
Once you’ve calculated the number of segments you’re going to work with, the next step is to ensure that the segmentation will be effective i.e. that each segment is genuinely different from the others.
First, you need to determine if the segments are large enough to be profitable. Your research findings will help you to quantify how many consumers are in each segment and their size in the market.
The segments must also be actionable – that is, they must allow you to develop an appropriate and unique marketing mix to attract and meet the needs of each individual segment.
Finally, consider whether it”s possible to easily reach all your different segments. To target them effectively the segments need to be linked to information that you already hold or which can easily be acquired – such as location, age, gender and socio-demographics (Mosaic and Acorn).
Tip: Make sure your segments are profitable, actionable and accessible.
How do you start to segment?
Various techniques can be used for segmenting a market but the method adopted usually depends on a trade-off between the cost of undertaking the segmentation versus the level of accuracy you need.
A priori segmentation
This simply uses pre-existing data such as socio, age, gender or geo-demographics to segment the market. However, while these segments are easy to target, they are usually quite crude.
For example, if you segment your market on age then you are assuming that people within the same age band think alike and make the same purchasing decisions, which is an unrealistic assumption.
Tip: Review all the evidence which you already have from previous research and focus on the factors that appear to drive differences in behaviour.
Customers are segmented according to their usage/purchase behaviour, such as heavy, medium, light users or buyers. This is a common technique which assumes that different usage patterns reflect underlying differences in needs and attitudes.
The segments are then profiled to establish which demographic variables drive purchase or usage behaviour within each segment.
Financial segmentation for loan use
For example, we carried out a segmentation study for a financial client where we segmented their customer base according to their past, current and further loan usage. This enabled the client to determine which segments were the most loyal, the ones most and least likely to take up product offerings, which segments had competitor products and which competitors they used.
This information was used to create a more effective marketing strategy, based on the information known about the profile of customers who fell into these segments.
Tip: Give the segments memorable names that summarise each segment”s key characteristics.
Attitudinal needs-based segmentation
The most powerful type of segmentation combines usage data with attitudinal research. The aim is to understand how people’s attitudes are linked to their usage and purchasing behaviour.
This is typically the most actionable form of segmentation as it allows you to understand what drives usage as well as which product/service attributes drive overall satisfaction within each segment. The technique for this type of segmentation is called ‘cluster analysis’; it ensures that members in each segment have similar characteristics but the segments differ from one another as much as possible.
We carried out a successful segmentation campaign for a pharmacy chain by creating an attitudinal needs-based segmentation. The segmentation was based on customers’ attitudes towards factors that would influence their choice of pharmacy and their attitudes towards health and lifestyle.
This enabled us to create a segmentation where customers within a segment had similar characteristics and needs yet each segment’s needs and characteristics were very different.
Tip: Use qualitative research prior to the segmentation to help identify all variables that might discriminate between different types of consumers.
Which segmentation method to use
There are advantages and disadvantages to each of the segmentation techniques detailed above. While ‘a priori’ segmentations are the simplest to apply and can be based on existing data, making them relatively inexpensive to carry out, they are the most crude and least powerful of the three techniques.
‘Usage’ segmentations are slightly more difficult to undertake, and yet are also relatively low in cost as they too are generally based on existing usage/purchasing data. However, the extra work required is rewarded by the production of a more meaningful segmentation which incorporates additional data over and above basic demographics.
‘Attitudinal needs-based’ segmentations are the most expensive and complicated of the three techniques. They normally involve undertaking a substantial number of usage and attitudinal surveys, which can be costly. The segmentation is based on advanced statistical techniques and is therefore the most difficult of the three techniques to undertake; however it will produce an insightful and powerful segmentation.
While the technique chosen will vary by company, in all cases the overall objective is the same – to identify significant groups of customers with differing needs and to address them effectively in order to protect and increase profits.
Tip: Review the segmentation regularly to reflect any major changes in the industry sector or market.