Deductive Logic. A Market Segmentation Tool

As a continuation of our earlier piece on “How To…Segment Your Market”. , we would like to introduce you the process of Deductive Logic as a Market Segmentation tool.

Market Segmentation Tool

Segmentation should be the basis of a Product Marketing strategy.

Segmentation ensures that the product roadmap and its respective marketing plan targets a specific market segment, bringing a closer alignment between customers’ needs and the marketplace offering.

Segmenting a market usually posses both a challenge and an opportunity for Marketers and Product Managers alike since collecting the relevant data and understanding the critical segmentation variables are pivotal to gain reliable insight to identify the ‘market opportunity’. Jumping into quantitative segmentation can be overwhelming especially for products of mass consumption where millions of units are sold through multiple channels and in different consumption occasions (i.e. how do we segment the market of Diet Coke drinkers?). Thus, selecting relevant variables and proper segmentation methods is the first step to define a strategic segmentation that could drive refined quantitative research.

The vast majority of segmentation variables are related to the consumer market: demographic (age, gender, marital status, etc), psychographic (values, attitudes and lifestyle) and behavioural variables (uses of Internet, buying patterns, etc) and requires substantial information to create strategic segmentation that can be used as the foundation for market analysis.

Deductive Logic Segmentation

A simple yet powerful segmentation method to identify relevant segments is the “Deductive Logic Segmentation” method. One way of applying deductive segmentation is through segmentation trees. They work the same way as a classification tree but instead of using statistical criteria, it uses deductive logic (Bock 2009). Deductive logic is powerful because it results in a valid argument as long as the premises are true.

Conclusions will follow a set of premises that are neither wrong nor right, as they are based on our own premises. Its quality depends on the validity of such premises. It could be compared to a mental map that will allow Product Managers to see all possible segments to better calculate the real market opportunity.

In the case of Diet Coke, customer profitability variables are: frequency of consumption (number of units/time) or volume of consumption (size of product/time), and suppliers’ bargaining power is how many similar products can the customer obtain at a similar price and in a similar place. This segmentation will help identify segments based on frequency and place of purchase and also consider the consumers’ choices against competitors. Choosing other variables will also be valid but as deductive logic explains, the validity of the outcome will depend on the variables.

Selecting Segmentation Variables

Selecting variables to segment a market can be intimidating. There is a myriad of options and if we are only at the market opportunity evaluation stage, where little or no data is available then the challenge is greater.

Uncles and Bock (2002) propose a simple qualitative taxonomy of variables that aid in identifying differences between groups of customers. These qualitative categories are: preference for product benefits; consumer interaction effects; choice barriers; bargain power and profitability.

Let’s look at how the previous qualitative taxonomy can help us identify relevant segments for the Diet Coke consumption example

Firstly; we identify variables that are relevant to the product supplier or variables that are only relevant to the product in question. For example, customer profitability and the supplier’s bargaining power. Once all supplier variables are identified, we can further analyse customer specific variables such as preference for product benefits; consumer interaction effects and choice barriers.

Importantly, relevant variables selected need to help answer the original question. How do we segment the market of Diet Coke drinkers in order to identify an opportunity to increase consumption?

Examples of variables are:
i. Choice barriers variables – product availability (sales/channel) – customer knowledge of the product benefits (brand awareness/frequency of consumption).
ii. Variables related to consumer interaction effects could be who is the household decision maker? And preference for product benefits could be – preference for sugarless colas – or caffeine versus caffeine free drinks.

Building The Segmentation Tree

After choosing the relevant segmentation variables, important for both supplier profitability and fulfilling the customer need we rank them logically from more important to least important.

For example, frequency and volume of consumption should be placed high in the tree because it will show profitable segment of consumers directly related to the market opportunity .Brand awareness although important, has less direct influence on the product’s consumption and is more difficult to quantify (although in the case of Diet Coke brand awareness and product availability are key for sales).

First, we have to ask the obvious question: Does the customer drink colas at all? And does the customer drink sugar-free colas? These logic basic questions raise at least 6 macro segments. At this stage, a Product Manager can start discarding or reconsidering segments not relevant to the opportunity she is trying to evaluate.

Continuing with the logical variable ranking; the following order is suggested:

  • Frequency of purchase
  • Volume of purchase
  • Brand awareness
  • Product availability

Discovering Segments

Once all segmentation variables are displayed in a segmentation tree (or classification tree), relevant segments are consequently ‘discovered’.

The point of this exercise is to find the segment or segments that will deliver business return.

As shown in the tree above, consumer segments are identified by following the arrows and multiplying the number of options per segmentation variable . For example, Diet Cola drinkers that consume the product 3 times a week form 27 segments (3 options of product presentation (X) 3 options of brand awareness (X) 3 options of product availability). The same process is applied for all three frequency of purchase options, resulting in 27 additional segments (18 for one time a week diet cola drinkers and 9 for non cola drinkers).

All 54 segments are not necessarily valid or worthy of further investigation.

For example the ‘3 times a week’ diet coke drinker that has high brand awareness and access to the product is not a segment that the business should spend additional resources on. There is no point increasing the consumption of that particular segment given that their propensity to purchase is already high.

On the other hand, it is worth evaluating segments where diet coke is consumed only once a week and brand awareness and product availability is low.

Validating Segments

It is necessary to validate segments that have been ‘discovered’ through this process.

Here are some ways of validating the segments:

  • Differential strategy development: do the differences between the segments suggest that different strategies should be developed for different segments?
  • Accessibility: can the different strategies be targeted towards the different segments?
  • Communicability: will the people responsible for implementing the segmentation be able to understand it?
  • Technical validity: are the true differences between the segments known?

This simple yet comprehensive taxonomy can help us narrow the number of quantitative analysis needed to identify the true potential of a market.

Once the segments have been identified and validated, narrowing them to the valid ones will help Product Managers quantify the real market opportunity and support their product’s business case. Additionally its helps in prioritizing the number of product improvements needed since it brings deeper understanding of the customer group the product addresses.