In this post about product-market fit, I briefly mentioned market segmentation in the context of understanding whether a product is a fit for a specific market segment. However, (market) segmentation deserves its own dedicated post, as it plays crucial role in a product development.
(Market) Segmentation definition from Wikipedia:
…the process of dividing a broad consumer or business market, normally consisting of existing and potential customers, into sub-groups of consumers (known as segments) based on shared characteristics1
How I would clarify it further:
…dividing a group of people or companies into ‘distinct’ categories, within which each member is expected to behave similarly.
The Influencer Segmentation Example
In the post about product-market fit, I talked about the product idea to help influencers monetize their presence effectively, and provided provided a basic framework for segmentation to help identify unique problems of each influencer segment. There are approximately 50 million influencers in the world1 in different categories, countries, follower number etc. In this large group of target audience, there are several major differences among certain groups, thus they have different problems. This makes segmentation exercise crucial for the product team to prioritize the segments of influencers to focus on, identify their top problems and solve them.
Why is segmentation important?
To sum up, segmentation helps us
- Determine smaller and distinct groups of users/businesses and understand their unique problems and needs better
- Analyze our product’s market fit for each segment seperately, so we can determine which segments are more receptive to our product vs others
- Decide whether we want to evolve our product beyond the MVP to address unique needs of different segments
- Prioritize segments based on their size, opportunity and our product’s abilty to win them
But how to do it effectively?
There are a few important considerations for the best use of segmentation in product development:
Decide on the right ‘attributes’
An attribute is a differentiation point to divide the target audience (people, customers, influencers etc) into different segments. The same target audience can be divided into different groups using different attributes.
In the influencer example above, we talked about countries, number of followers etc as attributes for segmentation but we can also add category (sport, gaming, beauty etc), as well as language. In another example, let’s say we are building a SAAS solution, and want to target companies to use our new internal messaging tool. Then, it is critical to understand what size of businesses, or which industries this product will resonate most and solve messaging problems. While deciding on the segmentation attributes, we need to think of the clear differentiation points between two members of this large group in terms of decision making: why would one company make ‘significantly’ different decisions compared to another company? It is important to define ‘significantly’, as it will help us prioritize the most important differentiation points. A Telecommunications company makes advertising investment decisions significantly different than an Ecommerce company. Even though there are still nuanced differences among Ecommerce companies themselves in terms of the same decisions, they are ‘similar enough’ to be categorized under one segment, and targeted accordingly.
Ensure attribute ‘values’ are dividing the market effectively
Now, we defined the attributes for the segmentation exercise, we know what differentiation point we will use to split the audience. However, for that specific attribute, how do we define the values? The attribute values are defined as the way attribute is used to divide the target audience.
For some attributes, values are clear. For example, if the attribute is gender, there are clear definitions of genders we can use. Or, if the attribute is age, then there are common age brackets we can use for attribute values like <18, 18-25, 25-35 etc.
However, for some other attributes, definitions need to be customized. For example, if we want to use ‘business size’ as the attribute, there are several ways to potentially use attribute values. Does it make sense to use values as companies with 0 to $1bn annual revenue & companies with more than $1bn annual revenue? Clearly, it would not make sense as those two categories are massive, and among the companies in each of them, there are significant differences. Therefore, a more ideal attribute value set should be something like 0-$100K, $100-$500K, $500K -$1m, $1m – $50m, $50m-$500m, $500m – $1bn, $1bn+. Now, you can argue that some of these groups can be merged, assuming the difference between them is not significant, I would not disagree. Especially in certain categories, such deep segmentation might not be required. For example, if we are segmenting CPG companies, there are clearly finite number of global conglomerates with significantly higher revenue figures than the local brands. The attribute value set can be limited to only a few for this particular industry.
End up with optimal number of segments and size of each segment
It would be ideal for a product to target all addressable segments, meet their needs, and eventually scale across all to gain maximum growth opportunity. However, in almost all cases, there are certain segments that will be more receptive to the product than others. Therefore it is important for us to see which segments see value of the product, so we can scale within them, and which segments do not see the value and require a different solution to address them. The question though is how many segments we should have.
Considering the formula: Total number of segments = Number of attribute values1 x Number of attribute values2 x ……Number of attribute valuesn , n being the number of attributes, if we have 3 attributes with 3 values in each, that would give us total of 27 segments. It would be very challenging to analyze 27 different segments in terms of their interest and fit to our product. Therefore, keeping the segmentation to the most important few attributes and values will make it easier for us to analyze the reaction of each segment to our product, better understand their needs and potential for market fit.
=====================
After all, segmentation should be fully customized based on the vision of the product and the problem(s) it is aiming to solve for. It must be done early on in the product development cycle to ensure that progress can be tracked, and analyzed effectively, and actions can be taken according to the insights coming from each segment.
References;
- Market Segmentation by Wikipedia
- The Rise Of The Influencer: Predictions For Ways They’ll Change The World by Forbes