Which Segmentation Method Will Yield the Strongest Business Results for Your Brand?

Customer segmentation has always been something pharma have implemented. But until now the most common traditional approaches were either based on how much of your brand they use, or by personality profile or market research response, or by demographics, or a combination of these.

However, now with the breadth and depth of customer data we have available to us, we can use far more powerful behavioral segmentation. Using artificial intelligence to implement behavioral segmentation is changing the way segmentation is done with far more powerful results. Behavioral segmentation is segmentation based on patterns of behavior shown by individuals as they interact with a company when making decisions about using/buying their brand. It allows you to determine the customer segments in a way that allows you to address the needs or desires of the customer and maximize your results as your customer travels along their customer journey.

There are many types of behavioral segmentation used today but they all confer advantages in a few key areas such as personalization of the experience, prediction of the next action and need of that customer, prioritization of the best content, channel, sequence of events for that customer and they provide significantly stronger, measurable, business results.

Job to be Done(JTBD) Approach
The way we have found that helps our clients the most when planning where opportunity spaces are for new brands or brand portfolios as well as for providing insight into how to position a portfolio of brands, is utilizing a ‘Jobs to be Done’ framework. This framework was developed by Prof Clayton Christensen at Harvard Business School as a complement to his theory of disruptive innovation which is about competitive responses to innovation. The disruptive innovation part of the story explains why your company will be disrupted and the ‘Jobs to be Done’ framework shows you how to create products and services your customers want. It completely transforms our understanding of customer choice because it clarifies the causal driver behind the decision – which traditional correlations never could. For fun, you should check this website out http://tylervigen.com/spurious-correlations – it is interesting how strong the correlation is between the per capita consumption of cheese is with the  number of civil engineering doctorates awarded.

Back to the topic at hand, within a ‘Job to be Done’ is a job. It is called a ‘job’ as it is centered in the thinking that someone ‘hires’ a product or service to get a job done. For example, we ‘hire’ a computer to help us when working, surgeons ‘hire’ scalpels to cut into soft tissue, and so on. We also ‘hire’ a drug to help us alleviate certain conditions in certain circumstances. While this is obvious, most companies are still not using this powerful approach when planning their drug marketing and still wander through customer interviews trying to find nuggets of insight often using techniques such as correlation to do so. Many pharma companies, instead of systematically applying a ‘Jobs to be Done’ framework to their data to uncover how and why your customers are ‘hiring’ your drugs, and for which situations, generally stick to the more traditional approaches.

What is interesting about the ‘Jobs to be Done’ approach is that you when you apply it, you come to understand that the circumstances around a decision of which drug to ask for, or prescribe, are far more important than the product attributes or customer attributes. This really throws a lot of the traditional old types of segmentation on their heads. It is the deep understanding of the range of circumstances facing a physician and their patient that shed light on how the drug should be marketed most effectively to capture maximum sales.

This approach is very useful for drug portfolio marketing planning. Often we get the question, ‘How can I maximize the uptake of a new drug in my portfolio, while continuing to maximize my existing blockbuster drug? How can I understand more fully the ‘what’ of how the market is structured now and the ‘why’ behind the structure with the dynamic changes occurring in the market place due to the new products entering the space? Or ‘Why are there not more physicians prescribing my drug? How can I get more uptake?’

For these types of projects, we leverage the existing internal data, collect new data and add in big data such as claims data, and analyse with Artificial Intelligence. This approach assesses the relationships between symptoms, tests, treatment changes and decisions while also linking it to an understanding of both the patients and physicians and their influencers as well as which brand choice they make when and why (both the conscious and subconscious why’s). Following this process enables us to more completely characterise the market structure as well as understand the reasons and causes underlying this structure. As a result, we can determine what is happening, why it is happening and where the opportunities lie and how large each opportunity space is.

This approach is essentially a new way of determining physician and patient segments using behaviour and rationale for behaviours. You no longer have to use the old methodology of looking at past prescribing behaviours, disease symptoms and treatments but can understand what the physician is needing to be done with or by a treatment when faced with a specific patient circumstance.