A million different ways to cut a cake
Customer Segmentation can be a real head-scratcher. The whole principle of taking your entire customer base and then splitting it up into smaller, more manageable homogeneous chunks is widely practiced and understood. That being said, organisations don’t always get it right. For most companies, the biggest obstacle is understanding the best way of approaching it. Rather like cutting a cake, there are a million different ways in which you could do it.
Historically, many product or manufacturing orientated businesses did this along product lines. However with customers often buying different products from across their portfolio – and demanding some kind of discount or loyalty bonus as a result – businesses soon started to realise that they needed to think about their customers in a different way. This naturally led to a financially based approach, with businesses organising and prioritising customer segments on the basis of revenue and profit potential. The downside of this was that they started dehumanising customers by thinking about them as units of money rather than people, thereby generating a distance between them and their buyers.
Over time, this led to the introduction of other approaches and information, including geographical, socio-demographic, psychographic and behavioural data sets. These, of course, added another layer of complexity to the segmentation process, with each having their individual strengths and weaknesses. There are, of course, inherent dangers from categorising people on the basis of their location or academic attainment. Today it would be wrong to automatically assume that an upper middle class ‘chap’ would be more likely to be reading Nietzsche than your average working class ‘bloke’!
Psychographics have also come under fire for being inaccurate from a behavioural perspective, as you can often observe a significant difference between what customers state as being their preferences and what they do in reality. One specific example of this was a survey taken of London city commuters, asking them what newspaper they read. This representative sample indicated that a very high proportion of people claimed to read a broadsheet newspaper on their way into work. However these figures did not tally with the newspaper sales data showing the reality of a significantly higher proportion of red-top tabloids being purchased.
On the other hand, whereas behavioural data provides a very accurate representation of reality, it does not consider the aspirations and desires provided by psychographics. After all, perception is reality in the mind of the customer, so something can be missed if we only consider people in the way that they ‘are’ rather than the way in which they would like to be seen.
A combination of these data sets is, of course, likely to give the most accurate picture of the customer, thereby theoretically aiding the segmentation process. However in today’s Big Data world, questions are starting to be raised, not only about the management of data; but also the relevance and future direction of segmentation.
The segment of one
With the growth of the digital age, organisations can now have enormous volumes of data flowing into them on a daily basis. This presents the challenge as to how these businesses can make sense of it all; and how they are able to differentiate between the really relevant information and all the useless ‘white noise’ threatening to drown it out.
If the organisation is, however, able to solve this problem, by using big data (which will no doubt get bigger and bigger through social listening, the internet of things and other sources), propensity modelling and predictive analytics; organisations will be able to move ever closer towards an omni-channel ‘nirvana’. In this scenario, there would be enough information to truly create a ‘segment of one’, meaning that with so much information available at an individual customer level, there may be no need to segment the customer base at all.
Personalisation vs. segmentation
Within this personalisation vs. segmentation debate, irrespective of how well the company is able to identify and profile the customer; the question remains as to the extent to which the business would be able to personalise the experience to meet individual customer needs, yet at the same time make a profit from the relationship. The whole purpose of segmentation for organisations has typically been to strike the right balance between meeting customer segment needs and having sufficient volumes to be able to achieve the critical mass and drive the scale economies needed to be competitively viable in a mass market scenario. Surely this will always be the stumbling block. Or will it?
The principle of ‘segment of one’ is certainly possible in the longer term, however the technologies available to deliver the experience will need to play catch-up with the technologies and data currently available to analyse it. We’re certainly moving in that direction. The increasing digitisation of the channel experience via mobile technologies; combined with the increasing cost effectiveness of modern production techniques such as 3D printing are bringing that reality ever closer to home.
In the future, the cost savings people will be able to achieve by not having to get things delivered – as they will be able to manufacture many things at home on their own personal 3D printers – may outweigh the cost savings gained from mass production. That being said, it is unlikely that customer needs and wants will be that disparate, and personal manufacturing will be that cost effective as to completely eradicate mass market production, certainly within the next few decades. Segmentation has undoubtedly got a good few years left in it yet.
Segmentation vs. personas
Some would argue that segments and personas are completely different (as suggested in this article from 2012 by the UX design company Foolproof), whereas others might suggest that a persona is an evolution of a segment. Having worked with both, my recommendation would be that the organisation needs to be getting its segmentation right, before building personas around them. It certainly doesn’t make any sense to create personas that bear no relation to a meaningful and accurate segmentation exercise. Segmentation validates the identity of homogeneous customer groups, whereas personas help to then enrich those identities for other purposes.
Personas ‘lift’ segments by providing a much richer qualitative picture of a typical ‘fictional’ customer within that segment, animating their personality and values. They enable the business to build a detailed story around the actions and motivations of that person through a combination of the extensive quantitative data used for the segmentation exercise and the rich qualitative data that can be gathered from other sources, such as interviews and focus groups with individuals recruited from representative segment samples.
The information not only relates to the needs and wants of the individual relative to the product, service and channel experiences offered, but can also help build a much better understanding of the customer’s backstory. This might include how branded customer experiences influence their lifestyle as a whole, clearly subject to the nature of proposition and its potential emotional influence on their psyche. When you think about brands such as Nike, Mercedes, Apple and Rolex, you start to understand the extent to which some customers might define themselves according to their relationships with these brands.
All the information is collated and expressed in writing and pictorially (e.g. pen portraits, cartoon caricatures and story boards) to present an easy to understand snapshot expression and chronological/ historical representation of that persona. This considers their life from both a functional and emotional perspective, detailing their personalities, values, actions, motivations, frustrations, etc. The personas are then used by the business to help build future-state customer journeys and experiences as part of a CX service design exercise.
So, in summary:
Segments: Here to stay? Well, for a little while longer anyway…
Personas: Bringing those segments to life.