The variables used to define LSMs were biased towards durables. Unfortunately, these are not necessarily an indicator of material wellbeing. Individuals may have purchased the respective durables on credit and thus become indebted. Technology also featured among the LSM variables.

Unfortunately the rate of technological change now is too rapid for such items to provide a stable measure over time. A smartphone was an expensive item for a niche market. Now one can be bought for several hundred rand.

SEMs were developed to reflect the way people live in South Africa, a country with a particularly high Gini co-efficient.

Initially, 113 variables were taken from the existing LSMs, Afrobarometer, the International Wealth Index and from the input of an expert panel, to ensure that they were relevant to the continent. These were then subjected to a robust and objective process of correspondence analysis to arrive at a parsimonious set of differentiating variables, which captured the socio-economic landscape of South Africa.

Structural items and infrastructure proved to be important differentiators of how people live. The type of shelter, roofing material, type of floor, way in which you access water, whether you have hot running water or a flush toilet are some of the reliable variables that differentiate people’s lives in South Africa.

To acknowledge these realities requires that the industry ceases to look at the market through Western – and particularly “Sandtonian” – eyes.

A further example of this is that proximity to a police station and post office is an indicator of infrastructure development. The question about these have been asked in terms of “nearness”. Although this might seem a loose term, it deliberately allows respondents to 'self define' this, because a car driver will have a different sense of what is close, compared with a walker.

SEMs however, do not disregard durables completely. The variables include microwaves, freezers, washing machines and floor polishers or vacuum cleaners.

A reassuring point is that the SEM segmentation approach delivers a flatter picture of South African society than LSMs did, with their bell-curve shape. This is a more probable depiction of an unequal society with a low GDP per capita.

A further improvement that the SEM system offers is that it acknowledges that trying to divide society into “piles of people”, in the belief that these piles are based on discrete differences, is flawed.

The reality is that people mostly fall along a continuum. For this reason, although there are 10 basic groups, there is also a scoring continuum, running from one to 100, which allows users to define their own bespoke segments.

SEMs are an improved approach to segmentation, which is relevant to South Africa. They have no correlation with LSMs, so there is no quick conversion factor that can be applied to change the one measure into the other.

It is a new system that media owners, marketers and agencies need to learn to use to gain a true representation of local target audiences.

It is time to move forward and it is incumbent on all three industry stakeholders to work on the adoption of SEMs.

For more information, visit www.mediamark.co.za. You can also follow Mediamark on Facebook or on Twitter.