Geography, the most powerful segmentation tool

Many marketing strategies look for small percentage characteristic deviations from competitive brands within the category. Ie. MasterCard - 48% male, 52% female. Visa - 53% male - 47% female. Therefore, the marketer's response is that 'MasterCard has a female skew. We should aim our efforts at females'. As a result, many miss the "big" picture. These differences are statistically irrelevant and could result in MasterCard not talking to 48% of its customer base!

No longer completely loyal

In mature markets, customers are no longer 100% loyal to one brand. Ninety nine percent of consumers shop from repertoires of brands they know and think well of. Each brand has some 100% loyal customers, but most buy or consider other brands more often (especially over time).

I mostly buy my groceries from Pick 'n Pay. However, when I come home late from work, I will shop at Woolworths or Spar, which stay open later - all these stores are accessible and convenient. I also buy fresh produce from Woolworths and enjoy the bakery section of Spar.

As consumers, we may choose different brands or retail outlets within our repertoires based on different expectations, needs, moods or wants but these variables are not predictable and are based on random behaviour. However, it is still the same person displaying this promiscuous behaviour.

Similarly, we all cross-shop for clothing at Edgars, Woolworths, Mr Price etc. So do I fit the "Mr Price profile" which is thought to be different to an "Edgar's shopper profile"? This not only holds true for retail outlets, but across brands in any category. A coffee drinker may buy Nescafé and Frisco over a period - which "profile" of user do they fall into?

The reality

The reality of any shopper behaviour is that brands compete in a market, not as individual brands. Marketers and media planners should define more clearly the profile of the market/category of users in which they operate.

As a product category, nappies should be aimed at 'people with babies', and not relegated to single/married mothers in LSM 7 - 10, living in urban areas etc.

Where we have found differences in the profile of users is by location. A retailer such as Shoprite may have an idea of what its shopper profile is based on demographics. In truth, a Shoprite shopper in Edenvale is very different to a Shoprite shopper in Chatsworth. However, the Woolworths, Checkers, Spar and Pick 'n Pay shopper in Edenvale all resemble the profile of Shoprite consumers in that area, and similarly the same is true for Chatsworth. An Edgar's shopper in Vereeniging has a very different demographic profile to the Edgars shopper in Sandton.

Very similar

In reality, shopper profiles of competing brands or stores within an area are very similar across all options within a category and in turn, these profiles resemble the profile of the area, and not as a generic 'one shoe fits all' demographic profile.

At NAB, our epiphany is that consumers with mental availability (high salience levels) and physical availability of the brand or retail outlet are more likely to use these brands or shop at these stores.

Understanding the category users/shoppers within different areas is more important to marketers than trying to own a segment based on demographics. Geo-demographic segmentation is invaluable when determining what product offerings to feature or what tone to use in your communication.