The MAPS survey will be conducting 20 004 interviews annually using an area-based national sample. There are two waves in a year, Wave 1 and Wave 2, with 2 cycles per wave. Each wave is made up of a period of 6 successive calendar months. The first wave of 2020 will run from July to December. The data and insights that we are sharing in this article reflect the results from cycle 1, N= 5 000 across 830 EA’s
Why does South Africa’s marketers need this data?
Geocoded consumer data allow brands to “see their consumers” allowing them to dig deeper, see a bigger picture, grow market share and make strategic decisions faster with more accuracy through a detailed view of consumer data down to a suburb level. Further to this adding a geographic approach to segmenting ensures relevant messages are targeting the right audience, in the right areas, using the right media, thus empowering detailed consumer profiling.
In today’s competitive market brands want to be able to find and target new consumers. One would expect that brands have a good handle on who their existing consumer are, the challenge is to successfully target new consumers with similar profiles, interests, needs and buying behaviors.
And of course, the silver bullet at the end of this process is to maximize your marketing and media ROI through spending most profitable geographic areas!
AfricaScope is a specialist GIS research company and as such we have over the years developed a number of techniques that for the foundation or the GeoMAPS product. The first technique Circle Point Thematics which is a qualitative method to rapidly map and view the results of survey data for quality control & research purposes.
The center of each circle shows where the interviews were conducted. Each point in the circle represents a randomly selected respondent. The color of each point indicates the answer given by the respondent to a question. The question could for example be, “Do you own a Cellphone?”, the red indicates NO and the green indicates YES. This could be true down to a brand question such as “Do you drink Nescafe Coffee?” – thus allowing marketers to get a very quick visual view of the competitive landscape as well as brand or category penetration down to a suburb level.
The second technique is called small area estimation which is a quantitative method to accurately to map survey data to local areas. The next image shows a very visual view of the LSM profile for the whole of Gauteng. In the context of GeoMAPS this view would be down to a suburb level.
The color-coded areas are those that were surveyed. Each color-coded area has multiple weighted variables associated with it from the survey. Secondary datasets (e.g. updated 2020 demographics) contained in all areas is linked to the survey data. A statistical relationship is created between secondary & survey datasets in color coded areas. Machine learning/artificial neural networks are used to generate the data for all local areas.
A look at the GeoCoded Data - Demographics
Demographics - Race
The insert to the left shows a view of the demographic profiling looking at race. Of all the suburbs shown here, Florida has the highest rate of racial integration, with all 4 race groups having good representation.
Hyde Park, Victory Park & Blairgowrie, Cresta, Randburg and Rand Park are predominantly white areas with some colored representation. Alexandra Township is predictably predominantly black.
Demographics - Age
Florida has the highest prevalence of ages 15-25 indicating a slightly older family structure in this suburb, i.e., teenagers and young adults still living at home. Victory Park, Montgomery Park & Randburg have a middle age profile indicating that younger up and coming families dominate this area. Alexandra shows a spread of age groups. This could possibly indicate very little movement out of the family structure.
Demographics - Income
Florida is predominantly a middle-income suburb. If one looks back to the race view, it is clear that this middle income is spread across the race groups. Hyde Park, Victory Park & Blairgowrie are predicable high-income households. The suburbs to the West have a mixed income profile. Alexandra Township has an interesting income profile in that townships are historically low-income per capita areas, but we are seeing higher income households growing in the township areas. This demonstrates the importance of better understanding the SA township market. It is possible in the GeoMAPS environment to ringfence all SA township and analyse and profile each township individually.
Bringing the data to life at a suburb level
In this initial release of data, a few industries have been analyzed: Telecoms, Medical Aids, Stokvels, Food Retail, Media – Radio. It is important to know that all categories and brands can and will be GeoCoded.
Telecoms: Mobile Network Providers
The markers have been color coded according to the brands colors, MTN: Yellow / Vodacom: Red, etc. Florida for example is MTN dominant. Vodacom and MTN appear to be strong in Alexandra, but the other networks have a good presence.
The power of GeoMAPS is the ability to visually see brand and category consumption down to the smallest spatial level, in this case suburb level. This is powerful data for brand strategy planning and understanding the competitive landscape!
Finance: Medical Aids
Florida has the highest prevalence of ages 15-25 and category penetration is low. Affordability may be a factor, but high potential growth area for lower priced medical aids or plan options. Hyde Park, Cresta & Randburg are saturated – brand switching campaign opportunity can be considered. New market opportunity exists in Victory Park, Blairgowrie and in the Rand Park areas.
Thinking back to the mixed income profiles of Alexandra and looking at the penetration of people engaging in this category, it is clear that growth opportunities exist here as penetration is low in the medical aid category.
Shopping Behavior: Food Retail
This specifically focuses on shoppers who do or don’t do their bulk shopping at Pick n Pay. The blue icons represent the point locations of all the Pick n Pay stores in this area shown in this map inset. This is an example of a secondary data set that can be integrated into the GeoMAPS environment.
The insert to the left gives an interesting view of how consumers shop for groceries. Residents of Alexandra and Cresta are more likely to engage in bulk shopping at Pick n Pay.
Thinking back to the income map, Hyde Park had the greatest number of high-income earners. This could explain the trend of No Bulk Shopping, thus inferring that these consumers are engaging in more regular top up shopping.
Closing the loop by understanding how GeoMAPS can be used to maximise marketing and media ROI
Media: Radio 702
As a marketer or media planner, ensuring maximized ROI on media spend is critically important.
GeoMAPS will allow you to layer media penetration by media type, publication, station or channel to get an accurate view of brand or category pentation vs media usage down to a suburb level. Resulting in targeted media campaigns and maximize ROI.
The inset to the right gives a view of the listenership of Radio 702 in the greater JHB area. Good listenership in area like Protea Glen, Randburg, Crest & Randpark Ridge. On the other end of the scale there is very low in Alexandra.
When considering Alexandra, as a media strategist, media spend would not be maximized in this area. However, from a media owners point of view, this is potentially a suburb where the marketing team may want to look at how to increase listenership so as to benefit from value media spend in the future.
GeoCoded Big Data is an innovation that will change the way marketers and media strategists “see their consumers”.
Originally published on africascope-sa.com.