Major stores still dominate SA FMCG shopping

Major retail outlets are responsible for more than half South Africa's sales turnover of groceries, toiletries and confectionery - yet they constitute only 2% of the stores selling these products.

Research company ACNielsen's recently completed 2002 census of grocery, toiletry and confectionery (GTC) stores shows that a total of just 1 375 "majors", which include hypermarkets and supermarkets, account for a whopping R35,2 billion in annual GTC sales. This is 54,5% of total national sales.

Among the chain stores, SuperSpar and Spar have the most outlets in South Africa, with 492. Next is Shoprite with 248, followed by Clicks 228, Pick 'n Pay supermarkets 112, Woolworths 102, Pick 'n Pay Family stores 87, Checkers/Hyperama 92 and Pick 'n Pay Hypermarkets at 14.

By contrast, 68 379 "non-majors" - which include cafes, superettes, forecourt outlets and rural counter-service stores - ring up just R29,5 billion, or 45,5%, a year. Many of these stores are extremely small, independent shops in both urban and rural areas, and do not contribute significantly to South Africa's overall retail trade figures.

ACNielsen conducts its store census on behalf of FMCG manufacturers, who use the resulting database to calculate sales efficiencies and draw up distribution strategies.

The valuable information contained within the census statistics includes detailed breakdowns of product line turnovers in particular store types.

This year's census employed enhanced technology involving the utilisation of 'enumerated areas' (EAs), as used by Statistics SA in its latest national population census. Statistics SA has defined more than 80 000 EAs in South Africa - these are the smallest geographical regions for which population data is released.

In the current methodology the census team visited a sample of about 6 000 EAs and physically counted all the stores it could find, enabling ACNielsen to enumerate more stores than in the previous census. These results were used to project to the final store universe.

The selection of EAs was made using scientific probability sampling, and was "stratified" - with a greater proportion of the sample being chosen from areas with higher population. "Much of this work was done using the latest technology, including the Global Positioning System (GPS) plotting of stores, which has resulted in tighter controls," says Ian Hobson, ACNielsen marketing manager for retail measurement systems.

For a store to be included in the ACNielsen census "universe" it must stock at least four food or four confectionery product classes, it must have a fixed location, it must be plainly recognisable as a trader, and it must be accessible from a public thoroughfare. Specifically excluded are school tuck shops, canteens, kiosks, hawkers, and mobile stores.

In the 'majors' category, in addition to Checkers/Hyperama, Pick 'n Pay Hypermarket, Pick 'n Pay supermarket, Shoprite, SuperSpar and Spar stores, the 2002 census has included Pick 'n Pay Family, Clicks and Woolworths stores for the first time. This has added 9,4% to GTC coverage in South Africa.

Two factors have driven a change in store classification. These are the rapidly changing retail trade structure through the development of branded superettes and forecourt stores, and manufacturers' need to identify channels differently. Among the non-majors, the defined sales channels reported are: branded superettes, such as Kwikspar, Foodies and Score; filling station forecourt stores; urban counter-service stores; urban self-service stores, and rural counter- and self-service stores.

Although the adoption of new methodology limits accurate comparisons with the 1998 census, the new statistics do suggest a marked fall-off in GTC turnover in KwaZulu-Natal and increases in Gauteng and Western Cape, says Hobson.

Users of the census, both manufacturers and retailers, have welcomed ACNielsen's enhanced methodology.

Rory Rochat, marketing director of confectionery manufacturer Nabisco, says: "It is a positive change in the context of a rapidly changing trading environment in South Africa. It will enable marketers to track the changing performance and the changing patterns in the market place."

Rochat said he particularly liked the inclusion of the garage forecourt category for the first time, as well as the differentiated self-service and counter-service measurements.

"Companies like us rely on these outlets - what we call 'down-the-street' - for the majority of our confectionery sales. So a more realistic gauge of what they are selling is a big improvement," says Rochat.

Russell Mackie, senior manager: Consumer and Market Knowledge for consumer goods maker Procter & Gamble SA, says: "The changes are all good steps towards a more accurate read of the South African market. The adoption of more retailers into the retail audit sample really gives us a more realistic picture of the market."

Mackie says Procter & Gamble has already noticed some significant changes in the company's national share figures as quoted under the new controls versus the old.

"Granted, the once-off change in the audit universe makes trend reporting a bit more difficult; however, our ACNielsen account team have been walking us through the changes and pointing out the impact of the change. Dealing with a blip in the trends is a small price to pay for a more accurate read."

He agrees that the adoption of new definitions for store types is more meaningful. "It makes targeting these channels more actionable."