The Fundamentals of Market Research in Shopping Centre Marketing

By Dr Dirk A Prinsloo

Urban Studies


  • You will often hear the following questions:
  • “Who is your market?”
  • “Where is your market coming from?”
  • “How frequently is your market supporting your particular centre?”
  • “What is the strengths of your competitors?”
  • “What are the needs of your customers?”
  • “Where does your shopping centre fit into a bigger picture of shopping centre options?”

These questions will be asked by centre managers, centre owners, brokers, possible new tenants, advertising agencies, advertisers, local newspapers and any other community group that would like to use the centre for their own promotions or local community sales, festivals or bazaars. You as a marketer should then be able to base your answers on quantified information obtained through market research. The market research information, however, is of utmost importance for you to fully understand the market you are operating in, to use this information in drawing up marketing communication and promotional events, and to segment your market according to the most viable market alternatives.


The learning objectives of this lecture are the following:

  • to understand the market research process;
  • to identify various types of market research;
  • to focus on the fundamentals of basic market research at your centre;
  • to clearly understand how to interpret the information, and to apply the findings in your day-to-day actions in marketing a particular centre;
  • market segmentation will be used as a broad indication of how market research could be used to divide a centre according to specific market segments.


Any market research process consists of a number of different steps. No matter what type of research is conducted, all these steps are usually included (Diagram 1).


3.1 Identifying the Problem

The problem definition is the first step towards launching a research study. The first sign that a problem exists in a shopping centre is:

  • complaints from customers;
  • the under-performance of the centre or the anchor tenants;
  • vacancies and difficulty in filling empty spaces;
  • or the offering of merchandise other than what is required by the target market;
  • ineffective communication;
  • the tenant mix does not satisfy the needs of the customers
  • the centre experiences very strong competition from competing/new centres.

The problems and challenges and the importance thereof will vary from centre to centre. Decreasing sales and increasing expenses or decreasing profits are all broad indications of problems associated with a shopping centre.

The research problem specifies the information required to progress towards the achievement of the research goals. The most critical research questions in most South African centres are the following:

  • how is my shopper profile changing and
  • how should I position the centre to cater for the changes?

3.2 Research Objectives

The objectives of the research must be identified and formulated with great care to ensure that all those aspects required identifying and solving the set problem will be addressed.

To solve a marketing problem of a shopping centre the main objectives with any research project are to:

  • demarcate the existing primary and secondary trade areas of the centre;
  • establish the socio-economic and demographic profile of the customers living within these trade area;
  • establish the shopping patterns of the customers;
  • identify the image of a centre compared to the major competing centres;
  •  identify the strengths and weaknesses of the centre in comparison to other centres, and
  •  to focus on specific issues like media use and propensity to support specific stores.

3.3 Collecting Data

Collecting data is the most important part of any research process. The information must always be of a very high quality and absolutely reliable. The fieldwork process must therefore be well executed and supervised.

3.3.1 Primary Data

Primary research data are observed and recorded directly from respondents. The information collected is directly related to the specific research problem identified. All the questions that one asks the respondents must be totally unbiased and formulated so that all the different respondents understand it.

Primary research data can further be divided into observations and surveys. In the case of a competitor analysis where it is difficult to obtain direct information observations can be useful determining the size of the store, the number of employees, range of merchandise, level of activity, advertising and promotions and the appearance of the store.

3.3.2 Secondary Data

Secondary data is compiled inside or outside the organisation for some purpose other than the current investigation. Examples of secondary data would include internal sources of sales data by each tenant in the shopping centre.

Sales data – most tenants have to provide centre management with their actual sales figures for the purpose of assessing turnover clause rentals. A simple comparison of sales performance between tenants will highlight those tenants who are prospering and those who are battling. Based on this sales data, the rentals and the floorspace and a number of ratios can be compiled.

Rent/Sales Ratio – This is the ratio of rental to sales experienced by a store. It is simply the rental expressed as a percentage of sales. Stores within the same merchandise category should experience similar rental ratios. Differences do occur as a result of pricing, stock turns, location, quality, and of course the rental rate per square metre paid.

Rent/Sales Index – This is the ratio represented by a store’s share of the centre’s total and divided by the store’s share of the centre’s sales. When a store’s share of sales is more favourable its rent/sales index will be less than 1,00, and when its share of sales is smaller than its share of rent, its rent/sales index will be greater than 1,00.

Trading Density – This is total annual sales divided by the space occupied by the store. One uses annual sales because of seasonal fluctuations with monthly data. Trading density can be calculated on two levels, i.e. using gross store area – the total area for which the tenant pays rent, or nett selling area – the area to which shoppers have access. A comparison of the two is often useful in revealing imbalances in space distribution.

In South Africa trading densities vary widely between store types, and their are also big differences between trading precinct types.

This is an extremely important ratio because it reflects the relative efficiency of space utilisation as well as the adequacy or otherwise of a store’s spaces.

External sources of information refer to the 1996 Population Census conducted by Stats SA. Demographic, socio-economic and house information is available for metropolitan areas and cities on suburb basis. Except for income most of this information does not change rapidly. The data, however, gets outdated after a number of years. With the changes taking place in South Africa at the moment, this information will become outdated faster than in the past. The information is however of critical importance to fully understand your market place.

Another external source that is vital for shopping centre development/management is the information available from local authorities. Population numbers, population growth, the possibility of new commercial or industrial developments and road changes might have a direct influence on a specific shopping centre.

A variety of other sources may provide centre mangers with some information about their centres or competing centres or the environment they operate in. AMPS provide information on media usage, Socio-monitor psychographic life style profiles, and Vision some suburban characteristics. All this information can add value to a better understanding of the customers.

3.4 Interpreting and Analysing Research Findings

3.4.1 Statistical Analysis

The first step in drawing conclusions from most surveys is the tabulation of the data. In most cases cross-tabulation may be quite useful. Statistical interpretation focuses on what is typical, outstanding or what deviates from norms or averages. Various other statistical techniques could be used to further analyse the data. The analysis must be straightforward for everybody to understand.

3.5 Reporting Research Findings

The penultimate step in the market research process is preparing a report of the research findings. The researcher must take a clear objective look at the findings to see how well the gathered facts answer the fundamental research question posed in the beginning. Market researchers must avoid complex studies and language. The researcher must recognise the needs and expectations of the report user. This does not mean that the researcher must give the user the answer he expected but must be objective and unbiased.

3.6 Use as Strategic Input

Research findings are not ‘nice-to-haves’. They must be used on a regular basis as input into issues like tenant mix, communication, design of PR and marketing events, to attract new tenants and to position the centre correctly for new market segments.


Two broad types of data are available for shopping centre market research (Diagram 2).



Diagram 2 clearly distinguishes between primary market research and secondary market research. There is a wide variety of primary market research types that could be used, but for purpose of this lecture the focus will mainly be on shopper intercepts, so-called shopper surveys.


In this case there will continuously be referred back to Diagram 2. The best way to understand the fundamentals of market research is based on the example: Glenvista Shopping Centre, Southern Suburbs of Johannesburg.

5.1 Defining Specific Problems

The Glenvista Shopping Centre is functioning as a convenience centre (±9 000m²) and is very well tenanted with a Checkers, Woolworths Food and other convenience stores. The main question to be answered is, how loyal are the customers and what are there specific needs.

5.2 Research Objectives

The main objectives with the research will be the following:

  • to determine the shopper profile;
  • to determine how frequently the shoppers support the centre;
  • to identify strengths and weaknesses associated with the centre, and
  • to use the market research information to draw up a segmentation strategy for the centre.

5.3 Design of the Questionnaire

The following is an example of a questionnaire to achieve the abovementioned objectives. The questions must be easy to understand, in a logical sequence and not of any leading nature. The key aspect in the questionnaire design is to be objective. The questionnaire can be used in a paper format or on a palm top. Each of these methods has its pros and cons.


5.4 Data Collection

5.4.1 Sample Size

The sample size of any market research project is of utmost importance. As a rule of thumb the number of questionnaires to be conducted at a particular centre very much depends on the size of the centre. The following criteria is used:

  • centres <10 000m² 100 – 130 questionnaires
  • centres up to 20 000m² 100 – 200 questionnaires
  • centres 20 – 50 000m² up to 300 questionnaires
  • centres between 50 – 100 000m² up to 600 questionnaires
  • centres larger than 100 000m² 800 – 1 000 questionnaires

It is important that the specific sample according to statistical significant levels be determined for your particular centre. There are statistical programmes that can assist you in determining the right sample size. In most cases a large sample size would be preferred but budgets are restricted in most cases.

5.4.2 Fieldwork

It must be clearly decided when is the best time for the fieldwork to be conducted. The normal situation is to interview shoppers during a normal weekday (Mondays to Thursdays) covering the whole day from early morning, lunchtime, afternoon, late afternoon, and especially in the case of Glenvista Shopping Centre, late afternoon trade. The research should also be conducted during Fridays and Saturdays because of the difference in shopping behaviour on these two days. For larger centres Sundays are also included as part of the research process.

It is of utmost importance that the fieldwork is controlled and that backchecks be done after completion of the questionnaires. This whole process is very important with a lot of detailed planning, control, supervising and management.

5.5 Data Analysis

The following table is an example of a computer printout for a shopper survey. In most cases the level of understanding of the end user is limited to percentages, averages and the use of simple tables and graphs. The analysis phase is very important because the input by means of the questionnaire design, the fieldwork, calculations, etc. will influence the analysis part of the survey.


Each heading (table number) represents a question asked in the questionnaire.

The total responses represent the number of respondents who answered a specific question (204 in the example). This could vary according to the number of responses on the specific questions. This is the first line under the banner headings and always adds up to 204, i.e. heavy users (85 respondents or 42%) + medium users (60 respondents or 29%) + light users (41 respondents or 20%) + first time visitors (18 respondents or 9%). The percentage adds up to 100%.

The total represents the percentage of variables which fall in each category (e.g. 21% support the centre 2-3 times per week, while 7% seldom support the centre). This column always adds up to 100.

The banners on top of the table represent various analyses which refer to information per category (e.g. 38% of the primary trade area support the centre 2 to 3 times per week. This is statistically 99,9% (+++) higher than the average of the total sample, which is 21%. The 7% in the secondary trade area who support the centre on a 2 to 3 times per week basis, is significantly 99,9% (—) lower than the average of 21%.



 5.6 Report Writing

There are various ways to write reports. This should be done by an experienced person to fully understand the data analysis, the centre as a whole, and to add value to the whole survey process. Reports can be in the following formats:

  • a full detailed report with graphs, tables and explanations;
  •  an executive summary with the necessary recommendations, and
  •  a PowerPoint presentation to summarise the whole process and present the main findings.

The following is an indication of some findings regarding a shopper survey conducted at Glenvista.




fund03-slide3 fund03-slide4 fund03-slide5 fund03-slide6 fund03-slide7 fund03-slide8 fund03-slide9


The objectives set for this particular survey were to determine how frequently the centre is supported and how loyal the shoppers are. From the abovementioned note the following:

  • a high socio-economic status with 70% of the shoppers in the LSM 9 and 10 categories;
  • 87% of the shoppers support the centre on at least a weekly basis, which is substantially higher than the benchmark figure of 76%.

This information, together with the media use, would be used to highlight the strengths of the centre.


There are a number of ways to segment the support from a particular area to a particular shopping centre.

6.1 Selecting a Specific Market Segment

Table 6.1 displays the four most common ways to segment a market, based on demographic, geographic, psychographic, and product attribute factors. Each of these approaches, or the combination with the others, represents an opportunity that can be satisfied with a product or service.

bm05table2 fund03-table6

Demographic Segmentation

Demographic variables are among the most widely used segmentation approaches. They owe their popularity to two facts:

  • They are easier to observe and/or measure than most other characteristics.
  • Their breakdown of demographic and socio-economic features like sex, age, family lifecycle, population/ethnic group, education, income, occupation, family size, religion, and home ownership are often closely linked to differences in behavioural patterns.

In many instances, you can combine demographic variables to produce a more meaningful breakdown rather than relying on a single criterion. For example, it is common to combine the age of the head of the household with the family size and the level of household income.

If four age levels, three family sizes, and three income levels are distinguished, a total of 36 segments result. Using a combination of primary data, secondary data, and judgement, you can then determine the value of each segment and thus arrive at a well-thought-out conclusion about which segments warrant your efforts.

Lifecycle stages together with income categories offer an ‘easy to market’ segment for specific shopping centres. Shopper-tainment is focussing more on a younger adult and a family profile. Some centres cater more for the empty nesters. Life cycle determines needs and shopping behaviour.

Demographic information is available from Stats SA (1996 Census) on a suburb basis. This is used to give a broad indication of the characteristics of people living in a particular area. This type of information is very useful to give a broad description of the trade area of a particular shopping centre.

Geographic Segmentation

Geographic segmentation is relatively easy to perform because the individual segments can be clearly defined on a map. It is a sensible strategy to employ when there are distinct differences in climatic conditions, access to transportation, proximity to round-the-clock service or repairs – as well as with such geographic considerations as varying regional tastes or unique culture-based habits and behaviours.

Geographically, you can segment by region, city size, by population density, or by other geopolitical criteria. However, such segmentation is effective only if it reflects differences in need and buying patterns. Many firms, for example, adjust their advertising efforts to as small an area as a number of suburbs.

The demarcation of the primary catchment area of a shopping centre is therefore very important. It is so important to understand the geographical distribution of different age groups, life cycle stages or income categories. In the case of Cresta ±58% of the market in the primary catchment area live in suburbs well above average, 25% in very affluent suburbs and 17% in average and below average priced suburbs. This is an easy way of segmentation. Different socio-economic status groups reside in specific suburbs. The segment approach can easily be implemented. This is an indication of the geo-demographic approach. The distribution of marketing material can be distributed according to this spatial distribution.

The user of desktop mapping programs linked to a well-developed geographic information system (GIS) can clearly indicate the different spatial patterns present in the catchment area of any shopping centre.

Psychographic Segmentation

The most difficult form of segmentation results from the application of psychographic variables, such as life style, personality, and self image. Banks, car manufacturers, and liquor producers, to name a few, benefit from the advantages of psychographic segmentation. It is a branch of market segmentation that is still evolving and promises great vitality in the future. This type of segmentation is however very difficult to measure. Psychographic or life style research attempts to segment customers according to their activities, interest, opinions, and very importantly, their shopping behaviour. Socio-monitor used to be a very popular life style segmentation in South Africa.

Product Attributes

Product attributes include usage rates defined as: non-user, ex-user, potential user, first-time user, and regular user groups. This is a very applicable segmentation method for shopping centres.

In practice, such information is further broken down to distinguish non-users, light, medium and heavy users of the shopping centre. Heavy users of a shopping centre can be identified as those shoppers who support the centre at least on a weekly basis.

The following gives an indication of the support levels of specific size shopping centres. This information can be used to segment specific user groups and to reward loyal shoppers with specific loyalty programmes.

Between 1998 and 2000 the proportion of shoppers visiting smaller centres (<10 000 m²) on a weekly basis has declined from 81 to 76 per cent. In larger (regional) centres (>30 000 m²) the decrease in weekly shoppers has been from 72% to 67%. (See Graph 6.1)


The average time spent (Graph 6.2) in the smaller centres has dropped from 49 to 42 minutes since 1998. By contrast the time spent in regional centres has increased to 108 minutes over the 95 noted in 1998.


From the two graphs it is evident that the smaller centres have lost out both in terms of time spent there and the number of weekly visits by customers; indicating the increasing degree of competition at this level of centre. In part the declines are attributed to the appearance of smaller convenience centres and isolated shops that together create a greater choice at the level of convenience shopping.

In the case of regional centres despite a small decrease in number of visits made time spent in them has increased. The reason for that is apparently due to an increase in both the size of the centres and their tenant mix. In addition more shopper-tainment stores have appeared along with coffee shops, as well as more browsing types.

‘Loyalty’ to a particular centre can be measured by the number of visits to it relative to the last 10 visits to any large centre. Regional centres enjoy very loyal shoppers with a factor of 7.4. In smaller centres the factor is only 4.8 indicating a strong level of competition. (Graph 6.3)


These breakdowns, in turn, would trigger different motivational appeals to improve the level of responsiveness from various segments. In other applications, companies with high market share might be especially eager to attract potential users, while smaller competitors with lower market share would devote their efforts to converting existing users.

Overall, you will find shopping centre attributes most practical in segmenting a market, and particularly applicable in deciding when and how to deploy a promotion event, to offer sales and how to communicate with your loyal shoppers.

6.2 Application of Market Segmentation

From the information obtained in the Glenvista survey, it is clear that the main segmentation could only be done based on the demographics. This is normally the case for smaller centres like this. In the case of larger centres the segmentation can be based on:

  • LSM classifications,
  • heavy, medium and light users,
  • to distinguish between households from the immediate vicinity and office workers coming from a wider area,
  • to distinguish between local and tourists visiting a centre,
  • to distinguish between different media user groups.

Based on the survey conducted at Glenvista most shoppers are heavy users of the centre, with 87% supporting on a weekly basis or more frequently. The most important therefore is to address all communication to a more affluent resident living in close proximity, with a LSM 9 or 10 profile, and an average monthly household of ±R20 000. All marketing and tenanting of the centre must be aimed at this particular demographic profile. In a small centre like this the support market and the target market is the same. In other centres a specific focus on a new target market to be attracted to the centre could be identified.


It is of utmost importance that the market research process is correctly structured and executed as indicated above. Once this has been done the market research information can be used in a number of ways. The most important are:

  • to draw up communication, marketing and promotional strategies;
  • to design specific events for the target market or the most important segments;
  • to use this information to draw the right tenants to the centre;
  • to benchmark your particular centre against other centres;
  • to get a better understanding of the competition, their strengths and weaknesses;
  • to quantify the likes and dislikes associated with the centre;
  • to use the strengths and to correct the weaknesses.

From all the above it is clear that the research must become a user-friendly document to be used by not only the marketing staff, but also tenants, management and owners.

It is also of utmost importance to track the changes at centres on a regular basis, especially in the new South African market, changes are occurring rapidly and therefore shopping centres need to understand and focus on their markets to identify changes in behaviour, needs and support. The role of market research can never be underestimated, therefore the importance to make use of the information from quality research.