Tourism has the potential to be a modern-day engine of growth because tourism is the largest service industry globally. This is especially true for developing countries, where tourism plays a significant role in the generation of foreign-exchange earnings (Srihadi et al. 2016). South Africa is no different, and government has identified tourism as being a key sector with the potential to assist with the country’s economic growth (National Department of Tourism 2011). In this regard, the South African government aims to increase tourism’s contribution to economic growth from R189.4 billion in 2009 to R499 billion by 2020 (National Department of Tourism 2011).

More than 10 million international tourists visited South Africa in 2016, which was 13% more than in 2015 (Cape Argus 2017). Various efforts are made by destinations to accelerate growth, for example the implementation of creative marketing strategies, promotional efforts, upgrading of tourism products – all to grow this industry as efficiently as possible. It is the key role of destination marketers to influence and manage demand (Middleton et al. 2009), and they are continuously searching for new ways to do this. To be more effective and efficient in one’s marketing strategy, it is vital to understand the market.

It is, however, not just about numbers or arrivals anymore but about securing significant economic benefits from these tourists for the benefit of the country. In the past government agencies focused primarily on the number of tourists rather than spending. And in this regard Saayman and Saayman (2006a) state that it is better for destinations to attract 10 tourists who spend $100 than 100 tourists who spend $10. The amount that tourists spend has therefore become a better criterion for measuring success than just arrival figures.

Australia, for example, attracts far fewer tourists than a country such as South Africa. However, the average tourist in Australia spends far more than those in South Africa. This becomes even more apparent when one analyses visitor arrival figures and their spending in South Africa, where research confirmed that a tourist from Africa spends far less than tourists from Europe. However, the sheer number of African tourists makes up for their lack of spending (Saayman & Saayman 2008).

The latter prompted this research, which aimed to identify the big spending tourist markets in South Africa. This research differed from that found in the literature, where most other studies analysed only tourist and visitor expenditure (see Craggs & Schoffield 2006; Kruger, Saayman & Saayman 2010; Letho et al. 2004; Saayman & Saayman 2006b, 2008). This study also used a list of 12 key attractions in the analyses to determine if spending differed from attraction to attraction – an innovative approach that to the best of the authors’ knowledge has not been followed before.

This type of research will assist tourism marketers and planners, because a new marketing paradigm is needed for South Africa. Such a paradigm should not focus on mass marketing but rather on targeted marketing for tourists to make a greater contribution to poverty-alleviation in this country. Research also confirms that the greater the economic impact, the greater the contribution to job creation (Saayman & Saayman 2006a, 2006b). Greater success follows from pursuing specific groups of tourists rather than from trying to appeal to the mass market (Morrison 2013).

The approach followed in this study was to target international tourists as they were waiting to board their respective flights back home. They were therefore in a favourable position to give a proper account of their spending during their stay and travels in South Africa.

Data for this study were collected by trained fieldworkers between 26 August and 01 September 2016. The research proposal was sent to the faculty’s ethical committee and was approved. The ethical code is EMS 15/10/15-02-01.

Table 1 presents the socio-demographic data that were collected from the survey at the O.R. Tambo International Airport. The majority of respondents were male (60.5%), had a tertiary-education background of either a diploma, degree or postgraduate qualification (85.7%), were married (54.2%) and had a professional qualification (22.8%). From the respondents, 49% indicated that the main purpose of their visit to South Africa was for holiday, 21% indicated it was for business, 19% indicated it was to visit family and friends, 2% indicated they had visited for medical reasons and 9% of visitors indicated they had visited for other purposes. Furthermore, 23% of respondents who visited South Africa for business also went on holiday to explore some of the sites South Africa has to offer.

In order to better understand how many tourists visited different South African tourist destinations, Table 2 presents the percentage of tourists who visited the following attractions: national parks, the Garden Route, Cape Town Victoria & Alfred Waterfront, Johannesburg, Robben Island, the Winelands, Soweto, the Cradle of Humankind, Table Mountain, Durban beachfront, Sun City and the cultural villages. Websites such as Viator, TripAdvisor, Places South Africa, Travelstart and Safari Now ranked these attractions among the top-rated attractions to visit when visiting South Africa. The most popular destination with 63.6% was Johannesburg, followed by national parks (58.6%) and Cape Town V&A Waterfront (40.1%). The destination that attracted the fewest visitors was the Cradle of Humankind, with 8.1%.

Because it was established which destinations attracted the most visitors, it would be meaningful to know, given a tourist’s socio-demographic information, which tourists are spending their money and on what. Table 3 reports the socio-demographic variables of marital status, level of education, occupation and gender, establishes the mean level of expenditure for each category in each variable and indicates whether the expenditure in each category in each variable differs significantly. The highest mean expenditure for the variable marital status was that of respondents who were engaged, with a mean level of expenditure of South African rand (R) 94 000.00, and respondents who were married, with a mean expenditure of R82 398.75.

Results further indicated that the mean expenditures of the different marital status categories differed significantly from one another (p < 0.05). Respondents who had a postgraduate qualification spent R40 299.57 on average, and the mean expenditures for the different levels of education differed significantly (p < 0.05). Furthermore, self-employed respondents spent R61 009.79 on average, followed by professionals, with a mean expenditure level of R41 743.35. The mean expenditure levels for the different occupations differed significantly (p < 0.05). The mean expenditures for males and females differed significantly, with males spending more than females (p < 0.05).

To elaborate on expenditure, Table 4 contains the mean expenditure per person on selected items, including airplane tickets, accommodation, activities, souvenirs, other transport, retail shopping, food and drink, other and spending per person in general. The mean expenditure per person visiting South Africa was calculated to be R31 297.25.

Table 5, on the other hand, presents the results from an independent sample t-test to determine whether there was a statistically significant difference in the mean scores for two groups (in this case tourists who had visited certain destinations or took part in certain activities, and those who had not). Table 5 shows that tourists visiting national parks, the Garden Route, Cape Town V&A Waterfront, Robben Island, the Winelands, the Cradle of Humankind and Table Mountain spent significantly more than visitors who had not visited these attractions and destinations.

Firstly, the research highlighted specific attraction preferences, the fact that all attractions were not equally important to tourists (confirming research by Morrison 2013) and that tourist spending per attraction differed significantly. It confirmed the popularity of attractions such as national parks, the Cape Town V&A Waterfront and Johannesburg. These attractions all form part of the list of top 10 things to see and experience in South Africa (www.southafrica.net) and are therefore well marketed, leading to optimal visitation.

The value of iconic, well-known attractions for big spenders should not be underestimated by the government and the tourism industry. To sustain the popularity of these attractions, it is important to constantly modify these products to serve the needs of potential, current and returning visitors. It is also evident that attractions such as the cultural villages and Robben Island are not favoured by the big spenders, even though they still spend significantly more at these attractions than the average spenders.

It would be interesting to determine the reasons for this. It could be that these big spenders have experienced these attractions before, or that these attractions do not interest them at all. From an economic point of view this is an interesting finding, because we need to understand the reasons why specific attractions are preferred and why tourists spend more at them compared to other top or well-known attractions.

Secondly, it was evident that socio-demographic characteristics influenced spending patterns significantly, confirming the notion that, in general, education (Snowball & Willis 2006), income (Saayman & Saayman 2006b; 2011), age (Kruger 2010), travelling distances (Pouta et al. 2006), length of stay (Thrane 2002) and being a foreign tourist (Saayman & Saayman 2006b) influence bigger spending, as indicated in the literature review. This research found that big spenders were neither engaged nor married, were better qualified, self-employed or in a professional occupation (earned more) and were male. The latter finding contradicts the research of Craggs and Schoffield (2006), who indicated that females were bigger spenders. These characteristics in themselves represent a specific group of tourists with particular preferences.

Thirdly, and significant to this research, spending was directly related to the type of attraction visited. Significantly more money was spent by those who visited national parks, the Garden Route, the Winelands and the Cradle of Humankind. The unique selling point of South Africa remains the natural attractions. It might also be that these attractions offer opportunities to spend and tourists are then willing to do so. One of the reasons why nature-based tourists spend more is because of the cost involved in staying at some of the lodges, which are extremely expensive but offer very unique experiences. Results on the motives of visitors and tourists also showed that 49% travel primarily for holiday purposes.

Fourthly, as the key finding of this research, two distinct markets were identified. This confirmed the notion that a combination of three markets exists, namely high, medium or low spenders (Spotts & Mahoney 1991). In this case, they were labelled average and big spenders. The importance of big spenders, as well as the relationship between this group of tourists and their attraction preferences, is evident. Big spenders spend significantly more (52.4%) than the average spenders, thereby increasing their appeal as a lucrative market for South Africa. An increase in the number of big spenders can have a significant economic impact, which from a poverty-alleviation point of view would be extremely beneficial to the country’s economic growth plan.

The first implication is that the big spenders have been identified and they should be the key target market if the South African tourism industry is focusing on generating more income from fewer tourists as indicated in the literature review.

The second implication is that marketing efforts should be directed at the high end of this market through direct and personal strategies showcasing these popular attractions and added value and changes or modifications to the attractions that will encourage first-time and/or repeat visitation. Marketing using Twitter and LinkedIn is therefore advised, with personalised messages and loyalty options.

The third implication is that tourism development should follow an ‘all-in-one’ approach providing for accommodation, restaurants, entertainment, souvenirs and unique elements that tourists want to experience and that provide the opportunity to spend money. In addition it remains paramount that tourism products need to ensure that they fulfil the needs of their visitors.

Lastly, for a country battling with high unemployment and high levels of poverty, it has become paramount to follow a more focused marketing approach. A two-pronged approach is therefore advised. In the first instance, a marketing focus on the big spenders is important, as their value lies in spending, and in the second instance the average spenders should be maintained, as their value lies in volume (number of average spenders).

The greatest limitation to this research is the fact that because of financial constraints only one survey could be conducted. A few such surveys over a period of a year might give different markets. It is recommended that one survey per quarter be conducted in order to get a comprehensive view of the spending behaviour of international tourists.