Of the 380 questionnaires distributed, 138 users provided responses. Two thirds of the respondents were female and 43% of respondents were in the 19-25 age bracket. The results show that all respondents visited the Internet at least once a day, with 69.57% spending several hours a day on the Internet (Figure 4.1). This confirmed them as the target population for the research work presented in this thesis. Interestingly, not even one respondent selected the available options of ‘Weekly’, ‘Monthly’, ‘Yearly’, ‘A few times’ or ‘Never’. Hence, the population segment analysed were clearly in possession of substantial Internet usage knowledge and experience, and thus could be relied on to provide insight into the type of advertisement necessary to address their needs.
51 Figure 4.1 How often the Internet was visited
When asked why they used the Internet, the largest number (over 26%) mentioned social interaction, which appears to be a major incentive for the younger generations (see Figure 4.2). Another large number (over 25%) stated that they used the Internet to help with studying. Additional reasons for surfing the Internet included working (19.36%) and shopping (17.87%), with a minority stipulating that they would used it for play (8.3%). The relatively large number answering unprompted that they used the Internet for shopping shows that a large proportion of transactions have moved from traditional shops to the e-market, and that businesses need to make better use of the opportunities such a market offers, including the potential of adaptive e-advertising, and ensure they are making the most of this trend. Such a response additionally confirms that these participants are an excellent target audience for the research in this thesis.
52 Figure 4.2 Purpose of visiting the Internet
When questioned specifically about shopping online, the majority of the respondents (54.35%) indicated that they sometimes shopped online. Moreover, 37.68% of the respondents indicated that they shopped offline, but would look up the items online first (see Figure 4.3). Both types of respondents are clear targets for adaptive advertising. Only 29.71% stated that they normally shop in offline shops, although the number of people declaring they normally shop online is low (13.04%), showing that most prefer a mixed approach.
Figure 4.3 Responses to shopping online
Considering that a large amount of research shows that online advertisements have negative connotations [12, 97, 101, 142], it is surprising that a majority of respondents (52.9%) found
53 advertising on the Internet useful on occasion, while 5.8% even indicated that advertising was always useful (Figure 4.4). This contrasts with the 26.81% who said that e-advertising was never useful. Explicit negative comments about Internet advertising provided by respondents included: “I find advertising annoying”, “I generally ignore it”, “It's just a load of rubbish”, and “Often deceiving”. These outcomes partially support hypothesis H1, as 79.71% (sometimes + never) do not feel that the advertising they are exposed to is useful very often.
Figure 4.4 Responses to whether advertising exposed to was useful or not
When asked if the advertising they were exposed to was adapted to their user preferences, 53.62% indicated that this happened sometimes, while 9.42% and 36.96% indicated mostly yes and mostly no respectively (Figure 4.5). From the open text responses, Amazon, Facebook and YouTube were mentioned by name as websites that did adapt advertising to user preferences. These answers provide support for hypothesis H3, in that social networks are a good source for user behaviour extraction.
54 Figure 4.5 Responses to adaptation of advertising to user preferences
When asked if the advertising adapted to any other characteristics, 63.24% said that sometimes it did adapt to their behaviour, product history and websites visited, while 31.62% said mostly no (Figure 4.6).
55 Of the respondents asked, 44.2% claimed that the social networks they were using did not provide useful advertising the majority of the time, but 43.48% also stated that on occasion the reverse is true. Social networks mentioned included Facebook, YouTube and Twitter. When answering a different question, the majority of respondents (73.19%) also indicated that social networks did not provide them with advertising adapted to their characteristics. From this statement one see that, although social networks can provide useful advertising, most of the time it is not adapted to user specific characteristics.
From the open user comments, three key categories of issues were recorded, regarding what attracted respondents to advertisements online. These included the following.
Design issues, including use of bright colours, large fonts, simple messages or short display time.
Relevance of the advertising message, i.e. is the advertisement relevant to the user or advertising a service the user is interested in?
Price, i.e. is there an on-going sale or possible discount available?
From user comments that mainly fell into one or more of the above three categories, hypothesis H1 can be supported, that users are more likely to accept adaptive advertising, which is suitable for their characteristics and environment. If the information is more relevant to their characteristics, either by design or relevant advertisement or price, the user is more likely to accept such an advertisement.
When asked about the display of advertising content, for instance, whether reasonable media was used according to bandwidth, 24.64% said mostly yes, 46.38% said sometimes and 28.99% said mostly no (see Figure 4.7). These findings suggest that alternative methods of adapting advertisements in circumstances of differing bandwidth require further exploration.
56 Figure 4.7 Use of reasonable media according to bandwidth
From the responses regarding whether advertising used a reasonable screen outline for the user devices, 40.58% said mostly no and only 14.49% said mostly yes (see Figure 4.8).
Figure 4.8 Use of reasonable screen outline for device