Setting the fundamentals to measure the effects of the ATP online marketing campaign

The first campaign of the partnership was launched on 15 December 2016 on the ADDMA Facebook account and was shared by the other two partners. The reactions to the post can be visualised to give a sense of how the triangular relationship can be shaped and used to disseminate information through social networks (Vlassi and Papatheodorou, 2018). Node XL software was used to download and visualise the data. Nodes represent the users that reacted to the posts, and lines depict the reaction (like, share, comment, etc.). The number of reactions totalled 4,510, and almost 60 per cent of them were shares. Greatdiffusion was observed, because only the shares of Aegean Airlines and Athens International Airport concentrated an important number of reactions that is worth mentioning (about one hundred each).

Further to a Memorandum of Collaboration signed between ADDMA and the University of the Aegean, the participatory observation started in early 2017 by first identifying how the three partners perceived success and what metrics should be used to evaluate their marketing action outcomes. On the basis of the available literature (Morgan et al., 2012); several conversations and unstructured interviews with representatives of each partner separately; and group discussions, five key performance indicators emerged as important in the minds of the partners: destination awareness, emotional proximity, intention to visit, incremental spending and ROI. A questionnaire was subsequently built by the authors (as representatives of the University of the Aegean) in cooperation with the ADDMA and its media-planning agency in 2018 to collect the appropriate data. The campaigns were disseminated over the internet by using AdWords and banners on social networks, among other strategies. Users exposed to the campaign were retargeted and asked to fill in the survey questions. Moreover, thanks to the nature of the internet and the possibilities offered by the media-planning agency, data about exposed users’previous online behaviour also became available. For example, the researchers are in a position to know whether users were interested in visiting Athens before the users saw the ad (by observing their keywords searches in Google) or whether the users had already booked a city break to Athens.

Increasing brand awareness can result in the inclusion of a brand in the consumer choice set and can also have positive effects on consumer final decision (Keller, 1993). To measure destination awareness, an unprompted question was placed in the questionnaire. Users were asked to mention a city-break destination that came to their minds. Moreover, a prompt question followed to detect the city-break destinations that already existed in users’ minds from a given sample of 12 cities. Respondents who changed perceptions and feelings about Athens were also examined by the survey. Participants were asked to state the extent to which they agreed or disagreed with 11 statements that intend to detect their emotional proximity to Athens. Intention to visit measured whether participants were interested in visiting Athens in the following 12 months. A question was included in the survey to detect whether visitors had already booked a holiday to Athens or whether they were planning to book one in the following 12 months.

Finally, to measure incremental spending and then ROI, a question was placed to detect the campaign’s influence. In other words, to detect whether the respondents’ actions were an outcome of the exposure to the campaign or whether these actions would have been undertaken anyway, the question ‘Thinking about the ad, how would you describe its influence on your deci-si on/plans to visit Athens?’ was set with alternative answers: ‘Would definitely have visited/planned to visit anyway’, ‘Would probably have visited/planned to visit anyway’, ‘Would have visited but for fewer nights’, ‘Would have visited but

Assessing online marketing campaigns 237 at a later date’, ‘Would probably not have visited/planned to visit otherwise’ and Would definitely not have visited/planned to visit otherwise’. Finally, visitor expenditure per night was estimated by asking respondents who had already taken a trip to Athens to state the amount of money they had spent and potential visitors to state the amount they intended to spend.