As it has been highlighted in several studies (Donohoe & Needham,
2009; Hsu & Sandford, 2007; Keeney et al., 2011) the selection of the
panellists plays a central role in Delphi technique as the results of the
study will be opinions of the participants. It is also noted that there is
no single right way for selecting the panellists and several terms can be
used to describe them. In this study the selection criteria for panellists
were not very tight but the channels that were used to induce product
developers were chosen in a way that they would reach people who
had experience of the area and/or topic/context. The number of panellist
was not decided beforehand because it was desirable to include all the interested
customers and to involve them in the process to gain as many
perspectives as possible and increase the chances of getting a heterogeneous
panel (see Turoff, 2006). The level of the expertise of the panellists
varied slightly as some of them had more experience of the area (visited
it more often) and some of them had more experience of nature-based
tourism and outdoor activities. This heterogeneity of the respondents
may benefit the results as it can be expected that different opinions and
ideas will emerge. Evaluating the expertise of the panellists by following
the requirements set by Adler and Ziglio (1996) it can be argued that the
panellist had experience and knowledge of the issues under examination.
The panellists registered themselves voluntarily as product developers
which indicate that the participants were willing, and had the time and
capacity to participate the process. However, during the selection process
the communication skills of the panellist could not be evaluated.
The Delphi technique requires constant data analysis throughout the
process. Hence the data analysis requires expertise and time from the
researcher. Several Delphi rounds also produce a lot of data that needs
to be carefully processed so that it enables the following rounds to be
conducted and in the end the results should be presented in such a
way that they contribute to the decision making processes.
5.4. Implementation in practice
As there are several issues that need to be consideredwhen a Delphi
study is conducted, the technique is labour intensive and it requires personnel/a researcher familiar with all the procedures needed. The
costs of conducting a Delphi may be considered quite reasonable
when online channels or tools are used to manage the process.However
multiple Delphi rounds make the process longer and require time from
the researcher. The costs of implementing a Delphi study may then be
higher. Because of these facts it might be that many tourism service providers
do not have sufficient resources such as know-how and the time
required to use the Delphi in their NSD.
5.5. Type of customer information — context and situation