Social media sites, like TripAdvisor, allow postings of User Generated Content (UGC) that provide restaurant customers with the opportunity of sharing their experiences and recording their satisfaction and dissatisfaction. These activities represent an essential form of electronic word-of-mouth that can influences the decision process of other customers. The paper investigates tourist satisfaction experiences as expressed in UGC on TripAdvisor to determine the role that wine plays in customer satisfaction with fine dining restaurants. A list of the world's top 1000 restaurants is used to identify five top- and five bottom-ranked restaurants and data in the form of six UGC narratives written by tourists on TripAdvisor for each restaurant are collected. Lexical analysis of the sixty narratives collected is undertaken via the Leximancer software. Results are reported, implications are discussed, limitations are noted, and directions for future research are indicated.

_{Higher levels of competition in online channels and the resulting information overload among customers is becoming an issue of increasing concern among marketers who seek to provide persuasive content. Past research has revealed that storytelling is an effective tool to achieve persuasion. Research has also suggested that consumers are essentially storytellers who narrate their experiences in the form of stories. Several researchers have looked at consumers’ stories and investigated the effect that these stories have on their receivers.}

 _{In the last decade the research about storytelling has shifted to the online world. There is a board agreement amongst researchers that this is indeed very persuasive when compared to other forms of promotional online content. Although online content can possess various characteristics, textual content is the dominant type of content employed by both consumers and sellers/producers. This thesis asks:}

 _{What effect do characteristics of online textual content have on persuading consumers?}

 _{This research question is grounded in Narrative Paradigm Theory that is used to present a conceptual model that allows for the development of more specific research questions dealing with characteristics of online textual content. To do this, a distinction is first drawn between user generated content (UGC) and producer generated content (PGC). In both cases, textual content can be story- or argument-based and can possess various other characteristics, that can be basic (valence and word count), stylistic (story-/ argument-based) and semantic (use of meaning in words). The thesis seeks to investigate what effect these characteristics have on the persuasion process of customers viewing online textual content. It seeks to understand what characteristics make textual content, (whether UGC or PGC) believable and influential and whether the knowledge learnt from UGC can be used by marketers to create persuasive content. A further aim of the research, is to identify an analytical tool that can help content creators identify the persuasive potential of any given text.}

_{The main research question and model gave rise to the following five sub-research questions:}

_{RQ1: What are the themes and concepts used by reviewers expressing differing customer satisfaction when posting UGC?}

_{RQ 2: Does the narrative force resulting from reviews posted as UGC differ by declared customer satisfaction ratings?}

_{RQ3: What is the impact of different content type and length of online textual reviews on narrative believability and purchase intention?}

_{RQ4: Does objective knowledge effect the impact of different content type of online textual reviews on narrative believability and purchase intention?}

_{RQ5: Does story-based textual PGC improve website stickiness?}

 _{To answer these research questions, four empirical studies are undertaken. Study one analyses the first two research questions whilst the remaining three studies each analyse a further research question. Different research methods for collecting and analysing data were used to address the research questions. Using different research methods is regarded advantageous because it allows for methodological rigorousness. The first study employs the qualitative analytical tools using the Diction and Leximancer software while the three other studies make us of an experimental approach. Experimental designs are preferred when the goal of the research is theory application. This doctoral thesis is presented in a monograph format comprising five chapters: Introduction, Literature review, Methodology, Empirical studies, and Conclusion. The latter indicates how the research contributes to the body of knowledge by providing multiple theoretical and managerial implications.}

PurposeInvestigates the extent to which wine tour-operating companies are successfully using their websites to communicate persuasive and meaningful differences to customers in their quest to clearly position their product offering.

Design/methodology/approachText data are collected from 250 websites that promote wine tourism in five different countries' winescapes. Lexical, regression and hierarchical clustering analyses are used.

FindingsLexical analysis using DICTION can distinguish among websites from different countries' winescapes and French wines obtain the best mean review scores from customers while US, Napa Valley, websites obtain the lowest scores. DICTION dimensions allow for meaningful clusters and can also predict TripAdvisor’s mean review scores.

Practical implicationsThere is a need to pay better attention in the development of website content and the critical role that both syntax and semantics can play in facilitating the use of a firm’s website, specifically in terms of clear positioning.

Originality/valueUses lexical analyses of website narratives to understanding current positioning of firms.