The Effect of Website Design Dimensions on Initial Trust: A Synthesis of the Empirical Literature

Abstract

ABSTRACT This paper aims to provide an integrative review of the experiment-based literature on the antecedents of initial trust in a business-to-consumer (B2C) e-commerce setting. To that end, we present a framework that classifies trust-inducing website features according to three broad dimensions, namely visual design, social cue design and content design, because comparing studies with different empirical set-ups requires conceptual clarity. To synthesize the literature we use an advanced vote-count procedure combined with a sign test. We find that the literature provides sound empirical support for our general hypothesis that web design cues effectively enhance consumers' initial trust towards unfamiliar online vendors. E-tailers should thus consider embedding human-like cues (i.e., facial photos, video streams) into their interfaces, as well as integrating assistive web applications (i.e., avatars, recommendation agents). Interestingly, we also find that internally provided e-assurance structures (such as privacy/security policies and vendor-specific guarantees) can be as effective as paid e-assurance mechanisms (such as third-party trust endorsements). Our overview also reveals that the effectiveness of certain trust-signalling features within the visual and social cue design dimensions is still under-researched. The support for the positive effect of such website atmospheric cues is therefore still weak. Keywords: initial online trust, website design, B2C e-commerce, literature review, vote-count method (ProQuest: ... denotes formula omitted.) 1. Introduction The goal of this paper is to provide a selective but systematic review of the empirical literature on initial trust in a business-to-consumer (B2C) e-commerce environment. In particular, we synthesize the experiment-based studies that investigate the effects of the e-tail interface on consumers' trust towards unfamiliar vendors. The choice of this specific topic is motivated by the need for a better understanding of how e-vendors can induce consumer trust by adequately choosing the characteristics of their interfaces. A lack of trust towards Web vendors is one of the main factors deterring consumers to engage in online shopping [Beatty et al. 2011]. Investigating the antecedents of initial trust is particularly important because in interactions where parties are unknown to each other and the internet is the channel of communication, initial trust determines whether or not a transaction will occur [Wakefield et al. 2004]. Thus, the ability to evoke initial online trust can determine the success or failure of an online retailer. There are already a few meta-studies in this domain that have summarized large bodies of evidence. Beatty et al. [2011] and Jourdan and Ingram [2011] offer a valuable discussion of antecedents of online trust such as usefulness, ease of use, and risk. However, a literature review that focuses on trust-inducing website design features is still missing. This holds a fortiori for such dimensions as visual, social cue and content design features. Moreover, empirical studies on these features often reach contradictory conclusions. For example, while some authors find a positive influence of institutional cues on consumer initial trust formation [e.g., Hu et al. 2010], others do not support this claim [e.g., Bahmanziari et al. 2009]. The main goal of this paper is therefore to identify what elements of website design play a crucial role in initial online trust formation, and to discuss the reasons for the discrepancies observed in the empirical research. We thus put forward the following research questions: What kinds of Web interface applications can be effective in stimulating initial online trust?, and also: What areas remain to be investigated further? Towards this end, we first briefly explain the concept of initial online trust, and how it differs from online trust in general (Section 2). …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 150 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.