Keywords

1 Introduction

Internet is offering new forms of business to companies, it can be used as a communication tool, to engage the customers and to sell products electronically. Through this tool, companies can spread information, seize opportunities and take advantage over their competitors [1]. The fast growing of E-commerce has created a new way to do transactions globally, because it breaks the geographical barriers and allow the users to get more customers with no great efforts [2]. For companies, electronic commerce is an easy tool to get information about trading, just using website [3].

When we analyzed metrics research sources and other aspects of E-commerce, you can find that, in the present, there are some tools to test usability of transactional websites. Although they are not metrics, they can determinate how they these applications work.

The lack of usability in E-commerce portals such: design, how easy to use and understand they are, the direct influence on users, because they find the portal less “friendly”. They do not feel comfortable using the website, so they decided not to use it. It turns into a technological failure, that has become an important market and a financial lost [3]. Due to these new demands, companies have decided to measure the usability level their commercial websites have. The measurement is achieved through techniques that allow an important feedback for companies about the design and the usability standards and interaction with users in the commercial web [4].

For all the reasons mentioned, we propose the elaboration of a group of metrics that can obtain numeric results that allow comparisons between different aspects of usability of E-commerce, starting with the elaborated and current proposals.

2 Methodology

In this section, we are presenting the description of the systematic review, the choice of the most relevant articles and the reviews and interviews done with the specialists for the proposal validation.

2.1 Systematic Review

We developed the systematic review of our research sources from previous work [5], which we are using the usability metrics present in the literature. Nevertheless, these metrics are very generic and oriented to general topics, so they can not be used in every domain or software categories, that is the reason why they have to be analyzed and we had to separate those that may be useful for the topic that is being discussed [6]. In addition, the most relevant articles are going to be taken as a source to evaluate the aspects of usability for E-commerce websites.

After analyzing the relevant documents [7,8,9,10,11], we are able of determinate twenty five usability aspects for E-commerce websites which have being compared against the identifying aspect with the existing metrics to know which one of the others is not been taken into account. The results are showed on the Table 1.

Table 1. Comparison of usability aspects for transactional websites.
Full size table

2.2 Selection

After analyzing the most relevant articles and information of metrics and usability aspects for transactional websites, we made a selection of metrics and the comparison between identifying usability aspects in existing metric and those found in literature.

To proceed with the selection of usability metrics we followed these selection and discard criteria:

  1. 1.

    Domain-oriented: We selected those metrics that let us to evaluate the applications of the E-commerce domain, because it is what this article is orientated to.

  2. 2.

    Metric goal: We selected those metrics with an objective according the website. For example metrics about functionality, errors and others.

  3. 3.

    Target: We select those, which were orientated to the user interaction with software, not the ones focused in developing.

In addition, we discarded the metrics that were too technical, because it is hard to analyze them in a simple way, when the user interacts with the website. For these specific cases, we could use some kind of software that helps to get results with this kind of metrics, for example, how fast the pointer moves, number of pixels used, and others. Likewise, we discard those metrics we cannot apply to the E-commerce, or those that have a similar measuring purpose, or if they are the same as the existing metrics already selected. The results show 25 different usability metrics total that can be used for the proposal.

To compare the usability aspects we found, we identified the aspects being covered with the existing metrics. This way, when compared with other aspects found in literature, we would be able to know which aspects are still not covered, and it would be useful when the new list of usability metrics is proposed.

2.3 Surveys and Interviews

After identifying all metrics and usability aspects, we did surveys and interviewed specialists in the usability subject, so they could review and validate our finds in the literature, and give us their opinion about the matter, so we could notice our possible mistakes and try to improve them. In addition, we did surveys and interviews after we proposed the usability metrics. Since this way our proposal could be validated and know if we directed the proposal to the right field, if they included the usability aspects that allowed to evaluate the transactional websites, and if there was some kind of mistake with the metrics, so we would be able to refine them, and present a completely right proposal.

The surveys and interviews were structured in such a way we could show the specialists the information we found in a clear precise and tidy way, so their opinions and validations would be useful to develop our research.

3 Proposal of Usability Metrics for E-commerce

With the metrics and usability aspects found in literature, we got a group of aspects not considered yet. With this information, we developed a first table with the proposal of usability metrics grouped by aspect. As a result, we created 49 metrics total. We put the metrics on a template that contents the features showed on the Table 2, with an example. We did this way to show them in a clear and tidy way. The results of the consolidated proposal are showed on Table 3 that shows 74 usability metrics, among the existing ones, and the ones created for E-commerce websites.

Table 2. Definition of the metric: “Product Information”
Full size table
Table 3. Consolidated list of usability metrics
Full size table

4 Validation of the List of Usability Metrics

This proposal was validated carefully with opinions from five usability experts. Each of these were shown and explained to them so they can give us an opinion and recommendations. All this process was developed based on interviews and questionnaires, where each expert evaluated each usability metric and rated them according to the importance they believed was feasible.

The opinion of the experts was diverse, for example, one of them told us that although the metrics and aspects of usability proposed served to evaluate a transactional website, they prefer to witness what results were obtained when evaluating a website within a real scenario. Another expert mentioned the proposed metrics covered very important aspects within the E-commerce websites, however, there could be more important aspects for a transactional website. In addition, a pattern should be developed to help to automate the evaluation of them, as it would be very useful for other people who would like to evaluate a transactional website using the metrics.

All these opinions served to revisit the proposal of metrics and to refine them. In the same way, we searched for more scientific articles related to usability aspects for E-commerce to cover all helpful information, for more accurate and reliable results when evaluating a transactional website.

5 Conclusion and Future Works

Usability is one of the most important aspects to take into account in e-commerce websites, since it is focused on the ease of use of the website in the face of interaction with users. Although companies use different evaluation methods to obtain usability, they are qualitative and do not provide numerical results to indicate the level of usability they possess. In this sense, a proposal of usability metrics was developed, which unites different perspectives of usability aspects to be evaluated in a website. Because this project is focused on metrics for transactional websites, a thorough analysis was made of the current metrics in the literature and the aspects they covered, in order to focus on those that were not taken into account by transactional websites and because next, there were no usability metrics. This analysis evidenced the relevant aspects that were not taken into account by transactional websites, for example the search by information, the purchase decision making, the post-sale behavior, and others.

As future work, we have to test the proposal within a real scenario, putting forward a method of evaluating usability metrics, evaluating different websites, finding their level of usability and comparing the results, going through a validation of specialists and thus provide the most reliable results possible.