Can a computer be sincere when it says "Have a nice day"?
Philip Calvert, Victoria University of Wellington, New Zealand
Whenever we go into a retail store to buy groceries, when we call a plumber to fix a leak, and when we talk to an insurance agent about our policies, we are probably quite aware of the service quality we expect from these providers. In-person service delivery, whether face to face, or over the telephone, is well enough engrained into our lives that we know what level of service we expect, and we can then go on to say what kind of service we think we receive – with the two sometimes being quite a distance apart.
Service quality in the digital environment
The emergence of service providers on the Web has launched the whole subject of e-commerce, yet while much attention has been paid to service process there is still not a great deal known about service quality in the digital environment. Although the services we look for might be the same – groceries and insurance can be bought and plumbers booked on the Web – the point of contact is sufficiently different that our, by now, established ways of evaluating service quality might no longer be adequate. For one thing there often isn't a person on the other side of our digital service experience, and that can be a little disconcerting. When a shop assistant asks: "Is there anything else I can do for you?" we might be able to judge the level of sincerity in the voice and body language, but if a computer system asks the same question then obviously there isn't any actual sincerity in it at all.
Most current ideas about service emanate from disconfirmation theory: if we know what we expect from the service and if our experience matches our expectations, then the expectations are confirmed; but if the experience falls short, then our expectations are disconfirmed. This is also known as gap theory, because we examine the gaps between expectations and our perceptions of the actual service.
Although there are many writers who contributed to the development of our current knowledge of service quality, none are better known that Zeithaml, Parasuraman and Berry, who will long be remembered for their SERVQUAL instrument developed in the late 1980s. After extensive research they decided that service quality was itself a construction of ten dimensions:
- and understanding the customer.
These were later refined by Zeithaml et al. to just five:
- assurance, and
In SERVQUAL they created a list of service encounter characteristics, each lying within one of the five dimensions, and respondents are asked to check off on a Likert scale how much each statement matches their expectations of an excellent service and how well they think an actual service is performing in practice.The aggregated results give the gap between expectations and performance on each one of the statements.
SERVQUAL has been used in libraries with some success, but the best results have come from adapted instruments using statements describing service encounters more specific to a customer's experience of libraries, and in this field the leaders have been Peter Hernon and Danuta Nitecki. Recently the Association of Research Libraries adopted work done at Texas A&M University Library to develop the Libqual+ instrument that is now used extensively in North America.
So far, service quality assessment using SERVQUAL and its successors has worked very well for what we might call traditional service encounters, but SERVQUAL was never designed to cater for the purely digital world, and whether it can cope with virtual service remains a moot point. Most electronic content is mediated with clients through a website, and it doesn't appear that the Web is a natural environment for service encounters – it was never designed to be so. When you buy groceries from a typical store you can chat with the people who are serving you and ask any questions about the fruit and vegetables that might come to mind. But when you buy groceries from an online vendor you have no idea who, or even what, is providing your service. As they used to say, on the Net no one knows you're a dog.
New dimensions of e-service quality
A few writers have tried to produce new dimensions of e-service quality, and none are better known that Zeithaml and Parasuraman, who together with Malhotra produced one of the best early passes (2000) at this new problem. Their model includes 11 dimensions:
- ease of navigation,
- site aesthetics, and
- price knowledge.
It seems like a long list, but it is worth remembering that the SERVQUAL started off with ten dimensions, to be reduced to five later on. There are some similarities with SERVQUAL, such as site aesthetics being quite similar to the tangibles dimension, and the two key dimensions of reliability and responsiveness occur in both lists. What the authors of the e-service dimensions have done this time around is merge the 11 units into just two higher-level abstractions, as they have called them: perceived convenience, and perceived control. This is based on a belief that it is convenience and control that, for the majority of customers, constitute e-service quality.
They have then introduced another step to say that convenience and control constitute e-service quality, and when this is combined with perceived price the result is perceived value, and when that is high the customer will make purchases and those all important repurchases (or repeat visits). This is a very useful starting-point for further thinking about e-service quality.
"Incubative dimension" and the "active dimension"
Another view on this topic has been presented by Santos (2003). In her model there is a divide between what she calls the "incubative dimension" that covers actions that take place even before the service is launched, and the "active dimension" which encompasses the ongoing activities of customer service.
Within the incubative dimension she lists:
- ease of use,
- structure and layout, and
In the active dimension her determinants are:
- security, and
If there are similarities between her list and the model presented by Zeithaml, Parasuraman, and Malhotra, it shouldn't be too surprising, because they are trying to describe the same thing!
E-service quality in libraries
Lastly, what of e-service quality in libraries? It is possible that one of the new models described above could work for libraries, or perhaps we shall see a blending of some of the existing models. It may be, though, that specialized services such as virtual reference will mean customers place a greater emphasis on accuracy as a factor. There will also have to be a way of providing good service for the hit-and-run customer who wants to visit the virtual service, see if it has a copy of a relevant document, e.g. an e-journal article, print it out, then leave.
For these people – often part-time students on very limited time – any delay in providing the service has a major impact. Delays might come in forms such as:
- being denied access because there is already the maximum number of users for the library's licence,
- being denied access by server downtime,
- very slow access speeds,
- incompatible file formats, and so on.
All these elements will become part of e-service for the library customer.
This is a relatively new topic yet it is clearly so important that there will be much more written and said about it in the near future. Watch it grow.
References and further reading
Santos, J. (2003), "E-service quality: a model of virtual service quality dimensions", Managing Service Quality, Vol. 13 No. 3, pp. 233-246.
Zeithaml, V.A., Parasuraman, A. and Malhotra, A. (2000), A Conceptual Framework for Understanding E-Service Quality: Implications for Future Research and Managerial Practice, Marketing Science Institute working papers, report number 00-115.