ONLINE SHOPPING AGENTS: AN EXPLORATORY STUDY OF

USERS’ PERCEPTIONS OF SERVICE QUALITY

Călin Gurău

GSCM – Montpellier Business School, 2300 Avenue des Moulins, 34185 Montpellier, France

Keywords: Shopping agents, users' perception, online service quality.

Abstract: The use of online shopping agents has increased dramatically in the last 10 years, as a result of e-commerce

development. Despite the importance of these online applications, very few studies attempted to identify

and analyse the main factors that influence the users’ perception regarding the service quality of online

shopping agents, and consequently, the elements that determine the users’ choice of online shopping agents.

The present study attempts to fill this literature gap, identifying on the basis of primary data analysis, the

various circumstantial or personal factors that can determine the choice of a specific searching strategy and

shopping agent.

1 INTRODUCTION

The sheer amount of information available on the

Internet has introduced new challenges for

companies and buyers. In the information-rich

online environment, the firms must increase the

distinctiveness of their offer, in order to make it

visible to their targeted consumer segments. The

buyers also have particular problems: although they

might know exactly the technical specifications and

the type of the desired product, how can they

quickly identify the best offer in terms of quality-

price? In many cases the amount and the variety of

information available to customers create a situation

of information overload.

Specific online applications have been created to

solve this problem: search engines and intelligent

agents. The use of these applications raises

important questions related to their quality and

reliability. The present study attempts to investigate

the users’ perceptions about the quality of shopping

agents, answering to the following research

questions:

1. What are the characteristics that define the

quality of shopping agents in users’ mind?

2. Are there any differences determined by the

customer profile in the usage of shopping agents?

3. What are the transaction-related factors that

influence the perceived quality regarding shopping

agents?

2 SHOPPING AGENTS:

A LITERATURE REVIEW

Shopping agents can be defined as specialised

software applications which help users to search

specific types of products or services offered online,

and to collect additional information about these

offers. After the users enters a query into the

shopping agent, the software visits various e-shops

or online merchant sites, collecting information

about the product or service demanded (Rowley,

2002). The search is usually done on the basis of

keywords or product categories, which are listed on

the shopping agent’s web page (Rowley, 2000a).

The use of shopping agents to search for the lowest

possible price offer will increase the pressure for

vendors to reduce the selling price, aligning it to the

smallest existing level (Iyer and Pazgal, 2003).

However, several studies showed that the differences

in the online prices remain significant (Iyer and

Pazgal, 2003, Sen et al, 2006). This can be explained

by the choice of some vendors to focus on quality

and personalisation rather than price, in order to

increase the loyalty of their customers. Other online

vendors decide to restrict the access of shopping

agents to their offers (Iyer and Pazgal, 2003) in

order to reduce the visibility of their price levels.

On the other hand, the services offered by shopping

agents to customers significantly reduce their time

and effort to search and compare online information.

Some authors have developed mathematical models

that attempt to predict the shopping behaviour of

207

Gur

au C. (2008).

ONLINE SHOPPING AGENTS: AN EXPLORATORY STUDY OF USERS’ PERCEPTIONS OF SERVICE QUALITY.

In Proceedings of the Tenth International Conference on Enterprise Information Systems - SAIC, pages 207-210

DOI: 10.5220/0001681402070210

 SciTePress

online customers when using shopping agents (Iyer

and Pazgal, 2003, Sen et al, 2006). Rowley (Rowley,

2000b) developed a model which presents the stages

of product/service search using a shopping agent. The

process is sequential and logical, but does not provide

any clues regarding the attitudes/behaviour of

consumers during the search process. In terms of e-

marketing, the factors that influence the consumers’

perceptions about the quality of shopping agents will

determine their choice and use of a specific shopping

agent, and finally, their level of satisfaction. A

somehow different model of consumer behaviour is

presented by Guttman et al. (Guttman et al., 1998)

which comprises six successive stages: (1) need

identification, (2) product brokering, (3) merchant

brokering, (4) negotiation, (5) purchase and delivery,

and (6) service and evaluation. Their analysis

indicates that online shopping agents are particularly

useful for the stages of product brokering, merchant

brokering, and, sometimes, negotiation.

It is also logical to consider various elements that can

influence the use of shopping agents and the

satisfaction derived from their services. Sproule and

Archer (Sproule and Archer, 2000) indicate a series of

factors that determine a change in the type of

information accessed by users and in their searching

strategies, such as the purchasing situation (new

versus frequent purchase) and the level of perceived

risk (high versus low risk). Cooke et al. (Cooke et al.,

2002) indicate that the online search strategy will be

different for familiar and for unfamiliar products,

focusing in their paper on the online marketing of

unfamiliar products.

3 RESEARCH METHODOLOGY

In order to answer to the three research questions

presented in Introduction, both secondary and

primary data have been accessed and analysed. In

the first stage of the research process, a series of

texts published online or in academic journals or

books have been used in order to obtain a

comprehensive overview of the state of research in

the area of shopping agents. In the second stage of

data collection process, 200 people have been

approached in Internet cafes located in London,

Southampton and Edinburgh, UK, and invited to

participate in a 20 minute face-to-face interview,

regarding their use of shopping agents and their

perception about the quality of shopping agents. 146

people accepted to take part in the survey, but only

121 had a direct experience of using shopping

agents, and therefore, only the data provided by

these people was retained and used for analysis. The

interviews have been recorded and then transcribed

in order to facilitate their analysis. During the

interviews, the respondents were asked to:

- describe their general experience in using

shopping agents;

- provide a series of elements that, in their

opinion, determines the users’ perception about the

quality of shopping agents;

- discuss the elements that could influence the users’

perceptions about the quality of shopping agents;

- indicate their usual searching strategy and the

type of shopping agent they prefer.

Data about the profile of respondents was also

collected, such as: gender, age, frequency of Internet

usage, frequency of e-shopping and frequency of

shopping agents’ usage. The transcribed interviews

were than analysed manually, the identified patters of

answers being codified and introduced into the SPSS

software for frequency and cross-tabulation analysis.

4 USERS’ PERCEPTIONS

REGARDING THE QUALITY

OF SHOPPING AGENTS

4.1 The Profile of Respondents

From 121 respondents, 79 (65.3%) were men and 42

(34.7%) women. Most of respondents were young,

85 (70.2%) having between 18 and 25 years old, 27

(22.3%) were aged between 26 and 40, and the

remaining 9 (7.5%) of respondents between 41 and

60. 66 respondents reported a high frequency of e-

shopping (at least once a week), 38 indicated a

medium frequency (at least once a month) and the

remaining 17 a low e-shopping frequency (less than

once a month). In terms of shopping agents usage,

42 respondents indicate that they use this service

regularly, 67 occasionally, and 12 rarely.

4.2 The Characteristics that Define the

Quality of Shopping Agents

The respondents have indicated during the

interviews the perceived characteristics that define,

for them, the quality of shopping agents. The

capacity of the shopping agent to find the lower

price offer is considered important by 81% of

respondents. These are probably the ‘bargain-

oriented shoppers’, which use the Internet primarily

of the opportunity to find competitive price offers.

Ease of interaction is outlined by 80.2% of

respondents as an important quality element of the

shopping agent. This characteristic is a rather

complex concepts which includes: the reliability of

shopping agent website, the ease of navigation

within the site, the clarity of various

ICEIS 2008 - International Conference on Enterprise Information Systems

208

information/indications provided on the website, the

tools available on the web site (search, select from a

list, order the results in relation to various headings).

76.8% evaluate as important the precision of the

shopping agent, defined as its capacity to find the

required product. 70.2% of respondents expect to

receive additional information about the offer and/or

about the vendor, either in the form of short reports,

or as links, that can then be explored by the user.

Finally, 58.7% of respondents like the shopping

agents that, beside the online offers, provide also

access to product rankings or product reports written

by previous customers. This service requires the

existence of a specific software function which has

to be actively implemented by the shopping agent.

4.3 The Differences determined by the

Customer Profile in the Perception

of Shopping Agents’ Quality

Considering the relative and subjective nature of

perceiving services quality, it can be expected that

the characteristics defining the quality of online

shopping agents will vary in relation to the personal

profile of respondents. In order to identify this

variation, the four elements which have been

selected to define the profile of the respondents in

this study: (1) gender, (2) age, (3) frequency of e-

shopping and (4) frequency of shopping agents

usage have been cross-tabulated with the quality

characteristics indicated by respondents. The results

indicate the influence of respondents’ gender on the

perceived quality characteristics of online shopping

agents. Male respondents seem to consider more

than the female respondents that the precision of the

shopping agent (in terms of finding the right product

and the lower price), represents an important quality

element. On the other hand, a large proportion of

women respondents, indicated as important quality

elements the provision of additional information

about the offer and/or vendor (92.5%), the ease of

interaction (97.5%) and the access to previous

clients’ ranking and/or recommendations. These

differences can be explained by the differences in

the shopping behaviour of men and women. Men

seem to be more focused on the efficiency of the

search process, while women like to have access to

additional data in order to make an informed choice.

The age of respondents is also determining a

variation in the respondents’ perception of the

defined quality characteristics. The younger

respondents are focused on the precision of shopping

agents (both in terms of price and specific product),

while the older respondents appreciate positively the

capacity of the search engine to provide additional

information and ease of interaction. At interesting

trend is the progression (or regression) of

percentages from one to another group age. The

regression is significant in what concern the capacity

to find the lower price offer (from 91.8% to

respondents aged between 18 and 25 years old, to

22.2% of respondents aged between 41 to 60 years

old). Similarly, the progression of percentages is

important in what concerns the capacity of the

shopping agent to provide additional information

about the product and/or vendor, ease of interaction

and the access to previous clients’ rankings and

recommendations. The results can be interpreted by

the need of older users to easily interact with the

software application, and to compare in more detail

the various offers selected by the shopping agent.

The frequent online shoppers seem to appreciate

particularly the capacity of the shopping agent to

find the lower price offer (97% of respondents from

this category). On the other hand, the people that are

using less frequently the online market are interested

in rich information and ease of interaction (the

totally of respondents defined as less frequent

shoppers indicated this element as important), but

also in the capacity of the shopping agent to identify

properly the required product (94.1% of respondents

from this category). These results are logical,

considering the fact that the less frequent shoppers

associate online transactions with high levels of risk,

and therefore require precise and extensive

information about the offer and the vendor.

The trends identified in relation to the e-shopping

frequency are similar with the results based on the

frequency of shopping agent usage. This might

indicate that frequent e-shoppers are also using

frequently online shopping agents, projecting in their

responses their specific behavioural profile.

However, this is not entirely true, because the

frequent users of online shopping agents represent

only 63.6% of the respondents that have indicated a

frequent e-shopping. Again, more research is

necessary to understand why some frequent

shoppers are not also frequent users of online

shopping agents. A possible explanation is their

preference to shop mostly from known and trusted

online vendor, which eliminates the need for

shopping agent services.

4.4 The Transaction Factors

influencing the Perceived Quality of

Shopping Agents

The relative importance of quality factors is different

in relation to various transaction-related elements.

The respondents indicated that the most important

transaction elements influencing the perceived quality

of shopping agents are: level of selling price,

familiarity of with the product, familiarity with the

vendor, length of product use, time available for

ONLINE SHOPPING AGENTS: AN EXPLORATORY STUDY OF USERS’ PERCEPTIONS OF SERVICE QUALITY

209

decision and purchasing. In a situation of a high

selling price the capacity of the shopping agents to

identify the lower price offer is considered important

by the majority of respondents. High percentages are

also provided for the capacity of the shopping agent to

select the required product, to provide additional

information about offer and/or vendor, as well as the

rankings and comments of previous customers. These

results can be explained by the high transaction risk

that is associated with a high product price.

Gradually, as the price level is medium and low, the

need for additional information becomes less

important. The need for a precise selection of the

lower price offer and of the required products is lower

when the price is small, but the percentage of

respondents that consider these factors as important

still remains relatively high. The proportion of people

that indicated the ease of interaction as important

remains quite stable in all three price situations.

As expected, a low familiarity of the customer with

the product determines a high request for

information related to the offer, vendor and previous

customers’ experience. Inversely, the shoppers

familiar with the product indicate a low importance

of additional information, focusing their needs on

the capacity of the shopping agent to identify

precisely the best price offer and a specific product.

When the product acquired online will be used a

long time, the respondents consider as particularly

important the capacity to obtain rich information

about the product, vendor and previous customers’

experience. This tendency is determined by the risk

associated with the transaction, taking into account

that, usually, the products with a long life-time are

also relatively expensive. However, the other quality

characteristics are also considered important by a

large proportion of respondents, the profile of results

being quite similar with the data provided in the case

of expensive products. This similarity is also

preserved in the evolution of percentages for the

situation of a product with a medium lifetime and for

a product with a short lifetime. Interestingly, in this

last situation, the shopper is specifically interested to

in the capacity of the shopping agent to provide the

best price offer, indicating that the shopping

decision might be determined by the level of price.

The answers provided are particularly interesting in

the situation when the shopper has a very short time

to find desired product and to conduct the

transaction. In this case, all the five quality

characteristics are indicated as highly important by a

very large proportion of respondents. The

percentages are decreasing progressively when the

customer has more time to conclude the transaction,

which creates better conditions for a thorough

information search and comparison of various

available offers.

5 CONCLUDING REMARKS

Despite the increased importance of shopping agents

for the development of online consumer markets and

transactions, the existing literature has not yet

considered the specific quality characteristics of

shopping agents, as they are perceived by the users,

in relation to their personal profile and the specific

circumstances of various transactions.

The study has a number of limitations, determined

by its exploratory approach. The sample of

respondents is relatively small, and the data

provided was analysed at a superficial level. The

focus of this project was more to understand what

are the perceived quality dimensions of the online

shopping agents, and what are the factors that

influence this perception (personal and

transactional), and not to find out why the

respondents are perceiving a specific way the quality

of e-shopping agents. This empirical dimension

requires further research, in which the statistical

analysis of quantitative data has to be complemented

by a qualitative approach, which can provide a better

understanding of the way in which online customers

interact with shopping agents.

REFERENCES

Rowley, J. (2002). ‘Window’ shopping and browsing

opportunities in cyberspace, Journal of Consumer

Behaviour, 1, 4, 369-378.

Rowley, J. (2000). Product search in e-shopping: a review

and research propositions. Journal of Consumer

Marketing, 17, 1, 20-35.

Iyer, G., & Pazgal, A. (2003). Internet Shopping Agents:

Virtual Co-Location and Competition. Marketing

Science, 22, 1, 85-106.

Sen, R., King, R.C., & Shaw, M.J. (2006). Buyers’ Choice

of Online Search Strategy and Its Managerial

Implications. Journal of Management Information

Systems, 23, 1, 211-238.

Rowley, J. (2000). Product searching with shopping bots.

Internet Research: Electronic Networking

Applications and Policy, 10, 3, 203-214.

Guttman, R.H., Moukas, A.G., & Maes, P. (1998). Agents

as mediators in electronic commerce. International

Journal of Electronic Markets, 8, 1, 22-27.

Sproule, S., & Archer, N. (2000). A buyer behaviour

framework for the development and design of software

agents in e-commerce. Internet Research: Electronic

Networking Applications and Policy, 10, 5, 396-405.

Cooke, A.D.J., Sujan, H., Sujan, M., & Weitz, B.A.

(2002). Marketing the Unfamiliar: The Role of

Context and Item-Specific Information in Electronic

Agent Recommendations. Journal of Marketing

Research, 39, 4, 488-497.

ICEIS 2008 - International Conference on Enterprise Information Systems

210