Shopper Age and the Use of Self-Service Technologies
Purpose – The effect of shopper age on attitudes toward and use of retail self-service technology
(SST) was investigated. The age variable has received relatively little attention in the literature.
Methodology – Questionnaire responses from three age groups are compared. Also, cluster
analysis is used to group subjects based on similarity in attitudes toward and use of SST.
Findings – Compared to younger consumers, older consumers had experience with fewer types of
SSTs, less confidence in using SST, reported missing human interaction to a greater degree, used
self-checkout less often when the option was available, were less willing to pay a premium for
express checkout, and were more likely to attribute a corporate self-interest for the introduction of
SST. For the total sample of 718 subjects, 40% reported using store self-checkout 15% of the time
or less when the option was available. Only 25% of subjects reported using automated store
checkout on more than half of their shopping occasions.
Research limitations – Only eight types of SST were studied and only one technology was
investigated in depth.
Practical implications – Based on the findings of this study, four managerial actions are
recommended that may potentially increase traffic throughput at automated retail checkout.
Originality/value – This is believed to be the first study to find significant differences among age
groups on multiple dependent variables associated with SST. Also, the identification of consumer
clusters based on attitudes toward and use of SST may be novel.
Keywords Age, Self-service technology, Attitudes, Technology adoption
Paper type Research paper
Self-service is growing. A research study conducted by IHL Consulting Group finds that
North American consumers are projected to spend over $525 billion at self-checkout lanes,
ticketing kiosks and other self-service machines in 2007, up from $438 billion in 2006 (IHL
Consulting Group, 2007). For the future, the group predicts an 18 percent increase in 2008 and
estimates the dollar value of self-checkout transactions should reach $1.3 trillion by 2011.
The primary driver behind the rollout of self-service technology (SST) by retailers is a
potential cost savings (Bitner et al., 2002). For example, in a grocery store, one cashier typically
oversees four self-service registers, so the store can serve the same number of patrons with a
reduced head count (Rosen, 2001). Additional benefits of SST include better management of
service demand fluctuation and providing a more consistent service, independent of employee
mood (Weijters et al., 2007). Also, there appears to be a consumer segment that prefers self-
service checkout over interaction with a human employee. The IHL Consulting Group study
previously mentioned found that 18 percent of shoppers use self-checkout “all the time” when
this option is available (Schuman, 2006). Indeed, another reason for the introduction of self-
checkout by retailers may be to better satisfy these customers, perhaps capturing this segment
from less innovative stores as well as retaining existing customers who prefer self-service.
Despite the almost omnipresence of self-service in the retail environment, it has been
noted that there is a relative lack of research on consumer attitudes toward and use of SST
(Weijters et al., 2007). This is especially true regarding the variable of age. Few studies have
examined perceptions toward and use of SST among different age groups. Yet, there is reason to
believe that differences may exist. Many older consumers came of age before SST was
introduced. Using self-service machines requires a behavioral change on the part of these
consumers. Also, it is possible that they may not view automated checkout as a “normal” level
of service. That is, this segment may have come to view full service as the “standard” service,
implying that self-service is something less than standard. If age-group differences exist,
retailers should be aware so they can plan accordingly. As pointed out by Weijters et al. (2007),
demographic variables such as age (as opposed to personality traits) are more readily identifiable
in practice and managerially actionable. For example, a particular age segment could be targeted
with a direct mailing and/or an incentive to use SST. It may be noted that older consumers
control more than 50 percent of discretionary income (Solomon, 2004), making this a segment
that retailers should especially want to attract and satisfy.
The primary goal of the present study is to investigate the effect of age on attitudes toward
and use of retail SST, especially self-service checkout at a grocery or other retail store. A second
goal is segmentation of the general consumer public based on their attitudes toward and use of
SST. A better understanding of how different segments perceive and use SST could help
retailers anticipate usage problems and plan remedies for such difficulties.
Automation of retail transactions is increasingly common. The trend began with the
introduction of the automated teller machine (ATM) several decades ago, and it continues today
with online banking, online purchase of goods from a retailer’s website, self-service checkout at
grocery stores, pay at the pump gasoline sales, in-room hotel checkout, automated airline check-
in and ticketing, and online stock brokerage transactions. The broad term of self-service
technology (SST) has been applied to technological interfaces that enable customers to produce a
service independent of direct service employee involvement (Meuter et al., 2000).
Meuter et al. (2005) notes that the concept of SST requires a paradigm shift by both
marketing practitioners and academics toward the view that separation of production and
consumption is no longer a desirable (or practical) goal. Rather, customers become co-
producers. The role of the customer in the production of service must be defined, just as the
organization defines the role of any employee. The authors suggest that successful SST co-
production requires customers knowing what is expected if them, being motivated to engage in
the behavior, and possessing the necessary knowledge and skills to carry out the behavior.
Adoption and Satisfaction
Walker and Johnson (2006) reviewed the literature to identify factors influencing the
adoption and use of SST. Their list includes: a) personal capacity (self-belief that the user is
capable of using the machine successfully), b) perceived risk (extent to which the machine is
believed to be reliable and personal information is believed to be secure), c) relative advantage
(extent to which SST is believed to be more convenient and faster than a traditional face-to-
face encounter), and d) preference for personal contact (the degree to which the consumer
prefers human interaction over interaction with a machine).
Bitner (2001), drawing from an analysis of 823 critical incident encounters with SST
(Meuter et al., 2000), suggested that the two major factors affecting the success of SST are
reliability (defined as dependability and user friendliness) and advantage (meaning that the
SST either saved time or money or delivered some other customer benefit).
With regard to possible age group differences, two factors highlighted in this section,
personal capacity and preference for human contact may be particularly relevant, and these two
factors appear as variables in the present study.
Shopper Age and SST
Although no study to date appears to have focused primarily on the effect of shopper age
on attitudes toward SST, the age variable appears as a minor factor in several investigations of
SST. Overall, the conclusions regarding user age as an influence have been mixed.
Dabholkar et al. (2003) interviewed grocery store shoppers regarding awareness of, level
of use of, and degree of liking for self-scanning checkout. No significant differences were
found across 6 different age groups. In a similar vein, Weijters et al. (2007) interviewed
supermarket shoppers in Western Europe finding that age had no main effect or moderating
influence on attitudes toward self-scanning checkout (perceived usefulness, ease of use,
reliability, and fun). Also, Phang et al. (2006) reported no significant effect of age on
intention to use an online bank fund transfer service. However, in this latter study, the sample
consisted of only mature consumers, 50 to 90 years old. A sample of more diverse ages might
have yielded different results.
The only study to report a significant effect for shopper age on attitudes toward SST is
that of Simon and Usunier (2007). These authors hypothesized and found age to have a strong
negative effect on preference for SST over employee contact. An additional hypothesis by
these authors, perhaps seeking to explain the negative correlation between age and preference
for use of SST, was that older consumers would perceive SST to be too complex to use.
However, the correlation between age and perceived service complexity was not found to be
Despite somewhat mixed findings from the scant literature on the topic, the assumption
here is that shopper age will influence attitudes toward and use of retail SSTs. Also, it is hoped
that the results of the present study may help to better explain the significant negative effect of
age on preference for SST found by Simon and Usunier.
Three frameworks have been proposed to explain the adoption and use of technology.
The Technology Acceptance Model (Davis et al., 1989) hypothesizes that two consumer
beliefs (perceived usefulness and perceived ease of use) drive attitude and behavioral intention
toward use of the technology. Alternatively, the Technology Readiness construct
(Parasuraman, 2000; Lin and Hsieh, 2006) suggests that the consumer has a mix of mental
enablers and inhibitors that collectively determine predisposition toward technology. The
construct has four dimensions: optimism (the belief that technology offers increased control,
flexibility and efficiency in daily life), innovativeness (the degree to which the consumer is a
pioneer and thought leader), discomfort (perceiving a lack of control over technology), and
insecurity (a distrust of technology and skepticism of its ability to work properly). Finally,
Cognitive Complexity Theory (Ziefle and Bay, 2005) hypothesizes that technology adoption is
a function of the difficulty of learning the procedural steps necessary to properly interact with
Although the three frameworks mentioned above differ somewhat in their focus, they all
appear to imply that confidence in using the technology would be an important factor in
adoption. Not surprisingly, recent studies of retail SST have emphasized consumer confidence
in using technology (labeled self-efficacy or control in the literature) as a factor in their
investigations (Dabholkar and Bagozzi, 2002; Walker and Johnson, 2006; Marzocchi and
Zammit, 2006; Yen, 2005). Accordingly, confidence in using SST was selected as an
important variable to measure in this study.
One of the rare studies to compare young versus mature consumers and how they differ
in interacting with technology is that of Ziefle and Bay (2005). These authors contrasted
young (age 20-35) and old (age 50-64) consumers for their ability to learn and perform certain
tasks on both a simple and a complex mobile phone. They found that a significantly lower
percentage of the older age group was successful in accomplishing the four assigned tasks and
that older adults required a significantly longer time to complete the tasks (on average, nearly
double the amount of time spent by the younger age group). The inference from this finding is
that, based on prior encounters with technological devices, older consumers may have less
confidence than younger consumers in using retail SSTs. Thus, the following hypothesis may
H1: Compared to younger consumers, older consumers will report
significantly less confidence in using retail self-service technologies.
Researchers investigating SST usage have noted that consumers may differ greatly in
their preference for interaction with a machine versus a human employee (Dabholkar and
Bagozzi, 2002; Walker and Johnson, 2006). Given a choice, some people prefer not to have to
deal with an employee while others enjoy the social exchange and/or believe that personnel can
render a more customized service. Would consumers of different age groups differ in their
desire for human interaction in a retail exchange? A conceptual perspective that might help to
answer this question is continuity theory. This theory states that people make adaptive
behavioral choices in order to preserve ties to their own past experiences and their social world
(Atchley, 1989). Continuity theory suggests that older consumers, growing up at a time when
SSTs were not available and purchases were normally completed with some human interaction,
may have come to view human interaction as a standard part of the service script. This age
group may interpret interaction with a mechanical device as a “reduction” in service.
Conversely, younger consumers have grown up with the option of self-service for much of
their consumer life and they may have come to view SST as a service standard. Based on
continuity theory, the following hypothesis may be advanced.
H2: Older consumers will report that they miss human interaction during
self-service to a greater degree than do younger consumers.
One additional hypothesis is proposed, derivative from both H1 and H2. Assuming
that older consumers lack confidence in using SST and that they also miss the human
interaction that occurs in traditional retail exchanges, then it follows that this age group
might have less experience with SSTs than younger consumers. That is, given a choice
between completing a retail transaction through SST versus interaction with an employee,
older consumers are viewed as more likely to choose a traditional service encounter. This is
stated more formally as the following hypothesis.
H3: Compared to younger consumers, older consumers will report
experience with a fewer number of self-service technologies.
A convenience sample of adults residing in the Albuquerque, New Mexico, metropolitan
area during October, 2006, was obtained. According to a 2004 study by Acxiom Corporation,
Albuquerque ranks 31 of out of 150 major metro areas in terms of representing the overall U.S.
population in consumer and demographic characteristics (Bremner, 2004). Subjects were
approached in public places and asked for a few minutes of their time to fill out a paper and
pencil questionnaire. A quota sampling technique was used to ensure an approximately equal
balance for gender as well as age groups (18 to 28, 29 through 48, 49 and older). Although the
questionnaire was mostly closed-ended, many subjects made verbal comments and/or wrote
explanations to their answers on the questionnaire, and these responses were also noted.
The focus of the questionnaire was a series of questions recording use of and attitudes
toward retail SST. Following an initial checkbox line where the respondent indicated their
gender and age category (18-28, 29-48, and 49 and older), there were 7 numbered sections.
Question 1 asked subjects to check which of the following forms of SST they had used:
ATM - automated teller machine, airline ticketing at airport check-in, telephone or internet bank
transactions, self-service checkout at grocery or other retail store, automated hotel checkout,
online stock brokerage transactions, online purchase from a web site, and pay at the pump
gasoline sale. These forms of technology were numbered 1 through 8, each on separate line with
a checkbox. Besides capturing which individual technologies the respondent had used, a
summation was made for each subject of how many of the eight technologies he/she had
experienced. Question 2 had two parts, asking which one of the listed technologies had given the
subject the most trouble, and then asking the subject to attribute the trouble with the machine to
one of four possible sources. The sources were: a) the design of the machine, b) the respondent
not operating the machine properly, c) faulty training of employees and/or maintenance of the
machine, or d) an attempt to automate a transaction that is just difficult to automate.
Question 3 asked respondents to report how “confident” they are, in general, in using SST.
This item was anchored by “no confidence” and “very confident” and scaled from -4 to +4 with
zero as neutral. Question 4 asked subjects to report their agreement with the statement “A
consequence of self-service technologies is that human interaction during transactions is mostly
avoided – do you ‘miss’ having human interaction during your business transactions?” This
question was anchored by “not at all” and “yes, very much” and also scaled -4 to +4. To avoid
dealing with negative numbers during analysis, the responses for Questions 3 and 4 were
rescaled 1 through 9 at data entry.
Question 5 presented the respondent with a list of three possible motivations for the
introduction of SST and asked the subject to check the one most agreed with. The motivations
were: a) to help the customer, b) to help the company, and c) to help both customer and
company. Question 6 stated “If a store has both self-service checkout and traditional cashier
lines, how frequently do you use self-service checkout?” To respond, the subject was asked to
write-in the percent of time on a line.
Question 7, the last in the questionnaire, was prompted by an article in Food Manufacture
magazine by Ian Bowman, a manager with Siemens Automation. The author writes that his
company is developing a smart barcode, a plastic chip, which is inexpensive enough to be placed
on everyday grocery items (Bowman, 2005). Unlike a traditional barcode, however, this chip
can transmit information (price, brand, type of item and quantity) when activated by a radio
wave, opening up the possibility of simply pushing a filled shopping cart through a checkout
station instead of unloading, scanning the barcode of each item individually, and then re-loading
the cart. The plastic chip is question is very similar to the radio frequency identification (RFID)
devices in current use, except that the latter are silicon chips and far too expensive to be placed
on single grocery items such as a tube of toothpaste. The advantage of the plastic chips is cost;
they are produced in a process similar to three-color printing rather than requiring the dust-free
room manufacturing environment of silicon chips.
The purpose of Question 7 was to present the idea of the plastic check-out chip to subjects
and ask if they would be willing to pay any additional percent on their grocery bill for this added
convenience. The question was intended to be an indirect measure of how much the consumer
values a speedy checkout. The full statement of this item as it appeared in the questionnaire is
A new technology is being developed to place electronic “chips” on
product packages. These chips would be programmed with the price of
the product – sort of a smart barcode. At checkout, say at a grocery store,
a radio wave would be passed over the cart and price and product
information would be wirelessly forwarded to the cash register. The un-
loading and re-loading of the shopping cart would be eliminated by this
technology. Would you be willing to pay any more for this technology?
(please write-in a number below; the number is the percent more you
would be willing to have added to your grocery bill to pay for the
A total of 718 complete, usable questionnaires were returned and this constituted the
final sample size. The gender categories were perfectly balanced with 359 subjects each.
However, the age categories were slightly unbalanced with 233 subjects aged 18-28, a total
of 234 respondents 29-48 years old, and 251 participants aged 49 and older.
Interaction with SSTs
Subject experience with the 8 forms of SST listed in the questionnaire varied widely.
For the total sample of 718, the individual technology and the percent of subjects with
experience using the technology is as follows: ATM – 95.7%, pay at the pump – 94.4%, self-
service checkout at a retail store – 84.4%, online banking – 77.9%, online retail purchase –
74.2%, automated ticketing at airline check-in – 71.7%, automated hotel check-out – 37.3%,
and online stock brokerage transactions – 13.0%.
When asked to choose the one technology they found most troublesome to use, 44
subjects (6.1%), indicated that they did not find any of the eight SSTs to be problematic to use.
For the 674 respondents who did choose a technology as troublesome, self-service checkout at
grocery or other retail stores was overwhelmingly chosen as the most problematic technology
to use, listed as worst by 49.3% of the 674 subjects. In attributing the source of difficulty,
43.0% of reporting consumers chose machine design, 22.4% attributed the problem to
themselves (improper use), 10.3% indicated maintenance was to blame, and 24.2% said the
problem was trying to automate a transaction that is inherently difficult to automate. For the
whole sample of 718 subjects, 40.1% reported using retail self-service checkout 15% of the
time or less; only 25.2% said they use self-service 50% of the time or more.
Initial data analysis regarding the effects of gender and age group on the metric
dependent variables in this study revealed highly significant effects for age, no significant
effects for gender for any of the variables, and no interaction between the two factors. This
preliminary inspection of the data was deemed necessary just to ensure that gender did not
interact with the age and confound the interpretation of results based on age groups.
Age Group Differences
A comparison of the three age groups is shown in Table I. For each of the five metric
variables (top half of the table), multiple means comparisons revealed that the older age
group (49 and older) significantly (p < .05) differed from the younger age group (18-28).
Compared to younger subjects, older subjects had experience with fewer SSTs, had less
confidence in using these technologies, missed human interaction more, used self-service
checkout in retail stores less often when the option was available, and would be less willing
to pay a premium to have express checkout (technology that could enable product barcode
information to be transmitted to a checkout register by radio wave, eliminating the unloading,
scanning, and re-loading of the shopping cart). The older age group also significantly (p < .
05) differed from the middle (29-48) age group on the variables of confidence and missing
human interaction. That is, the older subjects reported less confidence and a greater sense of
missing human interaction than subjects in the middle age group. The results support all
three hypotheses (H1, H2 and H3).
[Take in Table I]
The remaining portion of Table I details the percent of subjects in each age category
who have used each of the SSTs. To test for differences in age group use within each
individual type of technology, Chi-Square analysis was conducted on a crosstabs for each
technology, comparing the three age groups by users and non-users of each technology.
Significant (p < .001) age group by user associations were found for ATM, automated airline
ticketing at check-in, telephone or internet banking, self-service checkout at a grocery or
other retail store, online purchase from a web site, and pay-at-the-pump gas transactions. For
each of these technologies, the association is a trend toward less usage with increasing age.
An additional crosstabs was constructed to examine the number of subjects in each age
group who selected each of three possible motivations offered as a rationale for the
introduction of SST. The three motivations to choose from were: to help the customer, to
help the company, or to help both customer and company. The resulting crosstabs is shown
as Table II. A significant (p = .002) association between age group and motivation choice
was found. The older age group was much more likely to report a corporate self-interest
motivation for the introduction of SST.
[Take in Table II]
A cursory review of the SST literature did not reveal a prior instance in which a
moderately large sample had been clustered on perceptions toward and use of SST. Yet,
such an exercise may have practical usefulness, revealing attitudinal, demographic, and
behavioral associations that could provide a framework for anticipating and also developing
solutions for SST usage problems.
To conduct cluster analysis, a K-means technique was used, requesting a three cluster
solution based on the following metric variables: confidence in using SSTs, the degree to
which human interaction is missed, and the percent of time self-service checkout is used in
grocery or other retail stores when the option is available. The results are shown in Table III,
detailing final cluster centers on the three chosen variables and also profiling each cluster on
[Take in Table III]
Subjects in the first identified cluster (n = 159) were extremely confident in using
SSTs, did not mind missing human interaction, and chose to use self-service checkout the
vast majority of the time when the option is available. This cluster is further characterized by
having a disproportionately small percentage of subjects 49 and older in its membership,
having the highest mean for total number of SSTs used of any cluster, and attributing the
motivation for introduction of SSTs to a desire to help both customer and company.
Respondents grouped into the second cluster (n = 194) are characterized by having a
high level of confidence in using technology, a neutral attitude about missing human
interaction, and choosing to use self-service checkout only about half the time when they have
the option. This group had the second highest mean for number of SSTs used and attributed
the motivation for technology introduction to helping both customers and the company.
Participants in the third identified cluster (n = 365) exhibited the least confidence with
technology, missed human interaction more than any other group, and generally avoided the
use of self-service checkout when they had the option. This cluster had a disproportionate
number of members aged 49 and older, had the lowest mean for total number of SSTs used,
and attributed a less than altruistic motivation for the introduction of self-service technology.
The clustering exercise was fruitful because it revealed that the group most resistant to
using self-service store checkout also attributed a selfish motive for the introduction of SST.
This finding is interesting because it raises the possibility that low usage behavior could be
related to the perceived motivation for the introduction of SST by the company rather than to
procedural issues in using the machines.
This study confirms the finding of Simon and Usunier (2007) that increasing age has a
negative effect on preference for SST over employee contact. Further, the present study
builds on this finding demonstrating that the behavioral tendency of older consumers to avoid
SST is associated with a relatively diminished confidence in their ability to use SST, a desire
for human interaction, and an attribution that SST is there to benefit the company rather than
the consumer. Not surprisingly, older consumers reported significantly less willingness to
pay a premium on their grocery bill for express checkout (technology that could allow
product barcode information to be transmitted to the checkout register by radio wave,
eliminating the unloading, scanning, and re-loading of the shopping cart).
The finding that the older age group attributes a corporate self-interest motivation for
the introduction of SST is of particular interest. Continuity Theory (Atchley, 1989) might
imply that older consumers, growing up before the introduction of SST, have adopted a
service script involving a human cashier as the norm. This age group might view SST as a
“reduction” in service. Indeed, several respondents spontaneously mentioned that scanning
and bagging their own groceries was “extra work” for them. Also, some female subjects
interpreted the presence of SST to mean that employees must have lost their jobs, and they
viewed this as a negative consequence. Although these are only anecdotal comments, they
provide some basis for why subjects may attribute a selfish motivation for the introduction of
SST, and they also suggest a managerial action that may counter the attribution (addressed in
Across all age groups in the sample, self-service checkout at a grocery or other retail
store was overwhelmingly reported as the most problematic technology to use. Indeed, this
particular SST has been singled out in the press for its ability to frustrate consumers
(Santosus, 2004). Of the eight types of SST surveyed in this study, self-service grocery
scanners are arguably the most procedurally complicated to use, requiring scanned items to
be placed upright (not leaning against a vertical surface) on the weighing pad within a certain
time limit after scanning. Also, for vegetable items or other non-barcoded products, the
consumer must look up the proper code and keypad the code into the machine. Based on the
number of production rules required for successful use, Cognitive Complexity Theory (Ziefle
and Bay, 2005) would predict some consumer difficulty accepting this technology. Among
all subjects in this study, 43% attributed problems with these machines to design flaws. This
may suggest that consumers perceive the problem to be under the control of retailers and an
issue to be resolved by the store rather than through a change in consumer behavior.
This study identified and profiled three groups of consumers based on attitudes toward
and use of SST. These clusters cut across age categories. It may be noted that 15% of the
subjects in the cluster most accepting SST were members of the older age group. Also, 24%
of respondents in the cluster most resistant to SST were in the young age category. This is
mentioned primarily to emphasize that age (a demographic variable) is a simplistic label and
it does not necessarily describe an individual’s attitudes, values, or behavior. The cluster
results are also noteworthy for the size of the cluster resistant to using self-service checkout
(50% of the sample) and for the fact that this group perceives a corporate self-interest motive
for the introduction of SST.
Self-service technologies are expensive to set up and maintain. They only become
economically practical when the labor cost savings from the employees they replace equals
their cost of implementation and maintenance or they allow the business to capture a market
segment that would otherwise be lost. In this light, managers may find it disturbing that only
25% of subjects in this study chose to use automated store checkout on more than half of
their shopping occasions.
There are some managerial actions that may potentially increase traffic throughput at
automated retail checkout. First, the store might offer customers a discount for using
automated checkout. The incentive of getting a discount for self-checkout might deflect the
attribution that the motivation for introducing SST is only to help stores decrease costs.
Also, since customers perceive bagging their own groceries to be extra work, a discount may
help to make self-checkout more tolerable. Second, managers could station an older,
technology-capable employee to supervise the self-checkout area. The presence of an age
group peer might induce an older, technology-challenged customer to try automated
checkout. Third, managers should place multiple self-checkout lines into operation within
any one store. Younger, technologically-savvy customers do not want to be stuck in line
behind a slower moving, technologically-challenged customer. Fourth, the design of
automated checkout machines should continue to be improved toward more user-friendly
features. For example, some self-checkout terminals are not well-equipped to handle coupon
redemption (Carr, 2004) and this could be improved upon.
The next technology that might be introduced to automate retail transactions, plastic
barcode “chips” that function like RFID devices (Bowman, 2005), was briefly described
earlier in this paper. Although rolling a full shopping cart through a checkout terminal and
wirelessly transmitting product information to the checkout register saves time and effort, it
could be predicted that some consumers will be disturbed by the idea that their items are not
bagged (seeing this as a reduction in service). Retailers should pro-actively deal with this
perception, perhaps by offering a small discount for plastic chip checkout.
One final managerial point can be made, relating to both shopper age and self-service
checkout. The IHL Consulting Group report mentioned earlier noted that consumer purchase
of impulse items (such as chewing gum, magazines, salty snacks, soda, and bottled water) is
43% less when self-checkout is used versus traditional checkout (Schuman, 2006). This is
partly due to the fact that impulse items are not typically merchandised near self-checkout
lanes. Even if they were on display, consumers are so engaged in finding an open lane and
scanning their items that they may not notice them. This problem is made even worse by the
fact that the segment most likely to use self-checkout (consumers age 48 and younger, as
found in this study) are precisely the shoppers most prone to purchasing impulse items
(Wood, 1998; Chuang and Lin, 2005). Retailers doing ROI (return on investment)
calculations for self-checkout will need to factor-in these lost impulse merchandise sales.
Limitations and Future Research
This study had several limitations. First, only eight forms of SST were studied, and
only one technology (self-service checkout in a retail store) was investigated in depth. The
findings of the study may not generalize to other forms of SST. Second, the design of the
questionnaire included only a small number of variables. The brevity of the number of
measures limits the associations that may occur and the conclusions that may be drawn.
Third, given that SSTs may be re-engineered, improved, and made more user-friendly over
time, the findings in this study are time-context dependent. The findings may not generalize
to future forms of SST.
The selfish attribution for SST uncovered in this study as well as spontaneous
comments made by participants suggests that a qualitative study exploring additional reasons
for using or not using SST may be insightful.
Atchley, R.C. (1989), “A Continuity Theory of Normal Aging”, Gerontologist,
Vol. 29 No. 2, pp.183-190.
Bitner, M.J. (2001), “Self-Service Technologies: What Do Customers Expect?”,
Marketing Management, Vol. 10 No. 1, pp. 10-11.
Bitner, M.J., Ostrom, A.L. and Meuter, M.M. (2002), “Implementing Successful Self-Service
Technologies”, Academy of Management Executive, Vol. 16 No. 4, pp. 96-108.
Bowman, I. (2005), “Checkout Chips”, Food Manufacture, Vol. 80 No. 12, p. 22.
Bremner, K. (2004), “Albany Ranked No. 1 Test Market in Acxiom Study”, Direct Marketing
News, May 24, available at: www.dmnews.com/Albany-Ranked-No-1-Test-Market-in-
Carr, D.F. (2004), “Gotcha! The Problems with Self-Service Checkout Systems”,
Baseline, Vol. 1 No. 27, p. 34.
Chuang, S. and Lin, C. (2005), “The Effect of Individual Differences on Adolescents’
Impulsive Buying Behavior”, Adolescence, Vol. 40 No. 159, pp. 551-558.
Dabholkar, P.A. and Bagozzi, R.P. (2002), “An Attitudinal Model of Technology-Based
Self-Service: Moderating Effects of Consumer Traits and Situational Factors”,
Journal of the Academy of Marketing Science, Vol. 30 No. 3, pp. 184-201.
Drabholkar, P.A.., Bobbit, L.M. and Lee, E. (2003), “Understanding Consumer Motivation and
Behavior Related to Self-Scanning in Retailing”, International Journal of Service Industry
Management, Vol. 14 No. 1, pp. 59-95.
Davis, F.D., Bagozzi, R.P. and Washaw, P.R. (1989), “User Acceptance of Computer
Technology: A Comparison of Two Theoretical Models”, Management Science,
Vol. 35 No. 8, pp. 982-1003.
IHL Consulting Group (2007), “Consumers on Track to Spend $1.3 Trillion a Year at
Self-Service Machines by 2011”, available at:
Lin, J.C. and Hsieh, P. (2006), “The Role of Technology Readiness in Customers’ Perception
and Adoption of Self-Service Technologies”, International Journal of Service Industry
Management, Vol. 17 No. 5, pp. 497-517.
Marzocchi, G. and Zammit, A. (2006), “Self-Scanning Technologies in Retail: Determinants
of Adoption”, The Service Industries Journal, Vol. 26 No. 6, pp. 651-669.
Meuter, M.L., Ostrom, A.L., Roundtree, R.I. and Bitner, M.J. (2000), “Self-Service
Technologies: Understanding Customer Satisfaction with Technology-Based Service
Encounters”, Journal of Marketing, Vol. 64 No. 3, pp. 50-64.
Meuter, M.L., Bitner, M.J., Ostrom, A.L. and Brown, S.W. (2005), “Choosing Among
Alternative Service Delivery Modes: An Investigation of Customer Trial of Self-Service
Technologies”, Journal of Marketing, Vol. 69 No. 2, pp. 61-83.
Parasuraman, A. (2000), “Technology Readiness Index (TRI): A Multi-Item Scale to Measure
Readiness to Embrace New Technologies”, Journal of Service Research, Vol. 2 No. 4,
Phang, C., Sutanto, J., Kankanhalli, A., Li, Y., Tan, B. and Teo, H. (2006), “Senior Citizens’
Acceptance of Information Systems: A Study in the Context of e-Government Services”,
IEEE Transactions on Engineering Management, Vol. 53 No. 4, pp. 555-569.
Rosen, C. (2001), “More Retailers to Tell Customers, ‘Do It Yourself’”, Information Week,
February 26, 2001, Issue 826, p. 29.
Santosus, M. (2004), “Life in the Not-So-Fast Lane”, CIO, Vol. 18 No. 6, pp. 38-40.
Schuman, E. (2006), “Self-Checkout Killing Impulse Items”, July 25, available at:
Simon, F. and Usunier, J. (2007), “Cognitive, Demographic, and Situational Determinants
of Service Customer Preference for Personnel-in-Contact Over Self-Service Technology”,
International Journal of Research in Marketing, Vol. 24 No. 2, pp. 163-173.
Solomon, M.R. (2004), Consumer Behavior: Buying, Having, and Being, Pearson Prentice Hall,
Upper Saddle River, NJ, p. 513.
Walker, R.H. and Johnson, L.W. (2006), “Why Consumers Use and Do Not Use Technology-
Enabled Services”, Journal of Services Marketing, Vol. 20 No. 2, pp. 125-135.
Weijters, B., Rangarajan, D., Falk, T., and Schillewaert, N. (2007), “Determinants and
Outcomes of Customers’ Use of Self-Service Technology in a Retail Setting”,
Journal of Service Research, Vol. 10 No. 1, pp. 3-21.
Wood, M. (1998), “Socio-Economic Status, Delay of Gratification, and Impulse Buying”,
Journal of Economic Psychology, Vol. 19 No. 3, pp. 295-320.
Yen, H.R. (2005), “An Attribute-Based Model of Quality Satisfaction for Internet Self-Service
Technology”, The Service Industries Journal, Vol. 25 No. 5, pp. 641-659.
Ziefle, M. and Bay, S. (2005), “How Older Adults Meet Complexity: Aging Effects on the
Usability of Different Mobile Phones”, Behaviour & Information Technology,
Vol. 24 No. 5, pp. 375-389.
Statistics for Selected Variables by Age Group (overall n = 718)
n = 233
n = 234
49 and older
n = 251
Number of self-service
technologies have used
Confidence in using self-service
Degree to which human
interaction is missed
Percent of time subject opts for
self-service checkout in store
Percent more subject willing to
pay for express checkout
Percent of age group having
ATM 99.1 97.4 90.8
Automated airline ticketing 82.0 74.8 59.4
Internet banking 88.4 82.9 63.3
Self-service checkout in store 94.8 91.5 68.1
Automated hotel checkout 33.0 39.3 39.4
Online stock brokerage 10.3 16.7 12.0
Online retail purchase 90.1 78.2 55.8
Pay at the pump gas sale 98.3 97.0 88.4
Note: Paired entries are means (top) and standard deviations (bottom)
Motivation for Self-Service by Age Group (overall n = 718)
Age 18-28 Age 29-48
Age 49 +
To help customer 27 23 14 64
To help company 75 90 124 289
To help both customer
and company 131 121 113 365
Totals 233 234 251 718
Note: Cell entries are actual subject counts. No cells have expected counts less than 5.
Profile of Identified Clusters (overall n = 718)
n = 159
n = 194
n = 365
Final Cluster Centers
Confidence in using self-
service technologies 8.21 7.67 6.57
Degree to which human
interaction is missed 3.33 4.76 6.07
Percent of time opt for
self-service checkout 85.62 44.83 8.55
% Men 53.5 51.5 47.7
% Women 46.5 48.5 52.3
% Age 18-28 45.3 37.6 24.1
% Age 29-48 39.6 35.1 28.2
% Age 49 and older 15.1 27.3 47.7
Number of self-service
technologies have used
% say self-service introduced to
help customer 10.7 8.2 8.2
% say self-service introduced to
help company 32.1 35.1 46.6
% say self-service introduced to
help both customer and company 57.2 56.7 44.9
Note: Paired entries are means (top) and standard deviations (bottom)