IMPORTANCE-PERFORMANCE ANALYSIS: DIFFERENT APPROACHES

Abstract

In today's environment of rapid development of technology and global competition, where speed is a key feature of organizations' survival, managers are facing with difficult tasks. They are expected to regularly evaluate the level of customer satisfaction with provided services/products. Importance-performance analysis (IPA) is practical and useful method, that can help policy-makers to identify service/product elements which allocation of resources could contribute to higher satisfaction of users. The purpose of the study is to provide a comprehensive and systematic review of the literature which includes different approaches and modifications of IPA and determine the benefits and disadvantages of those approaches. The selection of relevant papers for this study was carried out by EBSCO database, including keywords Importance, Performance and Analysis. From the total number of available studies (1075), 13 papers fulfilled the criteria of final selection, as part of this systematic review. This study found that the traditional IPA is based on erroneous underlying assumptions. In terms of the IPA chart, the diagonal model (DM), in combination with Data centered quadrant model (DCQM), is given precedence over the quadrant model (QM) and Scale centered quadrant model (SCQM). Another very good suggestion is coordinate axis transformation, where fixed values are used for coordinate intersection. Customer satisfaction data obtained by direct method suffers from conscious inclinations of subjects, thus indirect estimation methods for importance and performance dimensions are recommended. Each approach has its advantages and disadvantages, and it was therefore difficult to say for any of the aforesaid IPA modifications that it was better than others and that it gave more relevant picture of the organization. The decision is made by managers to choose the method that will effectively and efficiently lead to desired information.

Full text

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
IMPORTANCE-PERFORMANCE ANALYSIS: DIFFERENT APPROACHES
Šemso Ormanović1, Alen Ćirić1, Munir Talović1, , Haris Alić1, Eldin Jelešković1, Denis
Čaušević1
1Faculty of Sport and Physical Education, University of Sarajevo, Sarajevo, Bosnia and Herzegovina
Review paper
Abstract
In today's environment of rapid development of technology and global competition, where speed is a key feature
of organizations' survival, managers are facing with difficult tasks. They are expected to regularly evaluate the
level of customer satisfaction with provided services/products. Importance-performance analysis (IPA) is
practical and useful method, that can help policy-makers to identify service/product elements which allocation of
resources could contribute to higher satisfaction of users. The purpose of the study is to provide a
comprehensive and systematic review of the literature which includes different approaches and modifications of
IPA and determine the benefits and disadvantages of those approaches. The selection of relevant papers for this
study was carried out by EBSCO database, including keywords Importance, Performance and Analysis. From the
total number of available studies (1075), 13 papers fulfilled the criteria of final selection, as part of this
systematic review. This study found that the traditional IPA is based on erroneous underlying assumptions. In
terms of the IPA chart, the diagonal model (DM), in combination with Data centered quadrant model (DCQM), is
given precedence over the quadrant model (QM) and Scale centered quadrant model (SCQM). Another very
good suggestion is coordinate axis transformation, where fixed values are used for coordinate intersection.
Customer satisfaction data obtained by direct method suffers from conscious inclinations of subjects, thus
indirect estimation methods for importance and performance dimensions are recommended. Each approach has
its advantages and disadvantages, and it was therefore difficult to say for any of the aforesaid IPA modifications
that it was better than others and that it gave more relevant picture of the organization. The decision is made by
managers to choose the method that will effectively and efficiently lead to desired information.
Key words: Satisfaction, Gap, TOPSIS, Three factor theory, PAT, KANO
Introduction
Understanding what makes the customer, or service
user, satisfied is the key to the success of any
organization, regardless of its activity. It is
undisputed, that higher quality of services/products
quality generates greater satisfaction of their users.
This sounds very simple, but in practice, it is
difficult to achieve the goal of having satisfied
customers, given the technological explosion and
fierce competition over the last four decades. At the
end of seventies of the twentieth century, Martilla
and James (1977) have developed a technique
called Importance-Performance Analysis (IPA),
which has become a very popular managerial tool
for organizational performance development. This
technique helps customer satisfaction
understanding, as well as detecting and placing
priority on those services/products which
improvement is necessary.
IPA is actually a graphic method which is showed in
a two-dimensional coordinate system, the average
values of importance and performance of different
services/products elements, which are calculated in
relation to one another, mainly in the area divided
into four quadrants (M. Feng, Mangan, Wong, Xu, &
Lalwani, 2014). In the traditional IPA, the average
value of importance and performance of different
service attributes are provided by direct evaluation
by users and calculated in the specified coordinate
system, where the horizontal axis represents
performance, and the vertical axis represents
importance (Martilla & James, 1977) (Figure 1).
Figure 1. Original IPA
Source: Lin et al. 2009

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
''Performance'' represents the user's perception of
the quality of services delivered by the organization,
while the ''importance'' refers to the assessment of
the importance of those services by users (Yildiz,
2011). Depending on in which quadrant a certain
service is located, managers can decide which
services are the top and low priorities for
improvement. As seen in Figure 1, quadrant 1 (high
importance/high performance) is called ''keep up
the good work'' and represents the strong side and
competitive advantage of companies, which task is
to continue to maintain the quality of those
elements contained in it. Quadrant 2 (high
importance/low performance) is called ''concentrate
here'' and includes elements that require immediate
corrective action. If the element is located in
quadrant 3 (low importance/low performance),
which is called ''low priority'', this element does not
represent any threat to the organization, but the
manager could rather think about the option of
transferring resources from these elements to those
requiring urgent action. Quadrant 4 (low
importance/high performance) is called ''possible
overkill'' and includes elements whose high qualities
have no impact on the total customer satisfaction,
so managers with these elements can also think
about the allocation of resources (Levenburg &
Magal, 2004). This chart allows managers an easy
insight into the overall picture of the situation, after
which they can start drafting action plans and
estimate costs of necessary improvements.
Given that IPA is a very simple and practical
method, which does not require excessive
knowledge and application of statistical methods,
and it is used in many different business areas such
as medical service (Piñeiro, Boubeta, & Mallou,
2006; Yavas & Shemwell, 2001), tourism (Hudson,
Hudson, & Miller, 2004; Oh, 2001; Wade & Eagles,
2003; Zhang & Chow, 2004), traffic and
transportation (Chen & Chang, 2005; Tam & Lam,
2004), education (Chang, 2014; Nale, Rauch,
Wathen, & Barr, 2000), production (Matzler, Bailom,
Hinterhuber, Renzl, & Pichler, 2004), services (C.-M.
Feng & Jeng, 2005; Joseph, Allbright, Stone,
Sekhon, & Tinson, 2005), and others. Although the
original IPA is a very useful and valuable method,
over the years, it has been subject of numerous
modifications and criticism by many researchers
(Abalo, Varela, & Manzano, 2007; Bacon, 2003;
Deng, Kuo, & Chen, 2008; M. Feng et al., 2014;
Johns, 2001; Kuo, Chen, & Deng, 2012; Lin, Chan,
& Tsai, 2009; Matzler et al., 2004). The biggest
debates revolved around two questions: 1) the way
to divide up the field containing elements of
different significance for decision makers, 2)
method of measuring and collecting the results of
importance and performance attributes (Abalo et
al., 2007).
The review of current literature resulted in the
following research questions: 1) which method of
division of the chart fields delivers the most relevant
information, 2) what is the best method for
calculating importance and performance, i.e., which
method results in the best prediction of corrective
action priority. The aim of this paper is to make a
comprehensive and systematic review of literature
referring different approaches and modifications of
IPA and determine the benefits and disadvantages
of these different approaches. This study should
help managers to select those modified IPA
methods which will ultimately contribute to the most
relevant information about the level of satisfaction
of their services/products users, as well as
indications of possible corrective actions.
Methods
Selection of studies
In order to make comparison of different
approaches to IPA, a comprehensive review of
current literature was carried out. Key words that
were used Importance, Performance, and Analysis
in the following EBSCO database: Academic Search
Complete, Business Source Complete, SPORTDiscus,
SocINDEX, CINAHL, MasterFILE Premier, ERIC,
MEDLINE, Health Source - Consumer Edition, Health
Source: Nursing/Academic Edition, Library,
Information Science & Technology Abstracts,
GreenFILE, GeoRef, Regional Business News, CAB
Abstracts. The search was conducted in May 2016,
and initial results showed a total of 1,075 papers
mentioning the keywords. The inclusion criteria of
the study were: 1) full text papers, 2) papers with
available references, 3) scientific (reviewed)
journals, and 4) papers from the year 2000 to date
(plus original research IPA (Martilla & James,
1977)). When the selected criteria were applied, the
resulting total number of selected papers was 73.
Following additional reading of titles and abstracts,
this number came down to the final number of 13
papers, which met the criteria included in this
systematic review, including original research by
Martilla and James. This research was focused on
those papers that were focused on various
modifications and transformations of the original
IPA (Figure 1).
Results and Discussion
Different approaches to dividing IPA chart
IPA is a method is useful management tool to easily
identify the strengths and weaknesses of an
organization, and to assess products/services
customer satisfaction. In the traditional IPA,
according to Martilla and James (1977), the
Quadrant model (QM) was used. This model consists
of a pair of coordinates and four fields containing
elements of importance and performance for a
particular service/product, and which average
values were calculated from direct assessment of
users of services/products. It was originally
suggested the cross section of coordinates to be in
the middle of the scale - Scale centered quadrant
model (SCQM) (eg. on a scale of 1 to 7, cross
section is the result of 4) (Figure 2) (Martilla &
James, 1977), which was widely criticized. Applying
SCQM, most elements of the service that had
average result for the importance attribute would be

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
placed in Quadrant 1 (keep up the good work),
resulting in a small discriminative value of the IPA
chart, which could lead managers in making wrong
decisions (Rial, Rial, Varela, & Real, 2008). Thus,
the first solution to this problem, which was
suggested in the same year of the method
occurrence by the same authors, and later other
ones (Hollenhorst, Olson, & Fortney, 1992; Martilla
& James, 1977; Rial et al., 2008), proposed the
cross section of coordinates using the importance
and performance average values - Data centered
quadrant model (DCQM) (Figure 3), therefore
contributing better discriminatory power of the IPA
chart. One of the biggest QM flaws lies in the
prioritizing of elements for improvement, where the
smallest element shifts in the coordinate system can
lead to drastic changes in the identification of
priorities (Bacon, 2003; Tontini, Picolo, & Silveira,
2014).
Since the DCQM encountered criticism, scientists
continue to search for the best graphic solution of
the IPA, so they incorporated the concept of ''gap''
(discrepancy or gap), which is calculated as the
difference between performance and importance
(still known as the gap between the expectations
and perceptions of service/products by users) (M.
Feng et al., 2014; Rial et al., 2008) in DCQM. The
result of this concept was adding a diagonal line at
an angle of 45˚, also called the iso-priority line, or
iso-rating line, where the values of importance are
the same as values of performance (no gap) (Figure
4) (Abalo et al., 2007; Rial et al., 2008). Elements
that are located above the diagonal line are those
with negative discrepancy
(performance<importance), while those below the
diagonal line are those with positive discrepancy
(performance>importance), and the element
distance from the diagonal line is taken as an
indicator of the satisfaction/dissatisfaction level
(Rial et al., 2008). This concept requires a different
interpretation of the chart, so the elements that are
above the diagonal line, regardless of them being
inside the quadrants 1 or 3, must be interpreted as
users’ dissatisfaction. Overall, the diagonal model
(DM) showed better performance than QM (Bacon,
2003).
In order to overcome the interpretation problems of
the chart with incorporated diagonal line, Abalo et
al. (2007) suggested a chart that includes a
diagonal line but which has a different division of
fields (Figure 5). According to this model, which is a
combination of QM and DM, the upper left quadrant
is considerably higher compared to the original IPA.
All elements that are above the diagonal line are
candidates for corrective action, with the distance of
elements from the diagonal line dictating the
priority (Abalo et al., 2007). The remaining three
quadrants have identical interpretation as the
traditional IPA.
Figure 3. Data centered quadrant model
Source: Albayrak et al. 2014
Figure 2. Scale centered quadrant model
Source: Yildiz, S. M. 2014
Figure 4. Diagonal line
Source: Rial et al. 2008

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
Trying to identify a unique model that would enable
managers of different profiles to get consistent
strategic indicators, Lin et al. (2009) proposed a
model called the Importance-performance gap
analysis (IPGA) (Figure 6), which integrates the IPA
and gap analysis, and in which the fixed values
obtained by transforming functions of two axes, and
using relative importance (RI) and relative
performance (RP), are used for the coordinates
cross section. The value of intersection points for
the RP is 0, while for RI it is 1, and the vertical axis
represents the ''best performance axis''. According
to the IPGA model, element's distance from the
intersection coordinates dictates the size of the
priorities for potential corrective action (eg. in
Figure 6, in quadrant 2 of the IPGA matrix, the
element A has a higher resources
adjustment/allocation priority in relation to the
element B).
Noting that in previous studies that have studied
the IPA, no one has considered the possible impact
of certain elements of services/products on
customers satisfaction, if they are offered or
improved. Tontini et al. (2014) suggests the
''Improvement-gaps analysis (IGA)'' model. A
specific feature of this model is the shaded part
around the coordinates, which represents 90%
probability of axis position (Figure 7b). It is up to
managers to decide whether a given element, if it is
in the shaded area, requires corrective action.
Different approaches to calculating importance and
performance attributes
As IPA was becoming a very useful, practical and
applicable tool for managers of different profiles,
scientists were trying to come up with the most
relevant results using different calculation methods
for importance and performance. Although IPA is a
very valuable method, numerous studies have
confirmed that it has several disadvantages, so
Matzler et al. (2004), Deng et al. (2008), Albayrak
and Caber (2014) state that the two basic
assumptions underlying the traditional IPA are
completely wrong: 1) attribute performance and
attribute importance are independent variables, and
2) the relation between attribute performance and
overall performance is linear and symmetric. In fact,
assumption is that, the relation between the
attribute performance and overall customer
satisfaction is asymmetrical, while the relation
Figure 7b. IGA matrix – shadow areas
Source: Tontini et al. 2014
Figure 7a. IGA matrix
Source: Tontini et al. 2014
Figure 6. IPGA matrix
Source: Lin et al. 2009
Figure 5. Revised IPA
Source: Rial et al., 2008

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
between performance and importance attributes is
causal (Deng et al., 2008), which indicates the
necessity of revising the original IPA. Numerous
studies have confirmed that the attribute
performance is less controversial in comparison to
the atribute importance (Abalo et al., 2007),
because the attribute importance, in some way, is a
reference point with which we compare the current
state of service/product element.
By analyzing the literature, two different methods of
measuring the above mentioned dimensions used in
IPA are: 1) direct measurement (based on the
Likert scale, or other measuring scale); 2) indirect
measurement (partial correlation analysis, multiple
regression analysis), obtained from the attribute
performance score, or by multivariate regression
analysis of an overall satisfaction rating as the
dependent variable with scores of individual
service/product elements as the independent
variable (Abalo et al., 2007; Bacon, 2003). In cases
where the dependent variable is not normally
distributed, the logistic regression analysis is also
applied for the attribute importance calculation
(Bacon, 2003).
According to the original IPA, the results for the
importance and performance attributes are reached
by direct assessment by users of services/products,
actually, a statement is requested from the users
about the level of satisfaction of certain
service/product element and the importance of the
same element (Martilla & James, 1977), and whose
averages are included in the final calculation. The
most commonly used measurement scale in the
survey questionnaire is the Likert five, seven and
nine-point scale (Bacon, 2003; Tontini et al., 2014).
This approach has been subject to numerous
criticisms by researchers, who state several
problems with this approach. The first one is that
direct evaluation might be the reflection of a desire
or conscious tendency of respondents, the second
one refers to typical tendency of respondents to
give high marks for attribute importance, and the
third one refers to the ''phenomenon of the crowd,''
where the participants face the problem of lack of
involvement and expertise when evaluating certain
services or products (Abalo et al., 2007; Bacon,
2003). While some authors claim that the direct
measurement method for performance and
importance attributes gives more relevant results
(Bacon, 2003; Rial et al., 2008), most authors
agree that the indirect method of measuring these
attributes, especially attribute importance, are more
exact (Abalo et al., 2007; Albayrak & Caber, 2014;
Deng et al., 2008; M. Feng et al., 2014; Johns,
2001; Lin et al., 2009; Matzler et al., 2004; Tontini
et al., 2014). Since different studies showed
disadvantages of the traditional IPA, modifications
of this model started to emerge, integrating one, or
more models. M. Feng et al. (2014) compared
different approaches to the integrated IPA with the
Gap 1 analysis (the difference between attribute
importance and attribute performance obtained by
direct measurement), Gap 2 analysis (includes a
performance difference between the focal firm and
competing firms, where the data for the importance
and performance are also obtained by direct
measurement), and the tri-factor theory (based on
the argument that not all elements of the
service/product are equally important). These
authors believe that each of these four approaches
has its advantages, and it is impossible to conclude
which one of them is the best. They are therefore
proposing the integration of traditional IPA with Gap
1 analysis (the results of Gap 1 analysis should be
integrated with average values of attribute
importance ranks), traditional IPA with Gap 2
analysis (identification of important elements of the
service/product whose performance is worse than
that of its competitors), and traditional IPA with tri-
factor theory (useful and practical instrument for
the classification of elements by importance), which
would contribute to greater relevance of the results.
However, the authors state that the most effective
solution is the integration of all four approaches.
The Gap 1 analysis has its roots in the work of
Parasuraman, Zeithaml, and Berry (1988), who
used the SERVQUAL scale to point out the
importance of identifying the gap between the
expected and observed quality of service. Many
researchers who have studied the problems of
quality of service, stated that the gap analysis is the
most effective indicator to improve overall customer
satisfaction of services/products of an organization
(Lin et al., 2009), which in combination with the
traditional IPA increases the strength of this
analytical method. Since Gap 1 Analysis shows a
problem of conscious tendencies of respondents,
Gap 2 analysis was suggested (also known as
performance ratio), taking into consideration the
gap between the performance of the focal company
and its competitors, where the elements with high
importance and performance are called ''prominent
factors'' and they represent competitive advantages
(M. Feng et al., 2014). The most relevant point of
reference, with which we can compare the current
state of the organization, is the best competitor
(benchmarking) (Deng et al., 2008). Ignoring the
competitive market state can lead the organization
to fatal results. Disadvantage of the Gap 2 analysis
is that it does not include the gap between the
importance and performance attributes of the focal
firm, which are very good indicators of the state of
an organization.
In this paper, Lin et al. (2009) used the integrated
model of IPA and Gap 2 Analysis (IPGA), where for
the attribute importance and attribute performance
they do not use absolute, but relative values
(relative importance RI and relative performance
RP). Besides, the transformation of the coordinate
axes functions, the concept of gap analysis was
included by using the ''index function''. The relative
dimensions are based on the transformation of the
gap direction (positive or negative) of the individual
service elements, with average function values.
Although IPGA has an advantage over the other
modified methods, because of the coordinates fixed
values, the complexity of the analysis is a feature
that makes this methodological approach
questionnable.

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
Another integrated model of the Gap and IPA
analysis (IGA) has been used in the research of
Tontini et al. (2014). This model, unlike the
traditional IPA, makes comparisons between the
significance of different services/products elements,
analyzing their potential impact on customer
satisfaction, if they are offered or improved.
According IGA, using a similar Kano model
structure (Kano, Seraku, Takahashi, & Tsuji, 1984),
users are asked about the expected satisfaction and
dissatisfaction. More precisely, a questionnaire is
composed of functional questions (the elements
have high performance), dysfunctional questions
(the elements have low performance), and
questions with the current satisfaction with a certain
element of service/product (Figure 8), and the
questions are arranged in randomized order to
avoid polarized responses. The IG dimension
(improvement gap – Figure 7a-b) is obtained by
subtracting the average value of satisfaction by
service/product element, from the average value of
functional questions for the same elements, which
results are then standardized and plotted on the
horizontal axis. The specificity of this method are
the dysfunctional questions which include ''expected
dissatisfaction'' of users as a measure of the
significance of an element, and whose results are
then standardized and plotted on the vertical axis
(Figure 7a). In dysfunctional questions, respondents
are asked to imagine a bad situation and give a
response in accordance with that situation.
According to the Kano two-dimensional model of service/product quality oriented toward customer
requirements, elements of services/products are divided into five different categories, with having a different
effect on the final customer satisfaction; 1) attractive quality: the presence of this element affects customer
satisfaction, while the absence does not encourage dissatisfaction, 2) one-dimensional quality: customer
satisfaction increases with fulfillment of these elements, and vice versa, 3) must-be quality: a basic measure of
a product/service where the absence of these elements generates extreme dissatisfaction of users, 4) indifferent
quality: the presence or absence of these elements does not cause satisfaction/dissatisfaction of users, and 5)
reverse quality: the presence of these elements causes dissatisfaction of users and vice versa) (Kano et al.,
1984).
Although this analysis facilitates the understanding
of user requirements, it ignores the fact that the
importance and performance dimensions of quality
can have an impact on the classification of
services/products elements, while IPA does not
consider two-dimensional quality characteristics.
The integration of these two models would
contribute to an increased accuracy of the Kano
method during services/products elements
categorization, eliminating flaws of the IPA method,
which considers only one-dimensional quality
elements, where the final result would be more
appropriate quality strategy (Kuo et al., 2012).
Recent customer satisfaction surveys (Albayrak &
Caber, 2014; Deng et al., 2008; M. Feng et al.,
2014; Matzler et al., 2004) suggest that elements of
the services/products fall within three categories
(basic factors, excitement factors, and performance
factors), where each of these factors has a different
impact on customer satisfaction, and this theory is
known as the tri-factor theory. Basic factors are the
minimum requirements that an organization needs
to offer to services/products customers, and if they
are not met, this is causing users’ dissatisfaction,
but if they are fulfilled, they do not lead to customer
satisfaction. In other words, the negative
performance of these elements has greater impact
on the total customer satisfaction than positive.
Excitement factors are those that cause customer
satisfaction, if they are present, but if they are not
offered, they do not cause dissatisfaction.
Performance factors result in customer satisfaction,
if they are offered, but also cause dissatisfaction, if
they are not offered. Therefore, the ratio
"performance of services/products elements - total
Figure 8. IGA sample question
Source: Tontini et al. 2014

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
customer satisfaction" for the basic factors and
excitement factors are nonlinear and asymmetric,
while for the performance factor this relationship is
linear and symmetrical, suggesting questionable
assumptions and applicability of the traditional IPA.
In fact, the tri-factor theory suggests a hierarchy of
attribute importance, where the basic factors are
the most important, followed by performance
factors, and finally the excitement factors. Matzler
et al. (2004) have calculated attribute importance in
their research by multiple regression analysis with
the total satisfaction as the dependent variable and
attribute performance as the independent variable.
Attribute performance is recoded in order to form
''dummy'' variables ''with low performance (0.1)'',
''high performance (1.0),'' and ''average
performance (0.0)'' , where two regression
coefficients (one to measure the impact when
performance is small, and the other when
performance is big) are obtained for each variable in
order to assess the impact of given element
performance on the total satisfaction. However,
subsequent studies determined that this method of
indirect calculation of attribute importance is
inadequate because of the potential impact of
multicollinearity between independent variables on
the regression coefficient estimate, and they state
that the partial correlation analysis is a more
appropriate tool to quantify the impact of
independent variables on dependent variables
(Deng et al., 2008). In the paper of Deng et al.
(2008), a method that integrates the partial
correlation analysis and natural logarithmic
transformation was used for indirect measurement
of the attribute importance.
This method consists of three steps: 1) the
transformation of all performance attributes in
natural logarithmic form, 2) set the natural
logarithm performance attributes and total users
satisfaction as a variable in multivariate correlation
model, 3) conduct a partial correlation analysis for
each attribute performance and total customer
satisfaction. The attribute performance of different
elements is calculated with the Gap 2 analysis,
dividing the average values of the focal company
performance by the average performance values of
the best competitor (performance ratios - a method
that accurately indicates the actual differences in
performance). Exploring the service quality of
fitness clubs, Albayrak and Caber (2014) calculated
asymmetry of the relation between the performance
of individual elements and the overall customer
satisfaction by a multiple regression analysis, using
factor scores of service quality as independent
variables, to explain the dependent variables of the
total satisfaction. Standardized regression
coefficients were used to assess the relative impact
of attribute performance on the total customer
satisfaction.
The Technique for order performance by similarity
to ideal solution (TOPSIS) is a technique which has
been used in the work of Sooreh, Salamzadeh,
Saffarzadeh, and Salamzadeh (2011), where aim
was to set priorities for various quality elements of
services/products. According to this method, the
best alternative should be within the shortest
distance from the ideal solution. Therefore this
study considered the distance of an certain element
from the ideal solution and the negative ideal
solution. The aim is to recommend the shortest
geometrical distance to the ideal solution for
individual elements of services/products. The
authors note that the combination of IPA and
TOPSIS methods reveal important and critical
elements of the organization.
In his research, Johns (2001) applied a new
methodological approach to IPA using the profiles
accumulation technique (PAT) to identify elements
of service and calculate attributes importance of
these elements. From the data obtained through
PAT, a closed type questionnaire was composed
with the aim of calculation of attributes
performance. This semi-quantitative quality
assessment methodology, which is based on free
responses of users, is asking them to list the
aspects of services that they like the most and not,
and explain their reasons. The number of responses
for every service quality element is an indicator of
attribute importance. By applying PAT, it is possible
to estimate the experience of service users without
any external influence on their answers.
Standardized results were used for both dimensions
Conclusion
Final considerations
Analyzing all of the methods and various
modifications of IPA, it can be concluded that these
tools are very useful assessment instruments for
the current state of organization, and their greatly
assist managers in making valid decisions. The fact
that is confirmed, by many researches, is that the
foundations of traditional IPA are based on wrong
assumptions, thereby the results of the original IPA
became the subject of doubt (Albayrak & Caber,
2014; Deng et al., 2008; Matzler et al., 2004).
During the implementation of IPA, it is necessary to
take into account those indicators that will
contribute to creating more comprehensive and
clearer picture of an organization state in terms of
services/products quality. Most organizations play
the market game with scarce resources. IPA can
help quality policy-makers in detecting those
elements of the services/products which allocation
of resources could contribute to more satisfied
users. As IPA has become a very popular tool, there
appeared an increasing number of simple and
complex modified methods. First of all, this method
needs to be and remain a simple assessment tool of
customer satisfaction and quality of
services/products. Each of these approaches has its
advantages and disadvantages, and it is therefore
difficult to say for any of the aforesaid modifications
of IPA that one is better than the other and that it
gives a more relevant picture of the organization.
The decision remains on managers to choose the
method that will be an effective and efficient way to
get to the desired information.

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
Although this is a comprehensive systematic review,
there are also some limitations, so it is
recommended that future studies take into account
all the previous researches related to IPA, from its
development, as well as extensive analysis of
methods for creating questionnaires and obtaining
information.
Acknowledgment
Authors of this paper would like to acknowledge Ms
Aida Šarić, prof., for the services of Academic
proofreading.
References
Abalo, J., Varela, J., & Manzano, V. (2007). Importance values for Importance–Performance Analysis: A formula
for spreading out values derived from preference rankings. Journal of Business Research, 60(2), 115-121.
Albayrak, T., & Caber, M. (2014). Symmetric and asymmetric influences of service attributes: The case of fitness
clubs. Managing Leisure, 19(5), 307-320.
Bacon, D. R. (2003). A comparison of approaches to importance-performance analysis. International Journal of
Market Research, 45(1), 55-73.
Chang, C.-C. (2014). An IPA-embedded model for evaluating creativity curricula. Innovations in Education and
Teaching International, 51(1), 59-71.
Chen, F.-Y., & Chang, Y.-H. (2005). Examining airline service quality from a process perspective. Journal of Air
Transport Management, 11(2), 79-87.
Deng, W.-J., Kuo, Y.-F., & Chen, W.-C. (2008). Revised importance–performance analysis: three-factor theory
and benchmarking. The Service Industries Journal, 28(1), 37-51.
Feng, C.-M., & Jeng, K.-Y. (2005). Analyzing airline service improvement strategy through importance and
performance analysis. Journal of the Eastern Asia Society for Transportation Studies, 6, 782-797.
Feng, M., Mangan, J., Wong, C., Xu, M., & Lalwani, C. (2014). Investigating the different approaches to
importance–performance analysis. The Service Industries Journal, 34(12), 1021-1041.
Hollenhorst, S., Olson, D., & Fortney, R. (1992). Use of importance-performance analysis to evaluate state park
cabins: the case of the West Virginia state park system. Journal of Park and recreation Administration, 10(1), 1-
11.
Hudson, S., Hudson, P., & Miller, G. A. (2004). The measurement of service quality in the tour operating sector:
A methodological comparison. Journal of travel Research, 42(3), 305-312.
Johns, N. (2001). Importance-performance analysis using the profile accumulation technique. Service Industries
Journal, 21(3), 49-63.
Joseph, M., Allbright, D., Stone, G., Sekhon, Y., & Tinson, J. (2005). Importance-performance analysis of UK
and US bank customer perceptions of service delivery technologies. International Journal of Financial Services
Management, 1(1), 66-88.
Kano, N., Seraku, N., Takahashi, F., & Tsuji, S. (1984). Attractive quality and must-be quality. The Journal of
the Japanese Society for Quality Control, 14(4), 39–48.
Kuo, Y.-F., Chen, J.-Y., & Deng, W.-J. (2012). IPA–Kano model: A new tool for categorising and diagnosing
service quality attributes. Total Quality Management & Business Excellence, 23(7-8), 731-748.
Levenburg, N. M., & Magal, S. R. (2004). Applying importance-performance analysis to evaluate e-business
strategies among small firms. E-service Journal, 3(3), 29-48.
Lin, S.-P., Chan, Y.-H., & Tsai, M.-C. (2009). A transformation function corresponding to IPA and gap analysis.
Total Quality Management, 20(8), 829-846.
Martilla, J. A., & James, J. C. (1977). Importance-performance analysis. The journal of marketing, 77-79.
Matzler, K., Bailom, F., Hinterhuber, H. H., Renzl, B., & Pichler, J. (2004). The asymmetric relationship between
attribute-level performance and overall customer satisfaction: a reconsideration of the importance–performance
analysis. Industrial marketing management, 33(4), 271-277.
Nale, R. D., Rauch, D. A., Wathen, S. A., & Barr, P. B. (2000). An exploratory look at the use of importance-
performance analysis as a curricular assessment tool in a school of business. Journal of Workplace Learning,
12(4), 139-145.
Oh, H. (2001). Revisiting importance–performance analysis. Tourism management, 22(6), 617-627.
Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). SERVQUAL: A multiple-item scale for measuring
customer perceptions of service quality. Journal of retailing, 64(1), 12-40.
Piñeiro, I. A., Boubeta, A. R., & Mallou, J. V. (2006). El análisis de importancia-valoración aplicado a la gestión
de servicios. Psicothema, 18(4), 730-737.
Rial, A., Rial, J., Varela, J., & Real, E. (2008). An application of importance-performance analysis (IPA) to the
management of sport centres. Managing Leisure, 13(3-4), 179-188.
Sooreh, L. K., Salamzadeh, A., Saffarzadeh, H., & Salamzadeh, Y. (2011). Defining and measuring
entrepreneurial universities: a study in Iranian context using importance-performance analysis and TOPSIS
technique. Global Business and Management Research: An International Journal, 3(2), 182-199.
Tam, M.-l., & Lam, W. H. (2004). Determination of service levels for passenger orientation in Hong Kong
International Airport. Journal of Air Transport Management, 10(3), 181-189.

Ormanović, Š. et. al.: Importance-performance analysis… Acta Kinesiologica 11 (2017) Supp. 2: 58-66
Tontini, G., Picolo, J. D., & Silveira, A. (2014). Which incremental innovations should we offer? Comparing
importance–performance analysis with improvement-gaps analysis. Total Quality Management & Business
Excellence, 25(7-8), 705-719.
Wade, D. J., & Eagles, P. F. (2003). The use of importance–performance analysis and market segmentation for
tourism management in parks and protected areas: An application to Tanzania's national parks. Journal of
Ecotourism, 2(3), 196-212.
Yavas, U., & Shemwell, D. J. (2001). Modified importance-performance analysis: an application to hospitals.
International Journal of Health Care Quality Assurance, 14(3), 104-110.
Yildiz, S. M. (2011). An importance-performance analysis of fitness centre service quality: Empirical results from
fitness centres in Turkey. African Journal of Business Management, 5(16), 7031.
Zhang, H. Q., & Chow, I. (2004). Application of importance-performance model in tour guides’ performance:
evidence from mainland Chinese outbound visitors in Hong Kong. Tourism management, 25(1), 81-91.
IMPORTANCE-PERFORMANCE ANALIZA: RAZLIČITI PRISTUPI
Sažetak
U današnjem okruženju brzog razvoja tehnologije i globalne tržišne konkurencije u kojem je brzina ključna
karakteristika opstanka organizacija, pred menadžerima je postavljen težak zadatak: u moru obaveza, redovno
evaluirati nivo zadovoljstva korisnika usluga/proizvoda kao i ukupno stanje organizacije. Importance-
performance analysis IPA je praktična i svrsishodna metoda kojom menadžeri mogu brzo i kvalitetno dobijati
uvid u veliku sliku, dali su na dobrom putu ili su potrebne određene korektivne akcije.Svrha ovog rada je izvršiti
sveobuhvatan i sistematski pregled dosadašnje literature koja se bavila različitim pristupima i modifikacijama
IPA metoda te utvrditi benefite i nedostatke različitih pristupa.Selekcija relevantnih radova za ovu studiju je
vršena u EBSCO bazi podataka, uključivanjem ključnih riječi Importance, Preformance i Analysis. Od ukupnog
broja od 1075 dostupnih dosadašnjih istraživanja, u finalno razmatranje je ušlo 13 radova koji su zadovoljavali
kriterije ove sistematske analize. Ovom studijom je ustanovljeno da je tradicionalni IPA baziran na pogrešnim
temeljnim predpostavkama, pa su stoga rezultati dobiveni ovom metodom dovedeni u pitanje. Po pitanju
grafikona IPA, dijagonalni model (DM) u kombinaciji sa Data centered quadrant model (DCQM) je metod koji ima
prednost u odnosu na kvadrant model (QM) i Scale centered quadrant model (SCQM). Jos jedan veoma dobar
prijedlog je transformacija funkcije osa koordinata, gdje se za presjek koordinata koriste fiksne vrijednosti.
Podaci zadovoljstva korisnika dobiveni direktnom metodom sa sobom nose pojavu svjesne sklonosti ispitanika te
se preporučuje indirektna metoda procjene dimenzija importance i performance. IPA može kreatorima strategija
kvaliteta pomoći u detektovanju onih elemenata usluga/proizvoda čija alokacija resursa može doprinjeti
zadovoljnijim korisnicima. Svaki od analiziranih modifikovanih pristupa ima svoje prednosti i nedostatke te na
menadžerima ostaje odluka da izaberu onaj metod kojim će na efektivan i efikasan način doći do relevantnih
informacija.
Key words: Zadovoljsto, Gap, TOPSIS, Teorija tri faktora, PAT, KANO
Corresponding information:
Received: 04 October 2017
Accepted: 22 December 2017
Correspondence to: Mr. Šemso Ormanović
University of Sarajevo
Faculty of Sport and Physical Education
Phone:
E-mail:osemso@hotmail.com