This a icle is dis ibu ed unde he e ms o he
C ea i e Commons A ibu ion 4.0 In e nacional License
ins i u ua
enp esa
Ins i u o de Economía Aplicada a la Emp esa
Managemen Le e s / Cuade nos de Ges ión
jou nal homepage: h p://www.ehu.eus/cuade nosdeges ion/ e is a/es/
ISSN: 1131-6837 / e-ISSN: 1988-2157
Managemen Le e s
Cuade nos de Ges ión
Enp esa Ins i u ua, UPV/EHU
Conocimien o en Ges ión/Managemen Knowledge
Volume 22 / Numbe 2 (2022) • ISSN: 1131-6837 / e-ISSN: 1988-2157
h p://www.ehu.eus/cuade nosdeges ion/ e is a/es/
Chu n in se ices – A bibliome ic e iew
Abandono en se icios - Una e isión bibliomé ica
Hugo Ribei o*, Belém Ba bosa
a, b, c
, An ónio C. Mo ei a
b, d, e,
, Rica do Rod igues
e, g
a Uni e si y o Po o. School o Economics and Managemen – belem@ ep.up.p –h ps://o cid.o g/0000-0002-4057-360X
b GOCVOPP - Resea ch Uni on Go e nance, Compe i i eness and Public Policies, Uni e si y o A ei o, Po ugal
c ce .UP – Cen e o Economics and Finance a UPo o, Uni e si y o Po o, Po ugal
d A ei o Uni e si y. Depa men o Economics, Managemen , Indus ial Enginee ing, and Tou ism – amo ei [email protected] – h ps://o cid.o g/0000-0002-6613-8796
e NECE-UBI - Resea ch Cen e o Business Sciences, Uni e sidade da Bei a In e io , Po ugal
INESCTEC—Ins i u e o Sys ems and Compu e Enginee ing, Technology and Science, Faculdade de Engenha ia da Uni e sidade do Po o, Po ugal
g Uni e sidade da Bei a In e io , Depa men o Business and Economics – g od [email protected] – h ps://o cid.o g/0000-0001-6382-5147
* Co esponding au ho : A ei o Uni e si y. Depa men o Economics, Managemen , Indus ial Enginee ing, and Tou ism–hugo. ibei [email protected] –h ps://o cid.o g/0000-
0002-0410-6430
ARTICLE INFO
Recei ed 01 June 2021,
Accep ed 29 Decembe 2021
A ailable online 4 Ma ch 2022
DOI: 10.5295/cdg.211509h
JEL: M30, M31
ABSTRACT
The pu pose o his a icle is o iden i y he mos impac ul esea ch on cus ome chu n and o map he concep ual
and in ellec ual s uc u e o i s ield o s udy. Da a we e collec ed om he WoS da abase, comp ising 338 a icles
published be ween 1995 and 2020. Se e al bibliome ic echniques we e applied, including analysis o co-wo ds,
co-ci a ion, bibliog aphic coupling, and co-au ho ship ne wo ks. R so wa e and he Bibliome ix/Biblioshiny
package we e used o pe o m he analyses. The esul s iden i y he mos ac i e and in luen ial au ho s, a icles,
and jou nals on he opic. Mo e speci ically, h ough co-ci a ions and bibliog aphic coupling, i was possible o
map he oldes a icles ( e ospec i e analysis) and he cu en esea ch on (p ospec i e analysis). The e o-
spec i e analysis, based on co-ci a ions, e ealed ha he ounda ions o his esea ch ield a e cons uc s such
as quali y o se ice, sa is ac ion, loyal y, and changing beha io s. The p ospec i e analysis, pe o med h ough
bibliog aphic coupling, e ealed ha cu en esea ch is embedded in p edic i e analysis, clus e s, da a mining,
and algo i hms. The esul s p o ide obus guidance o u he in es iga ion in his ield.
Keywo ds: Cus ome Chu n, Bibliome ic Analysis, Co-ci a ion Analysis, Bibliog aphic Coupling, Science Map-
ping, Biblioshiny.
RESUMEN
El obje i o de es e a ículo es iden i ica las in es igaciones más impac an es sob e la pé dida de clien es y aza
la es uc u a concep ual e in elec ual de su campo de es udio. Los da os han sido ecogidos de la base de da os
WoS, que comp enden 338 a ículos publicados en e 1995 y 2020. Va ias écnicas bibliomé icas ue on aplicadas,
incluyendo el análisis de co-palab as, coci aciones, acoplamien o bibliog á ico y edes de coau o ía. Pa a ealiza
los análisis se u iliza on el so wa e R y el Bibliome ix/Biblioshiny. Los esul ados iden i ican los au o es, a ículos
y e is as más in luyen es y ac i os sob e el ema. Más especí icamen e, a a és de las coci aciones y el acoplam-
ien o bibliog á ico, ue posible mapea los a ículos más an iguos (análisis e ospec i o) y la in es igación más
ac ual (análisis p ospec i o). El análisis e ospec i o, basado en las coci aciones, e eló que los undamen os de
es e campo de in es igación son cons uc os como la calidad del se icio, la sa is acción, la leal ad y el cambio de
compo amien os. El análisis p ospec i o, ealizado a a és del acoplamien o bibliog á ico, e eló que la in es i-
gación ac ual es á inme sa en el análisis p edic i o, los conglome ados, la mine ía de da os y los algo i mos. Los
esul ados p opo cionan una sólida o ien ación pa a segui in es igando en es e campo.
Palab as cla e: Chu n de Clien es, Análisis Bibliomé ico, Análisis de Coci ación, Acoplamien o Bibliog á ico,
Mapeo de la Ciencia, Biblioshiny.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
98 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
1. INTRODUCTION
Cus ome chu n is one o he mos challenging opics o
manage s in se e al sec o s in ol ing con ac ed se ices,
whe e cus ome s end o change se ice p o ide s epea edly
(Kuma e al., 2018). Also known as cus ome u no e , and
cus ome de ec ion o depa u e, cus ome chu n e e s o a
cus ome ’s decision o cease business wi h a se ice p o ide
(Adebiyi e al., 2016; Eshghi e al., 2007; Mahajan e al., 2015;
P ince & G eens ein, 2014), by swi ching o a new p o ide .
The e a e se e al clea indica ions o he ele ance o his
opic. Being widely accep ed ha a ac ing new cus ome s by
d i ing hem away om a compe i o is gene ally mo e ex-
pensi e han e aining cu en cus ome s by mee ing hei eal
needs (Kyei & Bayoh, 2017), he li e a u e u he s esses ha
highe e en ion a es esul in highe ma ke sha e, esul -
ing in highe e enues (Kyei & Bayoh, 2017). Pa icula ly in
ma ke s eaching high le els o sa u a ion and concen a ion,
inding and e aining new cus ome s is inc easingly di icul
and expensi e (Ca izo-Mo ei a e al., 2017; Hadden e al.,
2007; Mo ei ae al., 2016). Fo ha eason, companies such
as banks, elecommunica ions and ai lines, o name bu a ew,
use cus ome chu n o e en ion a e as a c i ical business me -
ic (Ami i & Daume III, 2016). In his connec ion, he li e a-
u e sugges s ha highe p io i y should be gi en o e aining
he mos aluable exis ing cus ome s a he han gaining new
ones (Ca izo-Mo ei a e al., 2017; Hadden e al., 2007; Mo ei-
a e al., 2016), which leads o a change in business pa adigm,
wi h a new emphasis on cus ome e en ion and ela ionship
de elopmen , a he han looking mainly a cus ome acqui-
si ion.
Unde s anding and p e en ing cus ome chu n, pa icu-
la ly by p o i able cus ome s, is c i ical o business su i al
(Adebiyi e al., 2016), including iden i ying he cus ome s who
a e mos likely o swi ch se ice p o ide s (Amin e al., 2019).
The de e minan s o cus ome chu n ha e been he ocus o a
signi ican s eam o esea ch, emaining cen al cons uc s in
ma ke ing ac i i ies (Schweidel e al., 2008). Kea eney (1995)
was one o he i s au ho s o s udy cus ome chu n, and
ound ha i s main causes included p ice, se ice ailu e, and
he company’s esponses o se ice p oblems. Rajan (2017)
added ha cus ome u no e may be based on dissa is ac-
ion, highe cos s, low quali y, lack o esou ces, and p i acy
conce ns. Sa is ac ion s ands ou as a p edic o o cus ome
e en ion (Ande son e al., 1994; Eshghi, Haugh on, & Topi,
2007), bu i also causes cus ome s o swi ch se ice p o ide s
(Becke e al., 2015), namely in esponse o posi i e wo d-o -
mou h om compe i o s’ mos sa is ied cus ome s (de Haan
e al., 2015).
Despi e he s eep inc ease in published a icles on cus ome
chu n, no one has ye p o ided a summa ized e iew o he
scien i ic landscape, p e en ing a clea iew o he s a e o he
a . To he bes o ou knowledge, no wo k so a has ocused on
analyzing he de elopmen o scien i ic p oduc ion on cus om-
e chu n. In o de o add ess his gap and p esen in o ma ion
on how chu n has been add essed in he li e a u e, his a icle
uses a bibliome ic analysis.
Once a scien i ic discipline has eached a ce ain deg ee o
ma u i y, i is common p ac ice o esea che s o ocus hei
a en ion on he li e a u e gene a ed by he scien i ic commu-
ni y o conduc li e a u e e iews o assess he s a e o he a
(Ramos-Rod iguez & Ruiz-Na a o, 2004). Indeed, syn hesizing
he esul s o pas esea ch is one o he mos c i ical asks o
ad ancing knowledge in a pa icula esea ch opic (Zupic & Ca-
e , 2015), namely by adop ing bibliome ic esea ch me hods o
map he s uc u e and de elopmen o scien i ic ields and disci-
plines (Zupic & Ca e , 2015). Speci ically, his echnique exam-
ines how disciplines, ields, expe ise, and indi idual documen s
and au ho s ela e (Small, 1999; Zupic & Ca e , 2015). Bibliome -
ic me hods adop a quan i a i e app oach o desc ibe, e alua e,
and moni o published esea ch, de e mining i s cogni i e s uc-
u e and e olu ion (Small, 1999). They ollow a sys ema ic, ans-
pa en and eplicable e iew p ocess (Zupic & Ca e , 2015) ha
sys ema ically ep esen s he na u e o speci ic scien i ic disci-
plines, highligh ing esea ch ends (Zhang e al., 2016). As such,
hey iden i y majo esea ch a eas, p o iding esea che s wi h a
solid basis o posi ioning signi ican cu en con ibu ions and
de ec ing new a enues o u u e esea ch (Fe ei a, 2018).
Following he bibliome ic analysis app oach, his a icle
aims o iden i y he mos impac ul esea ch on cus ome chu n
and o map he concep ual and in ellec ual s uc u e o i s ield
o s udy. I is in ended h ough his bibliome ic analysis o an-
swe he ollowing esea ch ques ions:
— Wha a e he speci ic opics associa ed wi h cus ome chu n
esea ch?
— Wha is he in ellec ual s uc u e o he ield?
— Who a e he cen al, pe iphe al, o b idging esea che s in his
ield?
— Wha is he in ellec ual s uc u e o ecen /eme ging li e a-
u e?
— Wha is he social s uc u e o he ield?
This a icle makes se e al con ibu ions o de elopmen o
he li e a u e on cus ome chu n in se ices. Fi s ly, i desc ibes
he s uc u e o he concep ual ield h ough co-wo d analysis
and maps he in ellec ual ield h ough a co-ci a ion analysis
o au ho s, a icles, and jou nals. Secondly, i iden i ies and o -
ganizes he mos ecu en hemes and cu ing-edge esea ch
h ough a bibliog aphic coupling analysis, which shows how
he opic is de eloping. Finally, i iden i ies he social s uc u e
o he esea ch ield. Hence, his a icle p o ides schola s wi h
guidance o u u e esea ch, by highligh ing he mos p omi-
nen con ibu ions on he opic and by iden i ying he ends in
his ield o esea ch. This app oach e eals ha he li e a u e is
a ied and co e s opics such as de ec ion, e en ion, cus om-
e chu n, and swi ching beha io , encompassing di e en and
some imes complemen a y concep s. I is also wo h no ing ha
cus ome expe ience, disappoin men , dese ion, cus ome en-
coun e and sa is ac ion, lack o se ice quali y and a ibu es a e
among he main de e minan s o chu ning. Finally, i is impo -
an o men ion ha i p edic i e models a e used ex ensi ely o
analyze chu n, beha io al models a e also g ea ly used o un-
de s and wha leads cus ome s o swap one se ice p o ide o
ano he .
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 99
2. BIBLIOMETRIC ANALYSIS OF THE LITERATURE
Bibliome ic me hods ( o example, co-ci a ion analysis,
bibliog aphic coupling, analysis o co-au ho s, among o he s)
use bibliog aphic da a om publica ion da abases o cons uc
s uc u al images o scien i ic ields (Zupic & Ca e , 2015), dis-
co e ing hei essence (P i cha d, 1969). Bibliome ic me hods
ha e wo p ima y uses: pe o mance analysis and scien i ic map-
ping (Cobo e al., 2011). Pe o mance analysis seeks o assess
he pe o mance o esea ch and publica ion by indi iduals and
ins i u ions. Scien i ic mapping aims o e eal he s uc u e and
dynamics o scien i ic ields (Zupic & Ca e , 2015).
Di e en app oaches ha e been applied o ex ac ne wo ks
using di e en uni s o analysis (au ho s, documen s, jou nals,
and e ms). Table1 shows he echniques ha will be applied
h oughou his documen .
Table 1
Mos common bibliome ic echniques by he uni o analysis
Bibliome ic echnique Uni o analysis used Kind o ela ion
Bibliog aphic
Coupling
Au ho Au ho ’s oeu es Common
e e ences among
au ho ’s oeu es
Documen Documen Common
e e ences among
documen s
Jou nal Jou nal’s oeu es Common
e e ences among
jou nal’s oeu es
Coau ho Au ho Au ho ’s name Au ho s’ co-
occu ence
Coun y Coun y om
a ilia ion
Coun ies’ co-
occu ence
Ins i u ion Ins i u ion om
a ilia ion
Ins i u ions’ co-
occu ence
Co-ci a ion Au ho Au ho ’s e e ence Co-ci ed au ho
Documen Re e ence Co-ci ed
documen s
Jou nal Jou nal’s e e ence Co-ci ed jou nal
Co-wo d Keywo d, o e m
ex ac ed om
i le, abs ac , o
documen ’s body
Te ms’ co-
occu ence
Sou ce: Adap ed om Cobo e al. (2011).
2.1. Resea ch me hodology and a icle selec ion
Two ypes o esea ch objec i es can be de ined by using he
bibliog aphy o scien i ic mapping: (1) o iden i y he knowl-
edge base o a esea ch opic o ield and i s in ellec ual s uc u e;
(2) o iden i y he main hemes and ends (concep ual s uc-
u e). The analysis ca ied ou in his s udy se es o iden i y he
concep ual and in ellec ual s uc u es o he esea ch ield. To
iden i y he exis ence o clus e s o a icles on cus ome chu n,
he mos in luen ial au ho s, hei geog aphical o igin, and au-
ho ship ne wo ks. To cons uc he bibliome ic maps, he so -
wa e used was Biblioshiny (A ia & Cuccu ullo, 2017), om he
R Co e Team (Team, 2021) Bibliome ix package. R is an open-
sou ce p og amming language, c ea ing a so wa e ecosys em ac-
cessible o he whole communi y. Indeed, all esou ces a e sha ed
by he communi y, and all knowledgeable use s can con ibu e
o he de elopmen o di e en so wa e packages. Biblioshiny is
an R so wa e package ha p o ides a g aphical en i onmen o
using he Bibliome ix package. The Bibliome ix package allows
he analysis and mapping o bibliog aphic da a.
Rega ding da a collec ion, and o ensu e ha all ele an a i-
cles we e conside ed in his s udy, we ini ially pe o med a sea ch
on he WoS1 da abase wi h he main keywo d “cus ome chu n”
and es ic ing he sea ch o a icles in English published un il
2020. WoS is by a he mos common sou ce o bibliog aphic da a
(Zupic & Ca e , 2015), wi h he oldes and mos comp ehensi e e-
co ds o ci a ion indexes (Ellegaa d & Wallin, 2015). This da abase
con ains enough da a o mos bibliome ic analyses. Da a, includ-
ing a icle i le, a icle ype, au ho s, hei ins i u ional a ilia ions,
keywo ds, abs ac , numbe o ci a ions, jou nal name, publishe
name and add ess, yea o publica ion, olume, issue numbe , and
a lis o ci ed e e ences a e a ailable o analysis (Zupic & Ca e ,
2015). This sea ch iden i ied 253a icles, whose keywo ds and ab-
s ac s we e ex ac ed and subjec o con en analysis using NVI-
VO2 quali a i e analysis so wa e. This p ocedu e allowed us o
iden i y synonyms, al e na i e ph asing, and combina ions o he
wo d “chu n” wi h o he wo ds ha could help iden i y addi ional
a icles on he opic. Con en analysis so wa e such as NVIVO
was conside ed help ul because i can explo e wo d sea ch and
analyze wo d equency. As a esul , addi ional keywo ds we e
iden i ied o his s udy, including “cus ome u no e ”, “cus ome
swi ching”, “chu n managemen ”, “chu n ac o s”, and “cus ome
de ec ion”. The esul ing se o keywo ds is p esen ed in Table2.
Table 2
Keywo ds used in he sea ch
Keywo ds
cus ome chu n
cus ome u no e
cus ome a i ion
cus ome o a ion
cus ome de ec ion
cus ome swi ching
consume swi ching beha io
cus ome swi ching beha io
s aye s * swi che s
chu n managemen
chu n de e minan s
chu n ac o s
chu n analysis
de ec ion managemen
Sou ce: Au ho ’s own elabo a ion.
1 h p://www.webo knowledge.com
2 h ps://www.qs in e na ional.com/n i o-quali a i e-da a-analysis-so -
wa e/home
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
100 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
A e iden i ying all he keywo ds, a new sea ch was ca -
ied ou in WoS, always using he Boolean ope a o “o ” be-
ween hem. The sea ch was based on he a icles’ i le, abs ac
and keywo ds. Only jou nal a icles we e conside ed, because
his ype o publica ion is a guably conside ed mos high-
ly by academics and p ac i ione s, due o ep esen ing mos
pee - e iewed and also he mos ci able con ibu ions. Hence,
con e ence p oceedings, books, book e iews, o con e ence
abs ac s we e no conside ed. In addi ion, only a icles w i en
in English and published up o 2020 we e conside ed, e u n-
ing 452a icles. Da a we e ex ac ed and analyzed in he i s
semes e o 2021.
Figu e 1
Sea ch lowcha used in his in es iga ion
Sou ce: Au ho ’s own elabo a ion.
Figu e1 p esen s he sea ch lowcha , c ea ed using he ool
de eloped by Haddaway and McGuinness (2021), desc ibing he
a ious s ages o sc eening and he numbe o a icles excluded
om he analysis. Fi s ly, all abs ac s we e ead and analyzed in
de ail. As a esul , 114a icles we e excluded, ei he because hey
we e ou side he scope o his s udy o because hey add essed
o he esea ch opics un ela ed o cus ome chu n. Hence, only
a icles ha had cus ome chu n as he objec o s udy we e in-
cluded. Consequen ly, 338a icles we e conside ed o his s udy.
In ela ion o bibliome ic me hods, di e en choices we e made
ega ding he ne wo k layou , da a no maliza ion, and he clus e ing
algo i hm o he analysis o co-wo ds, co-ci a ions, bibliog aphic
coupling, and co-au ho ship ne wo ks. Table3 p esen s he choices
made o each analysis ypology, as p esen ed in he nex sec ions.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 101
Table 3
Ne wo k Layou , No maliza ion and Clus e ing Algo i hm
Analysis Numbe o Nodes Minimum edges Ne wo k Layou No maliza ion Clus e ing Algo i hm
Co-Wo d 50 2 Kamada & Kawai Associa ion Walk ap
Co-Ci a ion & Bibliog aphic
Coupling
Au ho s 50 2 Kamada & Kawai N/A Walk ap
A icles 50 2 Kamada & Kawai N/A Walk ap
Jou nals 50 2 Kamada & Kawai N/A Walk ap
Coau ho
Au ho s 50 2 F uch e man & Reingold Associa ion Walk ap
Coun y o A ilia ion 50 2 F uch e man & Reingold Associa ion Walk ap
Ins i u ion om a ilia ion 50 2 F uch e man & Reingold Associa ion Walk ap
Sou ce: Au ho ’s own elabo a ion.
The da a we e no malized using associa ion s eng h as a
simila i y measu e. Acco ding o he heo e ical and empi ical
esul s o Eck and Wal man (2009), his measu e was consid-
e ed he mos app op ia e o no malize da a co-occu ence.
A clus e ing algo i hm was used o di ide he o e all ne wo k
in o di e en sub-ne wo ks o pe o m communi y de ec ion
(Cobo e al., 2011). A communi y, o clus e , is gene ally de-
ined as a subse o densely in e connec ed nodes ela i e o
he es o he ne wo k (Newman & Gi an, 2004). Walk ap,
de eloped by Pons and La apy (2005), was he algo i hm used.
As explained by i s au ho s, Walk ap o e s a measu e o sim-
ila i y be ween e ices based on andom walks, wi h se e al
impo an ad an ages: i cap u es he communi y s uc u e o
a ne wo k well, can be e icien ly compu ed, and can be used
in an agglome a ion algo i hm o compu e e icien ly he com-
muni y s uc u e o a ne wo k. This communi y de ec ion al-
go i hm was ound o ha e one o he bes esul s in iden i ying
communi ies e en o high alues o mixing coe icien (O -
man & Laba u , 2009).
The nex s ep was o selec co-occu ences, in ol ing ne -
wo k il e ing using wo as a minimum edge alue educ ion.
Only edges wi h a s eng h g ea e han o equal o wo a e
plo ed. In all analyses, only he 50 mos ci ed e e ences we e
plo ed.
2.2. Desc ip i e Analysis
We s a ed by pe o ming a desc ip i e analysis o he da a-
base used, o summa ize and explo e he da a. Table 4 shows he
main s a is ics o he da abase. The pe iod o analysis is om
1995 o 2020, p esen ing an annual g ow h a e o 17.84%. The
338a icles on he da abase ha e an a e age ci a ion o 33.24.
The a icles we e w i en by 796 au ho s, wi h an a e age o
2.36au ho s pe a icle.
Table 4
Main in o ma ion abou da a
Desc ip ion
A icles 338
Pe iod 1995 - 2020
Annual Pe cen age G ow h Ra e 17,84%
A e age ci a ions pe a icle 33,24
Au ho s 796
Au ho s Appea ances 995
Au ho s o single-au ho ed a icles 34
Au ho s o mul i-au ho ed a icles 762
Au ho s pe A icle 2,36
Co-Au ho s pe A icles 2,94
Collabo a ion Index 2,53
Sou ce: Au ho ’s own elabo a ion.
As shown in Figu e2, he numbe o a icles published be o e
2009 was ela i ely small, bu publica ions inc ease a e ha pe iod.
0
5
10
15
20
25
30
35
40
1990199520002005201020152020
Numbe o publish a icles
Yea
Figu e 2
Publishing end in he a ea o cus ome chu n
Sou ce: Au ho ’s own elabo a ion.
No e: Numbe o published a icles pe yea .
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
102 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
The ini ial s a is ics show ha he 338 a icles we e pub-
lished in 177 di e en jou nals. The op 10jou nals we e e-
sponsible o 110 a icles, abou 33% o he o al. Table5 shows
he leading jou nals whe e hese a icles we e published. “Ex-
pe Sys ems wi h Applica ions” s ands ou , wi h 12% o all
publica ions.
Table 5
The op 10 publishing jou nals con ibu ing o he a ea
o cus ome chu n
Sou ces No. o A icles % o A icles
Expe Sys ems wi h Applica ions 41 12
Eu opean Jou nal o Ope a ional
Resea ch 9 3
Eu opean Jou nal o Ma ke ing 8 2
Ma ke ing Science 8 2
Jou nal o Ma ke ing Resea ch 7 2
Jou nal o Se ice Resea ch 7 2
Telecommunica ions Policy 7 2
Decision Suppo Sys ems 6 2
Jou nal o The Academy o Ma ke ing
Science 6 2
In e na ional Jou nal o Ad anced
Compu e Science and Applica ions 5 1
Sou ce: Au ho ’s own elabo a ion.
Table6 p esen s he op 10 au ho s and he numbe o a icles
o which hey a e au ho s o coau ho s. Van Den Poel D, Baes-
ens B, and Coussemen , K ha e he g ea es numbe o a icles.
These h ee au ho s ha e hei publica ions mos ly in “Ope a-
ions Resea ch Managemen Science” and “Compu e Science
A i icial In elligence”.
Table 6
The op 10 con ibu ing au ho s and numbe o a icles
Au ho Numbe o published a icles
Van Den Poel D 13
Baesens B 12
Coussemen K 8
Ve beke W 6
Amin A 5
Anwa S 5
Liu Y 5
Ma ens D 5
Sou ce: Au ho ’s own elabo a ion.
Table 7 shows he op10 mos ci ed a icles. Those wi h
he highes numbe o ci a ions a e (1)Cus ome swi ching be-
ha iou in se ice indus ies: An explo a o y s udy (Kea eney,
1995); (2)Consume swi ching cos s: A ypology, an eceden s,
and consequences (Bu nham e al., 2003) and (3) Swi ching
ba ie s and epu chase in en ions in se ices (Jones e al.,
2000).
Table 7
Top 10-Mos ci ed a icles (Global)
A icle To al Global
Ci a ions (TC) TC pe Yea
Kea eney Sm, 1995, J Ma k 1154 42,74
Bu nham Ta, 2003, J Acad Ma k Sci 756 39,79
Jones Ma, 2000, J Re ail 648 29,45
Bansal Hs, 2004, J Acad Ma k Sci 427 23,72
Chen Py, 2002, In Sys Res 339 16,95
Aydin S, 2005, Eu J Ma ke 260 15,29
Neslin Sa, 2006, J Ma k Res 238 14,88
Kea eney Sm, 2001, J Acad Ma k Sci 237 11,29
Wei Cp, 2002, Expe Sys Appl 198 9,90
Bu ez J, 2009, Expe Sys Appl 189 14,54
Sou ce: Au ho ’s own elabo a ion.
I is also in e es ing o analyze he op 10 mos ci ed a icles
locally: a icles ha ecei ed ci a ions om documen s con-
ained only in he da abase s udied, as shown in Table 8. The
mos ci ed a icles a e (1)New insigh s in o chu n p edic ion
in he elecommunica ion sec o : A p o i -d i en da a mining
app oach (Ve beke e al., 2012); (2)Tu ning elecommunica-
ions call de ails o chu n p edic ion: a da a mining app oach
(Wei & Chiu, 2002); and (3)Cus ome a i ion analysis o i-
nancial se ices using p opo ional haza d models (Van den
Poel & La i ie e, 2004), wi h he exac quo es ha (3)Cus om-
e base analysis: pa ial de ec ion o beha io ally loyal clien s
in a non-con ac ual FMCG e ail se ing (Buckinx & Van den
Poel, 2005).
Table 8
Top 10-Mos ci ed a icles (Local)
A icle To al Local Ci a ions (TC)
Ve beke W, 2012, Eu J Ope Res 55
Wei Cp, 2002, Expe Sys Appl 54
Van Den Poel D, 2004, Eu J Ope Res 44
Buckinx W, 2005, Eu J Ope Res 44
Bu ez J, 2009, Expe Sys Appl 43
Tsai C , 2009, Expe Sys Appl 39
Ve beke W, 2011, Expe Sys Appl 39
Ahn Jh, 2006, Telecommun Policy 35
Bu ez J, 2007, Expe Sys Appl 34
Hadden J, 2007, Compu Ope Res 27
Sou ce: Au ho ’s own elabo a ion.
Rega ding he au ho s’ a ilia ion, Table9 shows he o gani-
za ions con ibu ing mos , based on he numbe o a icles pub-
lished. Compa ing his lis wi h he lis o he op10 au ho s,
in Table6, he Ca holic Uni e si y o Leu en, he Uni e si y o
Ghen , and he Uni e si y o Sou hamp on a e ep esen ed by
he mos p oli ic au ho s, Van den Poel, Di k (Ghen Uni e si-
y) and Baesens, Ba (Uni e si y o Sou hamp on and Ca holic
Uni e si y o Leu en).
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 103
Table 9
The op 20 con ibu ing o ganiza ions
O ganiza ion Loca ion A icles
Ka holieke Uni Leu en Belgium 31
Uni e si y Ghen Belgium 21
Uni e si y Sou hamp on Uni ed Kingdom 16
Ins Managemen Sci Pakis an 10
Uni e si y Ca holique Lille F ance 10
Uni e si y Teh an I an 10
Deakin Uni e si y Aus alia 8
Iq a Na ional Uni e si y Pakis an 8
Uni e si y Coll Dublin I eland 8
Geo gia S a e Uni e si y Uni ed S a es 7
Hong Kong Poly ech
Uni e si y
China 7
King Abdulaziz Uni e si y Saudi A abia 7
K.N. Toosi Uni e si y
Technology
I an 7
Uni e si y O iedo Spain 7
U ah S a e Uni e si y Uni ed S a es 7
Columbia Uni e si y Uni ed S a es 6
Helwan Uni e si y Egyp 6
Kuhne Logis ic Uni e si y Ge many 6
Lebanese Ame ican Uni e si y Lebanon 6
Na ional Cen al Uni e si y Taiwan 6
Sou ce: Au ho ’s own elabo a ion.
Figu e3 shows he a icles published by coun y. The in en-
si y o he colo is p opo ional o he numbe o publica ions.
In gene al, he geog aphical dispe sion indica es ha esea ch
and p ac ice ega ding cus ome chu n ha e a ac ed wo ldwide
a en ion.
Figu e 3
Scien i ic P oduc ion by Coun y
Sou ce: Au ho ’s own elabo a ion-Biblioshiny ou pu .
No e: The shade o blue ep esen s he p oduc ion o a icles pe
coun y. The da ke he colo , he g ea e he p oduc ion.
Table10 shows he op10 coun ies by publica ions, consid-
e ing he co esponding au ho , wi h China, he Uni ed S a es,
and Belgium accoun ing o 41% o he o al.
Table 10
Top 10 Mos p oduc i e coun ies
(based on i s au ho ’s a ilia ion)
Coun y Numbe o published a icles % o A icles
China 53 14
USA 51 13
Belgium 25 6
Ko ea 19 5
Pakis an 16 4
India 15 4
I an 14 4
Aus alia 11 3
F ance 10 3
Ge many 9 2
Sou ce: Au ho ’s own elabo a ion.
The mos equen ly used wo ds/ph ases we e also analyzed.
The analysis was pe o med o he wo ds/ph ases de ined by he
o iginal au ho s (Au ho Keywo ds) and he wo ds/ph ases de-
ined au oma ically by a compu e ized algo i hm (Keywo dsPlus).
These wo ds o ph ases ha appea de ined by he algo i hm a e
equen ly p esen in he i les o he e e ences o an a icle and
no necessa ily in he i le o he a icle o he au ho ’s keywo ds
(Ga ield & She , 1993). The Keywo dsPlus algo i hm can cap u e
he con en o a icles wi h g ea e dep h and a ie y (Ga ield &
She , 1993). I is as e ec i e as he wo ds de ined by he au ho s
in e ms o bibliome ic analysis in es iga ing he knowledge
s uc u e o scien i ic a eas. S ill, i is less comp ehensi e in ep-
esen ing he con en o an a icle (Zhang e al., 2016). The op20
au ho s’ wo ds can be seen in Table11, and he op20 o he wo ds
gene a ed by he Keywo dsPlus algo i hm in Table12.
Table 11
Top 20-Mos equen Au ho Keywo ds
Au ho Keywo ds (DE) No. o
A icles Au ho Keywo ds (DE) No. o
A icles
Chu n P edic ion 66 Cus ome Chu n
P edic ion
14
Da a Mining 47 Cus ome Loyal y 14
Cus ome Chu n 45 Telecommunica ions 12
Classi ica ion 29 Chu n Analysis 11
Cus ome Rela ionship
Managemen
28 Cus ome De ec ion 11
Cus ome Re en ion 26 Logis ic Reg ession 11
Chu n 22 Sa is ac ion 11
Machine Lea ning 19 Telecommunica ion 11
Cus ome 18 Ma ke ing 10
Cus ome Sa is ac ion 15 P edic ion 10
Sou ce: Au ho ’s own elabo a ion.
Rega ding he wo ds de ined by he au ho s, “Cus ome
Chu n”, “Cus ome De ec ion” and “Chu n Analysis” should be
ead wi h due con ex ualiza ion since hey we e pa o he se o
e ms used o build he esea ch.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
104 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
Table 12
Top 20-Mos equen Keywo ds-Plus
Keywo ds-Plus (ID) No. o
A icles Keywo ds-Plus (ID) No. o
A icles
Sa is ac ion 74 De e minan s 28
Re en ion 63 De ec ion 27
Model 60 Quali y 25
Loyal y 52 Managemen 24
Beha io 39 Cus ome Chu n 23
Se ices 38 Selec ion 22
Models 37 Algo i hm 20
P edic ion 37 A i ion 20
Classi ica ion 33 Cus ome Sa is ac ion 18
Impac 32 Dynamic-Model 18
Sou ce: Au ho ’s own elabo a ion.
F om analysis o he wo ables, e ms linked o da a science
—“Chu n P edic ion”, “Da a Mining”, “Classi ica ion”, “Machine
Lea ning”, “P edic ion”, “Model”, “Algo i hm” and “Selec ion”—
s and ou , ep esen ing he p ominence o da a analysis and da a
mining in in es iga ing cus ome chu n.
I is also in e es ing o no e ha cus ome e en ion, sa is-
ac ion and loyal y, and p edic ion also appea a he op o he
mos e e enced wo ds, ei he by he o iginal au ho s o by he
Keywo dPlus algo i hm, e ealing he impo ance o hese con-
s uc s in in es iga ing cus ome chu n.
Figu es4 and 5 show he e olu iona y end o he op10
wo ds/ph ases. Conce ning he au ho s’ wo ds, he e is a
g owing end in “Chu n P edic ion”, “Machine Lea ning”, and
“Cus ome Chu n”, once again ein o cing he p e ious obse -
a ions.
Conce ning Keywo dPlus wo ds/ph ases, and he e o e wi h
a b oade spec um, we obse e adjacen esea ch ields wi h
signi ican g ow h, “sa is ac ion”, “ e en ion” and “beha io ”, bu
also he same e ms ha we e obse ed in he au ho s’ wo ds,
ela ed o da a analysis, “Model” and “P edic ion”.
Figu e 4
Timeline Wo d G ow h (Au ho ’s Keywo ds)
Sou ce: Au ho ’s own elabo a ion-Biblioshiny ou pu .
No e: Numbe o occu ences o he au ho keywo ds o e ime.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 105
Figu e 5
Timeline Wo d G ow h (Keywo dsPlus)
Sou ce: Au ho ’s own elabo a ion-Biblioshiny ou pu .
No e: Numbe o keywo d plus occu ences o e ime.
2.3. Concep ual s uc u e o he ield
The concep ual s uc u e allows unde s anding o wha is be-
ing said in he esea ch ield, he main hemes, and ends (A ia
& Cuccu ullo, 2017). I is o en used o unde s and he opics
co e ed by esea che s and iden i y he mos impo an and
mos ecen issues. The mos equen co-wo d analysis was used
o he concep ual s uc u e and o unde s and wha was being
analyzed in he esea ch ield.
A. Co-wo d analysis
Co-wo d analysis is a con en analysis echnique ha uses he
mos impo an wo ds o keywo ds in documen s, o es ablish
ela ionships and build a concep ual s uc u e o a esea ch ield,
and o iden i y he main concep s analyzed in a gi en ield (Callon
e al., 1991; Callon e al., 1983). I can be applied o he keywo ds,
abs ac s, o ex s in hei en i e y (A ia & Cuccu ullo, 2017). The
uni o analysis is usually a keywo d o e m ex ac ed om he
i le, abs ac , o body o he documen (A ia & Cuccu ullo, 2017).
In scien i ic mapping, a ne wo k g aph is used o ep esen
co-occu ences be ween bibliog aphic me ada a. The g aph is
made up o nodes o poin s (each node is a wo d) connec ed by
lines. The size o each node is p opo ional o he occu ence
o he i em, and he size o he edges o he lines is p opo ion-
al o hei co-occu ence. The colo s ep esen he g oups o
which each wo d belongs. Co-occu ences can be no malized
using simila i y measu es o ob ain simila i ies be ween he da a
(Cobo e al., 2011).
Figu e6 shows he ne wo k diag am esul ing om he au-
ho s’ co-wo ds analysis.
Two majo g oups/clus e s o wo ds can be dis inguished.
The i s g oup deals wi h cus ome chu n based on p edic i e
analysis, ea u e selec ion, clus e ing, da a mining, and algo-
i hms such as decision ees, andom o es s, logis ic eg es-
sion, and suppo ec o machine. Also appea ing in his g oup
a e wo ds such as big da a, business in elligence, and elecom.
The second g oup o wo ds is ela ed o cus ome e en ion, sa -
is ac ion, loyal y, cus ome ela ionship managemen , se ice
quali y, change in en ions, and beha io . Rela ional ma ke ing,
us , and swi ching cos s also igu e in his g oup. These wo
clus e s show ha he esea ch ield is essen ially sus ained by
wo s eams o esea ch, he mos ecu en o which uses p e-
dic i e me hods.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
112 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
2.5. In ellec ual s uc u e o he ield - Bibliog aphic Coupling
The in ellec ual s uc u e e eals how an au ho ’s wo k in lu-
ences a de e mined scien i ic communi y. Co-ci a ion analysis and
bibliog aphic coupling ha e been used o analyze he in ellec ual
s uc u e o a ield o scien i ic esea ch (Cobo e al., 2011). Bib-
liog aphic coupling answe s ques ions such as; Wha is he in el-
lec ual s uc u e o ecen /eme ging li e a u e? (Zupic & Ca e ,
2015). Bibliog aphic coupling is pa icula ly sui able o analyz-
ing he esea ch on o a esea ch opic o ield (Zupic & Ca e ,
2015). The concep o a esea ch on is used o desc ibe cu en
scien i ic a icles ha ci e he knowledge base publica ions, p e i-
ously desc ibed as he se o a icles mos ci ed by cu en esea ch.
Two a icles a e linked bibliog aphically i a leas one ci ed
sou ce appea s in he bibliog aphies o e e ence lis s o bo h a -
icles (Kessle , 1963). Bibliog aphic coupling desc ibes he ex en
o which wo a icles a e ela ed because bo h e e o he same
a icle (Fe ei a, 2018). This analysis echnique can also be applied
o au ho s and jou nals. The au ho ’s bibliog aphic coupling aims
o disco e co-au ho ela ionships be ween au ho s who ci e he
same e e ences, while bibliog aphic coupling o jou nals aims o
disco e jou nals ha ci e he same e e ences (Cobo e al., 2011).
Fo bibliog aphic coupling analysis, he 50 mos ci ed au-
ho s, a icles, and jou nals we e selec ed.
A. Au ho s
Bibliog aphic coupling o au ho s is a me hod o mapping
ac i e au ho s, which can gi e a ealis ic iew o he cu en s a e
o esea ch (Zhao & S o mann, 2008). I can also analyze he
social s uc u e o a pa icula esea ch ield. Figu e11 p esen s
he esul ing ne wo k diag am.
The cen ali y measu es o he co-au ho ship ne wo k we e cal-
cula ed. Based on be weenness cen ali y and closeness cen ali y, as
explained p e iously, he au ho s wi h he g ea es measu e o p ox-
imi y a e: (1)Van Den Poel D; (2)Coussemen K; and (3)Hsieh YC.
Table17 shows he op 10 au ho s ega ding hese wo me ics ( he
comple e lis o he da abase i sel can be ob ained om he co e-
sponding au ho upon eques ). We can conclude ha hese au ho s
a e he main cons i uen s o he in ellec ual s uc u e o he esea ch
ield. They a e he au ho s o he esea ch on .
Table 17
Top 10 Au ho s wi h high Be weenness and Closeness (BC)
Au ho s Closeness Au ho s Be weenness
Van Den Poel D 0.242 Mahajan V 0.012
Coussemen K 0.241 Zhang Y 0.012
Hsieh YC 0.240 Kim D 0.010
Mahajan V 0.239 Van Den Poel D 0.009
Ge po T 0.239 Ge po TJ 0.008
S akho ych S 0.239 Coussemen K 0.008
Lee YS 0.238 Polo Y 0.006
De Bock KW 0.238 Ja ie Sese F 0.006
Ahmadi N 0.238 Gup a S 0.006
Ewing M 0.238 Ahmadi N 0.006
Sou ce: Au ho ’s own elabo a ion.
Figu e 11
Bibliog aphic coupling o au ho s
Sou ce: Au ho ’s own elabo a ion-Bibliome ix ou pu .
No e: The size o he ci cle indica es an i em’s weigh , he lines indica e he links be ween he i ems,
he dis ance be ween he i ems shows hei ela ionship, and he di e en colo s indica e he clus e s.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 113
B. A icles
As men ioned p e iously, bibliog aphic coupling desc ibes
he ex en o which wo a icles a e ela ed by bo h ci ing he
same a icle (Fe ei a, 2018). Figu e12 shows he ne wo k dia-
g am esul ing om bibliog aphic coupling o he a icles. Cen-
ali y measu es o he ne wo k and he espec i e nodes we e
calcula ed. The a icles wi h he g ea es cen ali y measu e a e:
(1)Imp o ing cus ome a i ion p edic ion by in eg a ing emo-
ions om clien /company in e ac ion emails and e alua ing
mul iple classi ie s (Coussemen & Van den Poel, 2009); (2)A
new hyb id classi ica ion algo i hm o cus ome chu n p edic-
ion based on logis ic eg ession and decision ees (De Caigny
e al., 2018); and (3)Regaining d i ing mobile communica ion
cus ome s: P edic ing he odds o success o winback e o s wi h
compe ing isks eg ession (Ge po & Ahmadi, 2015).
Figu e 12
Bibliog aphic Coupling o A icles
Sou ce: Au ho ’s own elabo a ion-Bibliome ix ou pu .
No e: The size o he ci cle indica es an i em’s weigh , he lines indica e he links be ween he i ems,
he dis ance be ween he i ems shows hei ela ionship, and he di e en colo s indica e he clus e s.
Table18 shows he op10 a icles conce ning hei deg ee o
closeness cen ali y, ha is, he a icles mos simila o each o he .
Table19 shows he a icles wi h a highe deg ee o be ween-
ness cen ali y: he a icles ep esen ing a b idge be ween he
di e en esea ch s eams. The a icles in bo h ables cons i u e
he esea ch on on cus ome chu n.
The majo i y o a icles come a e 2009, wi h 90% o he
a icles in he ables belonging o his pe iod. Mo e han hal
he a icles a e om 2015 onwa ds. This was also obse ed in
he desc ip i e analysis, in he e olu ion o a icle publica ion
o e ime. The esea ch on is ela ed mainly o p edic i e
analysis.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
114 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
Table 18
Top 10 A icles wi h high Closeness (BC)
Au ho s A icle Closeness
Coussemen K, 2009 Imp o ing cus ome a i ion p edic ion by in eg a ing emo ions om clien /company in e ac ion emails
and e alua ing mul iple classi ie s
0.309
De Caigny, 2018 A new hyb id classi ica ion algo i hm o cus ome chu n p edic ion based on logis ic eg ession and
decision ees
0.308
Ge po TJ, 2015 Regaining d i ing mobile communica ion cus ome s: P edic ing he odds o success o winback e o s wi h
compe ing isks eg ession
0.306
Coussemen K, 2014 Imp o ing cus ome e en ion managemen h ough cos -sensi i e lea ning 0.306
Mahajan V, 2015 Re iew o da a mining echniques o chu n p edic ion in elecom 0.306
Ahn J, 2020 A su ey on chu n analysis in a ious business domains 0.306
Jah omi AT 2014 Managing B2B cus ome chu n, e en ion and p o i abili y 0.305
Ge po TJ, 2015 Who is (no ) con inced o wi hd aw a con ac e mina ion announcemen ? - A disc iminan analysis o
mobile communica ions cus ome s in Ge many
0.304
Ve beke W, 2012 New insigh s in o chu n p edic ion in he elecommunica ion sec o : A p o i d i en da a mining app oach 0.304
Buckinx W, 2005 Cus ome base analysis: Pa ial de ec ion o beha iou ally loyal clien s in a non-con ac ual FMCG e ail se ing 0.300
Sou ce: Au ho ’s own elabo a ion.
Table 19
Top 10 A icles wi h high Be weenness (BC)
Au ho s A icle Be weenness
Ahn Y, 2020 Cus ome a i ion analysis in he secu i ies indus y: A la ge-scale ield s udy In Ko ea 0.014
Mahajan V, 2015 Re iew o da a mining echniques o chu n p edic ion in elecom 0.014
Coussemen K, 2009 Imp o ing cus ome a i ion p edic ion by in eg a ing emo ions om clien /company in e ac ion emails
and e alua ing mul iple classi ie s
0.013
Ge po TJ, 2015 New insigh s in o chu n p edic ion in he elecommunica ion sec o : A p o i d i en da a mining app oach 0.013
Benedek G, 2014 The impo ance o social embeddedness: chu n models a mobile p o ide s 0.011
Ahn J, 2020 A su ey on chu n analysis in a ious business domains 0.011
Al-Mash aie M, 2020 Cus ome swi ching beha io analysis in he elecommunica ion indus y 0.010
De Caigny, 2018 A new hyb id classi ica ion algo i hm o cus ome chu n p edic ion 0.009
Polo Y, 2009 How o Make Swi ching Cos ly: The ole o ma ke ing and ela ionship cha ac e is ics 0.009
Hadden J 2007 Compu e -assis ed cus ome chu n managemen : S a e-o - he-a and u u e ends 0.008
Sou ce: Au ho ’s own elabo a ion.
C. Jou nals
The bibliog aphic coupling o jou nals seeks o s udy he
ela ionship be ween common e e ences among he jou nals’
publica ions (Cobo e al., 2011). Figu e 13 p esen s he ne -
wo k diag am esul ing om he bibliog aphic coupling o
jou nals.
Cen ali y measu es we e calcula ed o he co-au ho ship
ne wo k. Rega ding closeness cen ali y, he jou nals wi h he
highes alue a e (1)Expe Sys ems wi h Applica ions, (2)Tel-
ecommunica ions Policy, (3)Eu opean Jou nal o Ma ke ing.
Expe Sys ems wi h Applica ions has he highes numbe o
a icles in he sample unde s udy, a 12%. Rega ding be ween-
ness cen ali y, he jou nals wi h he highes alues a e (1)In-
e na ional Jou nal o Bank Ma ke ing; (2) Expe Sys ems
wi h Applica ions; and (3) Eu opean Jou nal o Ope a ional
Resea ch. Table20 shows he op10 jou nals in e ms o cen-
ali y measu es.
F om he esea ch ield o he jou nals, we can conclude
ha he esea ch on has been p ima ily published by jou nals
whose ocus is on in elligen sys ems, echnologies, applica ions,
in elligence, and da a science. Also no ewo hy a e wo jou nals,
which a e mo e sec o -o ien ed, wi h publica ions on elecom-
munica ions and banking.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 115
Figu e 13
Bibliog aphic Coupling o Sou ces
Sou ce: Au ho ’s own elabo a ion-Bibliome ix ou pu .
No e: The size o he ci cle indica es an i em’s weigh , he lines indica e he links be ween he i ems,
he dis ance be ween he i ems shows hei ela ionship, and he di e en colo s indica e he clus e s.
Table 20
Top 10 Jou nals wi h high Be weenness and Closeness (BC)
Jou nal Closeness Jou nal Be weenness
Expe Sys ems Wi h Applica ions 0.326 In e na ional Jou nal o Bank Ma ke ing 0.035
Telecommunica ions Policy 0.321 Expe Sys ems wi h Applica ions 0.030
Eu opean Jou nal o Ma ke ing 0.321 Eu opean Jou nal o Ope a ional Resea ch 0.025
Eu opean Jou nal o Ope a ional Resea ch 0.320 Decision Sciences 0.025
Indus ial Ma ke ing Managemen 0.319 Eu opean Jou nal o Ma ke ing 0.021
In e na ional Jou nal o Bank Ma ke ing 0.317 Jou nal o Business Resea ch 0.016
Jou nal o Business Resea ch 0.317 Telecommunica ions Policy 0.015
Jou nal o Se ice Resea ch 0.317 Jou nal o Se ice Resea ch 0.015
Ma ke ing Science 0.315 Ma ke ing Science 0.015
In o ma ion Sys ems and e-Business Managemen 0.310 Indus ial Ma ke ing Managemen 0.014
Sou ce: Au ho ’s own elabo a ion.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
116 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
2.6. Co-au ho analysis
Co-au ho analysis looks a au ho s and hei a ilia ions
o s udy social s uc u e and collabo a ion ne wo ks (Pe e s &
Van aan, 1991). I is pa icula ly well sui ed o s udying esea ch
ques ions in ol ing scien i ic collabo a ion (Zupic & Ca e ,
2015). The mos commonly used me hod o s udy social s uc-
u e is he co-au ho ne wo k. Howe e , i is also possible o use
in o ma ion abou au ho s’ geog aphical loca ion and ins i u-
ional a ilia ions o examine collabo a ion a he le el o ins i u-
ions and coun ies (Zupic & Ca e , 2015).
A. Au ho
Figu e14 shows he co-au ho ship ne wo k g aph. The e a e
10 clus e s, and h ee s and ou wi h a highe numbe o co-au-
ho s. The i s was led by Baesens B and Ve beke W, he second
by Van Den Poel D and Coussemen K, and he hi d by AminA
and Anwa S. These au ho s also ha e he highes numbe o a -
icles in he sample s udied.
Table 21 p esen s he au ho ne wo ks, o de ing hem by
PageRank. Fo co-au ho ing ne wo ks, PageRank gi es g ea e
weigh o au ho s who collabo a e wi h di e en au ho s and
hose who collabo a e wi h ew au ho s bu do so equen ly (Yan
& Ding, 2011).
Figu e 14
Co-au ho ship analysis
Sou ce: Au ho ’s own elabo a ion-Biblioshiny ou pu .
No e: The ec angles deno e au ho nodes and a e labeled
by he au ho ’s las name and ini ials. The node size co esponds
o he numbe o publica ions, and he colo o he di e en clus e s.
Table 21
A icles o each co-au ho ship clus e : PageRank measu e
Clus e 1 PageRank Clus e 2 PageRank
Van den Poel D 0.037 S akho ych S 0.026
Coussemen K 0.028 Ewing M 0.026
Migueis Vl 0.026
Camanho As 0.026
Falcao E 0.026
Cunha J 0.026
De Bock Kw 0.024
De Caigny A 0.019
Clus e 3 PageRank Clus e 4 PageRank
Becke Ju 0.026 Amin A 0.040
Spann M 0.026 Anwa S 0.040
Shah B 0.021
Nawaz M 0.021
Hussain A 0.021
Al-Obeida F 0.016
Clus e 5 PageRank Clus e 6 PageRank
Baesens B 0.063 Kim Ys 0.026
Ve beke W 0.036 Lee H 0.026
Van hienen J 0.026
Ma ens D 0.023
Oska sdo i M 0.019
S ipling E 0.009
Zhu B 0.009
Clus e 7 PageRank Clus e 8 PageRank
Guillen M 0.026 Li H 0.026
Nielsen Jp 0.026 Liu Y 0.026
Clus e 9 PageRank Clus e 10 PageRank
Raza B 0.026 Khan Y 0.026
Malik Ak 0.026 Sha iq S 0.026
Ahmed S 0.026
Sa wan N 0.026
Sou ce: Au ho ’s own elabo a ion.
This analysis also shows ha he clus e s o au ho s wi h
he g ea es scien i ic collabo a ion a e essen ially hose iden-
i ied as mainly esponsible o he esea ch on h ough he
bibliog aphic coupling analysis, meaning he e is high collab-
o a ion on he esea ch on . Van Den Poel D and Cousse-
men K s and ou wi h he highes cen ali y measu e, mean-
ing hey can each o he au ho s in he ne wo k ia a sho e
pa h.
B. Coun y o a ilia ion
Rega ding co-au ho ship in he au ho ’s coun y o a ilia-
ion, Figu e15 p esen s i e clus e s, wo o which ha e a highe
numbe o co-au ho s.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 117
Figu e 15
Co-au ho ship analysis (Coun y)
Sou ce: Au ho ’s own elabo a ion-Biblioshiny ou pu .
No e: The nodes ep esen coun ies and links connec coun ies
in he o m o co-au ho ships. Bibliome ix so wa e a ibu es
a di e en colo o each clus e .
The i s clus e comp ises he Uni ed S a es and China wi h a
highe PageRank, and a second made up o he Uni ed Kingdom
and Belgium wi h a high PageRank. We ecall ha China, he Uni -
ed S a es, and Belgium a e esponsible o 41% o all publica ions.
Table22 p esen s he coun y ne wo ks, so ing hem by PageRank.
Table 22
Coun ies o each co-au ho ship clus e : PageRank measu e
Clus e 1 PageRank Clus e 2 PageRank
Uni ed Kingdom 0.121 Spain 0.036
Belgium 0.106 Denma k 0.023
F ance 0.042
Po ugal 0.016
Clus e 3 PageRank Clus e 4 PageRank
Ge many 0.041 Pakis an 0.0637
Swi ze land 0.026 Uni ed A ab
Emi a es
0.0214
Saudi A abia 0.0245
Clus e 5 PageRank
USA 0.170
China 0.148
Ko ea 0.057
India 0.031
Ne he lands 0.020
Canada 0.019
Aus alia 0.016
Tu key 0.016
Sou ce: Au ho ’s own elabo a ion.
Figu e16 shows he collabo a i e map be ween coun ies, whe e
he equencies be ween China and he Uni ed S a es and be ween
he Uni ed Kingdom and Belgium s and ou , as seen ea lie .
Figu e 16
Coun y Collabo a ionMap
Sou ce: Au ho ’s own elabo a ion-Biblioshiny ou pu .
No e: The blue colo on he map ep esen s esea ch coope a ion among
coun ies. The da ke he colo , he highe he coun y collabo a ion.
The scale o coope a ion is ep esen ed h ough he hickness o he line.
C. Ins i u ion om a ilia ion
Finally, conce ning he au ho s’ a ilia ed ins i u ions, he
co-au ho ship ne wo k e eals one p ominen clus e , wi h he
Uni e si y o Sou hamp on and he Ca holic Uni e si y o Leu-
en a he head, a ilia ed uni e si ies o Baesens B, who is also
he au ho wi h he second highes numbe o a icles on he
da abase. Figu e17 p esen s he g aph o he co-au ho ship ne -
wo k by educa ional ins i u ion.
Figu e 17
Co-au ho ship analysis (Ins i u ion)
Sou ce: Au ho ’s own elabo a ion-Biblioshiny ou pu .
No e: The nodes ep esen ins i u ions and a e labeled
by he ins i u ion name. The node size co esponds o he numbe
o occu ences o co-au ho ships publica ions, and links connec
ins i u ions in he o m o co-au ho ships. Bibliome ix so wa e
a ibu es di e en colo s o each clus e .
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
118 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
Table23 shows he ne wo ks o ins i u ions by clus e , o -
de ed by PageRank.
Table 23
Ins i u ions o each co-au ho ship clus e : PageRank measu e
Clus e 1 PageRank Clus e 2 PageRank
Ka holieke Uni e si y
Leu en
0.121 Ins Managemen Sci 0.084
Uni e si y
Sou hamp on
0.108 Zayed Uni e si y 0.039
Uni e si y o An we p 0.035 Taibah Uni e si y 0.050
V ije Uni e si ei
B ussel
0.026 Uni e si y S i ling 0.050
Sichuan Uni e si y 0.026
Uni e si y Ghen 0.017
Clus e 3 PageRank Clus e 4 PageRank
Columbia Uni e si y 0.056 Monash Uni e si y 0.003
Uni e si y o
Pennsyl ania
0.056 Deakin Uni e si y 0.003
Clus e 5 PageRank
Chinese Academy o
Sciences
0.056
Uni e si y Neb aska 0.056
Sou ce: Au ho ’s own elabo a ion.
3. CONCLUSIONS AND DIRECTIONS FOR FUTURE
RESEARCH
As ma ke s become inc easingly sa u a ed, academic e-
sea che s and companies ha e ecognized i is essen ial o iden-
i y he cus ome s mos likely o swi ch o ano he se ice p o-
ide (Kea eney & Pa hasa a hy, 2001). The in en ion he e was
o p esen a s uc u ed e iew o cus ome chu n, using biblio-
me ic echniques o analyze he esea ch ield’s in ellec ual and
concep ual s uc u e. As a as we know, a bibliome ic analysis
has ne e been ca ied ou o iden i y analy ically and objec i ely
he mos in luen ial s udies and au ho s, as well as he eme ging
esea ch clus e s, so we belie e ha his s udy ad ances knowl-
edge abou he ield o s udies analyzed.
This s udy syn hesizes insigh s om a conside able amoun o
ele an li e a u e on cus ome chu n. The da a show ha esea ch
on cus ome chu n has been published mo e and mo e, hus
demons a ing he g owing impo ance o his ield o esea ch.
As o he mos in luen ial ou le s o he dissemina ion o
esea ch abou cus ome chu n, Expe Sys ems wi h Applica-
ions is he jou nal o choice. Amongs he mos p oli ic au ho s,
we highligh Van Den Poel, Baesens B and Coussemen K, which
explains, in pa , why Belgium is one o he coun ies ha ha e
con ibu ed mos o de eloping and dissemina ing esea ch on
cus ome chu n. In e es ingly, when we ocus ou analysis on
he mos ci ed e e ences, he wo k by Kea eney S.M anks e y
highly, sugges ing he ele ance o his au ho ’s esea ch in his
a ea o knowledge. Howe e , when one analyzes he global and
local ci a ions, one ealizes ha Ve beke and Bu ez s and ou
among he op-10 mos local-ci ed a icles, Bu ez being in he
las place in he op-10 gobal-ci ed a icles. One possible expla-
na ion why hey a e no among he op-10 mos globally ci ed a -
icles is ha hei publica ions a e om 2011 and 2012 (Ve beke)
and om 2007 and 2009 (Bu ez).
In o de o complemen he desc ip i e analysis and ob ain
u he insigh s in o he cus ome chu n li e a u e, his a icle
analyzed he concep ual and in ellec ual s uc u e o he esea ch
ield. De e mina ion o he in ellec ual s uc u e and he esea ch
on o he scien i ic domains a e essen ial no only o esea ch
bu also o elabo a e policies and p ac ices (A ia & Cuccu ullo,
2017). The impo ance o ha ing “a concep ual and in ellec ual
map” is undeniable o he cons uc ion o a holis ic iew o a
ield o s udies. The bibliome ic analysis (1995-2000) ca ied
ou in his s udy allows mapping and syn hesizing he ela ion-
ships be ween au ho s, a icles, and undamen al jou nals in he
ield o cus ome chu n.
Rega ding he i s esea ch ques ion p oposed o his a -
icle, which was abou he speci ic opics associa ed wi h cus-
ome chu n esea ch, we pe o med a co-wo d analysis. We
obse ed in he i s ins ance ha he e a e wo di e en angles,
wo majo lines o in es iga ion, which a e complemen a y.
One, in which he esea che s’ objec i es a e o unde s and wha
leads o cus ome chu n and o de ine essen ial chu n ac o s,
such as sa is ac ion, se ice quali y and se ice a ibu es (e.g.,
A hanassopoulos, 2000; Kea eney, 1995; Reichheld & Sasse ,
1990; Rus & Zaho ik, 1993). The second, whe e esea che s o-
cus on imp o ing cus ome chu n o ecas ing models o boos
p edic i e pe o mance (e.g., Ve beke e al., 2012; Ve beke e al.,
2011; Wei& Chiu, 2002). These p edic i e s udies ypically apply
modeling echniques, such as a i icial neu al ne wo ks, decision
ees, and andom o es s, o la ge samples o subsc ibe s. Com-
plemen a ily, and h ough he cons uc ion o a hema ic map3,
chu n p edic ion was ound o be he mos impo an heme in
he ield o in es iga ion. This heme ela es o di e en concep s
such as “classi ica ion”, “machine lea ning” and “ elecommunica-
ion”, among o he s. Since elecommunica ions is an a ea sui able
o his ype o in es iga ion, u u e esea ch on cus ome u n-
o e is ecommended, using no p edic i e models, bu beha io-
al models, wi h an eceden s ocusing on sa is ac ion, quali y o
se ice, swi ching cos s, and socio-demog aphic da a, o name
jus a ew. Howe e , despi e hese indings, chu n seems o be an
un esol ed issue as cus ome expe ience, encoun e , disappoin -
men and in e ac ion a e con inuously p esen in he se ices
p o ided. Mo eo e , beyond he ela ional aspec s o si ua ional
in e ac ion, he e is a lack o knowledge on how cumula i e ex-
pe ience and cus ome encoun e s in luence chu n.
Rega ding he in ellec ual s uc u e o he ield, app oached
by he second esea ch ques ion, wo echniques o bibliome ic
analysis we e used, co-ci a ion analysis and bibliog aphic cou-
pling. Fi s ly, h ough analyzing a icles, jou nals and au ho s,
we sough o map olde wo ks, iden i ying he in ellec ual s uc-
u e a he base o he ield o s udies h ough analysis o co-ci-
3 Thema ic mapping consis s o a wo d co-occu ence ne wo k analysis o
de ine wha science alks abou in a esea ch ield, main hemes, and ends.
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 119
a ions. Secondly, we mapped he cu en on o he esea ch
ield h ough bibliog aphic coupling.
Co-ci a ion analysis e ealed ha he au ho s wi h he g ea -
es closeness and be weenness cen ali y a e Reichheld FF, Kea -
eney SM, Bol on RN, Rus RT, and Ganesh J, ha is, hese a e
he p ima y au ho s mos ci ed by cu en esea ch, and hey
a e he ounda ions o cu en esea ch. The a icle “Cus ome
swi ching beha io in se ice indus ies: An explo a o y s udy”
by Kea eney (1995), which in he in ellec ual s uc u e o he e-
sea ch ield eme ges as he a icle wi h he g ea es measu e o
closeness cen ali y, is one o he main ounding a icles o he
esea ch ield and is also he mos ci ed a icle globally. I was he
i s a icle o p esen a model explaining cus ome chu n in se -
ice indus ies conside ing a possible numbe o causal ac o s
and hei in e ela ionships These da a ackle he hi d esea ch
ques ion p oposed by his a icle, ega ding he cen al, pe iph-
e al, o b idging esea che s in his ield.
Rega ding cu ing-edge esea ch and he in ellec ual s uc-
u e o eme ging li e a u e ( ou h esea ch ques ion), biblio-
g aphic coupling e ealed ha he p incipal au ho s a e Cousse-
men K and Van den Poel D, wi h he mos p ominen a icle
being by hese au ho s, Imp o ing Cus ome A i ion P edic-
ion by In eg a ing Emo ions om Clien /Company In e ac ion
Emails and E alua ing Mul iple Classi ie s (Coussemen & Van
den Poel, 2009). Hence, p edic i e me hods a e ound o be a
he cu ing edge o cus ome chu n esea ch.
Finally, in ela ion o he social s uc u e o he esea ch
ield ( i h esea ch ques ion), he Uni ed S a es and China a e
he main collabo a ing coun ies. This collabo a ion should be
ex ended o o he clus e s o coun ies, since cus ome u no-
e is ans e sal o all coun ies, and a mo e comp ehensi e
iew o c oss-cul u al managemen is clea ly necessa y.
Th oughou he s udy, and in all he analyses pe o med, no
e e ences we e ound o he esea ch ield abou cus ome expe-
ience. A good cus ome expe ience ends o educe signi ican ly
he p opensi y o swi ch o ano he b and (Si ap acha & Tocque ,
2012). Expe iences can be seen as he “imp essions” ha emain
in cus ome s’ minds, as he esul o a holis ic encoun e wi h an
o e o objec (Iglesiase al., 2011), culmina ing in sa is ac ion o
disappoin men and dese ion o abandonmen (Meye & Schwa-
ge , 2007). A posi i e expe ience can p omo e an emo ional bond
be ween he b and and i s cus ome s, which, in u n, inc eases
cus ome loyal y (Gen ile e al., 2007). One ecommenda ion o
u u e in es iga ion is o s udy he ela ionship be ween he cus-
ome expe ience and cus ome chu n, he de e minan s and hei
impac on cus ome chu n. We sha e he same ques ion e e ed
o by Lemon and Ve hoe (2016) who ask i cus ome expe ience
can explain he cus ome ’s beha io and he company’s pe o -
mance. I is also impo an o add ess how disconnec ed he e-
la ion be ween cus ome sa is ac ion and in ol emen wi h chu n
is. Once again, we emphasize he need o app oach chu n add ess-
ing he beha io al side, no only wi h p edic i e me hods bu also
add essing how emo ional d i e s in luence beha io al esponses
leading o chu n, bo h in ela ional o ansac ional encoun e s.
Mo eo e , he e is lack o e e ences o how compe i o s’ ac ion
in luences chu ning beha io .
Al hough his in es iga ion con ibu es o knowledge in his
a ea, being he i s a emp o map he esea ch ield sys ema ical-
ly, se e al limi a ions and u u e esea ch oppo uni ies a e wo h
men ioning h ough a bibliome ic analysis. Da a collec ion was
ca ied ou exclusi ely on he “Web o Science” da abase and, con-
sequen ly, a icles and o he ypes o documen s indexed by o he
pla o ms we e no co e ed. This decision was because he so wa e
used o bibliome ic analysis is s ill unable o me ge he e e ences
ci ed co ec ly. In he u u e, o he da abases (e.g., Scopus) should
be analyzed oge he wi h WoS. Only scien i ic a icles we e con-
side ed, excluding books, con e ence p oceedings, edi o ial ma e i-
al, and o he s, so i will be o in e es o examine o he publica ions
no included in he sample, o complemen he esul s ob ained.
Mo eo e , he ocus o his a icle was on bibliome ic analysis
and he in ellec ual s uc u e o chu n. As such, a li e a u e e iew
based on con en analysis is manda o y o explo e he in icacies
ela ing consume beha io , cus ome sa is ac ion, cus ome loyal-
y, chu n p edic ion, cus ome chu n and hei emo ional d i e s.
4. REFERENCES
Adebiyi, S. O., Oya oye, E. O., & Amole, B. B. (2016). Rele an d i e s
o cus ome s` chu n and e en ion decision in he Nige ian mobile
elecommunica ion indus y. Jou nal o Compe i i eness, 6(3), 52-
67. h ps://doi.o g/10.7441/joc.2016.03.04
Ahn, J. H., Han, S. P., & Lee, Y. S. (2006). Cus ome chu n analysis:
Chu n de e minan s and media ion e ec s o pa ial de ec ion in
he Ko ean mobile elecommunica ions se ice indus y. Telecom-
munica ions Policy, 30(10-11), 552-568. h ps://doi.o g/10.1016/j.
elpol.2006.09.006
Ahn, J., Hwang, J., Kim, D., Choi, H., & Kang, S. (2020). A Su ey on
chu n analysis in a ious business domains. IEEE Access, 8, 220816-
220839. h ps://doi.o g/10.1109/access.2020.3042657
Al-Mash aie, M., Chung, S. H., & Jeon, H. W. (2020). Cus ome swi ching
beha io analysis in he elecommunica ion indus y ia push-pull-
moo ing amewo k: A machine lea ning app oach. Compu e s &
Indus ial Enginee ing, 144. h ps://doi.o g/10.1016/j.cie.2020.106476
Amin, A., Shah, B., Kha ak, A. M., Mo ei a, F. J. L., Ali, G., Rocha, A.,
& Anwa , S. (2019). C oss-company cus ome chu n p edic ion in
elecommunica ion: A compa ison o da a ans o ma ion me h-
ods. In e na ional Jou nal o In o ma ion Managemen , 46, 304-319.
h ps://doi.o g/10.1016/j.ijin omg .2018.08.015
Ami i, H., & Daume III, H. (2016). Sho ex ep esen a ion o de ec ing
chu n in mic oblogs. Pape p esen ed a he Thi ie h AAAI Con e -
ence on A i icial In elligence.
Ande son, E. W., Fo nell, C., & Lehmann, D. R. (1994). Cus ome sa is ac-
ion, ma ke sha e, and p o i abili y - Findings om Sweden. Jou nal o
Ma ke ing, 58(3), 53-66. h ps://doi.o g/10.1177/002224299405800304
A ia, M., & Cuccu ullo, C. (2017). Bibliome ix: An R- ool o com-
p ehensi e science mapping analysis. Jou nal o In o me ics, 11(4),
959-975. h ps://doi.o g/10.1016/j.joi.2017.08.007
A hanassopoulos, A. D. (2000). Cus ome sa is ac ion cues o suppo
ma ke segmen a ion and explain swi ching beha io . Jou nal o
Business Resea ch, 47(3), 191-207. h ps://doi.o g/10.1016/s0148-
2963(98)00060-5
Aydin, S., & Oze , G. (2005). The analysis o an eceden s o cus om-
e loyal y in he Tu kish mobile elecommunica ion ma ke .
Eu opean Jou nal o Ma ke ing, 39(7-6), 910-925. h ps://doi.
o g/10.1108/03090560510601833
Bansal, H. S., I ing, P. G., & Taylo , S. F. (2004). A h ee-componen
model o cus ome commi men o se ice p o ide s. Jou nal o
he Academy o Ma ke ing Science, 32(3), 234-250. h ps://doi.
o g/10.1177/0092070304263332
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
120 Hugo Ribei o, Belém Ba bosa, An ónio C. Mo ei a, Rica do Rod igues
Becke , J. U., Spann, M., & Schulze, T. (2015). Implica ions o minimum
con ac du a ions on cus ome e en ion. Ma ke ing Le e s, 26(4),
579-592. h ps://doi.o g/10.1007/s11002-014-9293-2
Benedek, G., Lubloy, A., & Vas ag, G. (2014). The Impo ance o So-
cial Embeddedness: Chu n Models a Mobile P o ide s. Decision
Sciences, 45(1), 175-201. h ps://doi.o g/10.1111/deci.12057
Bol on, R. N. (1998). A dynamic model o he du a ion o he cus ome 's e-
la ionship wi h a con inuous se ice p o ide : The ole o sa is ac ion.
Ma ke ing Science, 17(1), 45-65. h ps://doi.o g/10.1287/mksc.17.1.45
Buckinx, W., & Van den Poel, D. (2005). Cus ome base analysis: Pa ial
de ec ion o beha iou ally loyal clien s in a non-con ac ual FMCG
e ail se ing. Eu opean Jou nal o Ope a ional Resea ch, 164(1),
252-268. h ps://doi.o g/10.1016/j.ejo .2003.12.010
Bu ez, J., & Van den Poel, D. (2007). CRM a a pay-TV company: Us-
ing analy ical models o educe cus ome a i ion by a ge ed ma -
ke ing o subsc ip ion se ices. Expe Sys ems wi h Applica ions,
32(2), 277-288. h ps://doi.o g/10.1016/j.eswa.2005.11.037
Bu ez, J., & Van den Poel, D. (2009). Handling class imbalance in cus-
ome chu n p edic ion. Expe Sys ems wi h Applica ions, 36(3),
4626-4636. h ps://doi.o g/10.1016/j.eswa.2008.05.027
Bu nham, T. A., F els, J. K., & Mahajan, V. (2003). Consume swi ch-
ing cos s: A ypology, an eceden s, and consequences. Jou nal
o he Academy o Ma ke ing Science, 31(2), 109-126. h ps://doi.
o g/10.1177/0092070302250897
Callon, M., Cou ial, J. P., & La ille, F. (1991). Co-wo d analysis as a ool
o desc ibing he ne wo k o in e ac ions be ween basic and ech-
nological esea ch - The case o polyme chemis y. Scien ome ics,
22(1), 155-205. h ps://doi.o g/10.1007/b 02019280
Callon, M., Cou ial, J. P., Tu ne , W. A., & Bauin, S. (1983). F om ans-
la ions o p oblema ic ne wo ks - An in oduc ion o co-wo d anal-
ysis. Social Science In o ma ion Su Les Sciences Sociales, 22(2), 191-
235. h ps://doi.o g/10.1177/053901883022002003
Ca izo-Mo ei a, A., F ei as-da Sil a, P. M., & Fe ei a-Mou inho, V.
M. (2017). The e ec s o b and expe iences on quali y, sa is ac ion
and loyal y: An empi ical s udy in he elecommunica ions mul-
iple-play se ice ma ke . Inno a , 27(64), 23-38. h ps://doi.o g/
10.15446/inno a . 27n64.62366
Chen, P. Y., & Hi , L. M. (2002). Measu ing swi ching cos s and he
de e minan s o cus ome e en ion in In e ne -enabled businesses:
A s udy o he Online b oke age indus y. In o ma ion Sys ems Re-
sea ch, 13(3), 255-274. h ps://doi.o g/10.1287/is e.13.3.255.78
Cobo, M. J., Lopez-He e a, A. G., He e a-Viedma, E., & He e a, F.
(2011). Science mapping so wa e ools: e iew, analysis, and co-
ope a i e s udy among ools. Jou nal o he Ame ican Socie y o
In o ma ion Science and Technology, 62(7), 1382-1402. h ps://doi.
o g/10.1002/asi.21525
Coussemen , K. (2014). Imp o ing cus ome e en ion managemen
h ough cos -sensi i e lea ning. Eu opean Jou nal o Ma ke ing,
48(3-4), 477-495. h ps://doi.o g/10.1108/ejm-03-2012-0180
Coussemen , K., & Van den Poel, D. (2009). Imp o ing cus ome a -
i ion p edic ion by in eg a ing emo ions om clien /company
in e ac ion emails and e alua ing mul iple classi ie s. Expe Sys-
ems wi h Applica ions, 36(3), 6127-6134. h ps://doi.o g/10.1016/j.
eswa.2008.07.021
De Caigny, A., Coussemen , K., & De Bock, K. W. (2018). A new hy-
b id classi ica ion algo i hm o cus ome chu n p edic ion based
on logis ic eg ession and decision ees. Eu opean Jou nal o Op-
e a ional Resea ch, 269(2), 760-772. h ps://doi.o g/10.1016/j.
ejo .2018.02.009
de Haan, E., Ve hoe , P. C., & Wiesel, T. (2015). The p edic i e abili y
o di e en cus ome eedback me ics o e en ion. In e na ion-
al Jou nal o Resea ch in Ma ke ing, 32(2), 195-206. h ps://doi.
o g/10.1016/j.ij esma .2015.02.004
Eck, N. J. ., & Wal man, L. (2009). How o no malize co-occu ence
da a? An analysis o some well‐known simila i y measu es. Jou nal
o he Ame ican socie y o in o ma ion science and echnology, 60(8),
1635-1651. h ps://doi.o g/10.1002/asi.21075
Ellegaa d, O., & Wallin, J. A. (2015). The bibliome ic analysis o schol-
a ly p oduc ion: How g ea is he impac ? Scien ome ics, 105(3),
1809-1831. h ps://doi.o g/10.1007/s11192-015-1645-z
Eshghi, A., Haugh on, D., & Topi, H. (2007). De e minan s o cus-
ome loyal y in he wi eless elecommunica ions indus y. Tele-
communica ions Policy, 31(2), 93-106. h ps://doi.o g/10.1016/j.
elpol.2006.12.005
Fe ei a, F. A. F. (2018). Mapping he ield o a s-based managemen :
Bibliog aphic coupling and co-ci a ion analyses. Jou nal o Business
Resea ch, 85, 348-357. h ps://doi.o g/10.1016/j.jbus es.2017.03.026
Fischbach, K., Pu zke, J., & Schode , D. (2011). Co-au ho ship ne wo ks
in elec onic ma ke s esea ch. Elec onic Ma ke s, 21(1), 19-40.
h ps://doi.o g/10.1007/s12525-011-0051-5
Ganesh, J., A nold, M. J., & Reynolds, K. E. (2000). Unde s anding he
cus ome base o se ice p o ide s: An examina ion o he di e -
ences be ween swi che s and s aye s. Jou nal o Ma ke ing, 64(3),
65-87. h ps://doi.o g/10.1509/jmkg.64.3.65.18028
Ga ield, E., & She , I. H. (1993). KEYWORDS-PLUS(TM) - Algo i h-
mic de i a i e indexing. Jou nal o he Ame ican Socie y o In o -
ma ion Science, 44(5), 298-299. h ps://doi.o g/10.1002/(sici)1097-
4571(199306)44:5<298::aid-asi5>3.0.co;2-a
Gen ile, C., Spille , N., & Noci, G. (2007). How o sus ain he cus ome
expe ience:: An o e iew o expe ience componen s ha co-c ea e
alue wi h he cus ome . Eu opean managemen jou nal, 25(5), 395-
410. h ps://doi.o g/ /10.1016/j.emj.2007.08.005
Ge po , T. J., & Ahmadi, N. (2015). Regaining d i ing mobile commu-
nica ion cus ome s: P edic ing he odds o success o winback e o s
wi h compe ing isks eg ession. Expe Sys ems wi h Applica ions,
42(21), 7917-7928. h ps://doi.o g/10.1016/j.eswa.2015.05.011
Haddaway, N. R. A. U. P. C. C., & McGuinness, L. A. (2021). PRIS-
MA2020: R package and ShinyApp o p oducing PRISMA 2020
complian low diag ams (Ve sion 0.0.2): Zenodo. Re ie ed om
h p://doi.o g/10.5281/zenodo.5082518
Hadden, J., Tiwa i, A., Roy, R., & Ru a, D. (2007). Compu e assis ed
cus ome chu n managemen : S a e-o - he-a and u u e ends.
Compu e s & Ope a ions Resea ch, 34(10), 2902-2917. h ps://doi.
o g/10.1016/j.co .2005.11.007
Iglesias, O., Singh, J. J., & Ba is a-Fogue , J. M. (2011). The ole o
b and expe ience and a ec i e commi men in de e mining b and
loyal y. Jou nal o B and Managemen , 18(8), 570-582. h ps://doi.
o g/10.1057/bm.2010.58
Jah omi, A. T., S akho ych, S., & Ewing, M. (2014). Managing B2B
cus ome chu n, e en ion and p o i abili y. Indus ial Ma ke ing
Managemen , 43(7), 1258-1268. h ps://doi.o g/10.1016/j.indma -
man.2014.06.016
Jones, M. A., Mo he sbaugh, D. L., & Bea y, S. E. (2000). Swi ching
ba ie s and epu chase in en ions in se ices. Jou nal o Re ailing,
76(2), 259-274. h ps://doi.o g/10.1016/s0022-4359(00)00024-5
Kea eney, S. M. (1995). Cus ome swi ching beha io in-se ice in-
dus ies - An explo a o y-s udy. Jou nal o Ma ke ing, 59(2), 71-82.
h ps://doi.o g/10.2307/1252074
Kea eney, S. M., & Pa hasa a hy, M. (2001). Cus ome swi ching be-
ha io in online se ices: An explo a o y s udy o he ole o se-
lec ed a i udinal, beha io al, and demog aphic ac o s. Jou nal
o he Academy o Ma ke ing Science, 29(4), 374-390. h ps://doi.
o g/10.1177/03079450094225
Kessle , M. M. (1963). Bibliog aphic coupling be ween scien i ic pape s.
Ame ican Documen a ion, 14(1), 10-25. h ps://doi.o g/10.1002/
asi.5090140103
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
Chu n in se ices – A bibliome ic e iew 121
Kuma , V., Leszkiewicz, A., & He bs , A. (2018). A e you back o good
o s ill shopping a ound? In es iga ing cus ome s’ epea chu n be-
ha io . Jou nal o Ma ke ing Resea ch, 55(2), 208-225. h ps://doi.
o g/10.1509/jm .16.0623
Kyei, D. A., & Bayoh, A. T. M. (2017). Inno a ion and cus ome e en ion
in he Ghanaian elecommunica ion indus y. In e na ional Jou nal o
Inno a ion, 5(2), 171-183. h ps://doi.o g/10.5585/iji. 5i2.154
Lemmens, A., & C oux, C. (2006). Bagging and boos ing classi ica ion
ees o p edic chu n. Jou nal o Ma ke ing Resea ch, 43(2), 276-286.
h ps://doi.o g/10.1509/jmk .43.2.276
Lemon, K. N., & Ve hoe , P. C. (2016). Unde s anding cus ome expe i-
ence h oughou he cus ome jou ney. Jou nal o Ma ke ing, 80(6),
69-96. h ps://doi.o g/10.1509/jm.15.0420
Mahajan, V., Mis a, R., & Mahajan, R. (2015). Re iew o da a mining
echniques o chu n p edic ion in elecom. Jou nal o In o ma ion
and O ganiza ional Sciences, 39(2), 183-197.
Meye , C., & Schwage , A. (2007). Unde s anding cus ome expe ience.
Ha a d Business Re iew, 85(2), 116-26,157.
Mo ei a, A. C., Sil a, P., & Mou inho, V. (2016). Di e ences be ween
s aye s, swi che s, and hea y swi che s: A s udy in he elecommu-
nica ions se ice ma ke . Ma ke ing In elligence & Planning, 34(6),
843-862. h ps://doi.o g/10.1108/MIP-07-2015-0128
Moze , M. C., Wolniewicz, R., G imes, D. B., Johnson, E., & Kaushansky,
H. (2000). P edic ing subsc ibe dissa is ac ion and imp o ing e en-
ion in he wi eless elecommunica ions indus y. Ieee T ansac ions on
Neu al Ne wo ks, 11(3), 690-696. h ps://doi.o g/10.1109/72.846740
Neslin, S. A., Gup a, S., Kamaku a, W., Lu, J. X., & Mason, C. H. (2006).
De ec ion de ec ion: Measu ing and unde s anding he p edic i e
accu acy o cus ome chu n models. Jou nal o Ma ke ing Resea ch,
43(2), 204-211. h ps://doi.o g/10.1509/jmk .43.2.204
Newman, M. E., & Gi an, M. (2004). Finding and e alua ing commu-
ni y s uc u e in ne wo ks. Physical e iew E, 69(2), 026113. h ps://
doi.o g/10.1103/phys e e.69.026113
O man, G. K., & Laba u , V. (2009). A compa ison o communi y de ec ion
algo i hms on a i icial ne wo ks. Pape p esen ed a he In e na ion-
al con e ence on disco e y science. h ps://doi.o g/10.1007/978-3-
642-04747-3_20
Pe e s, H. P. F., & Van aan, A. F. J. (1991). S uc u ing scien i ic ac i i-
ies by coau ho analysis - An exe cise on a uni e si y- acul y le el.
Scien ome ics, 20(1), 235-255. h ps://doi.o g/10.1007/b 02018157
Polo, Y., & Sese, F. J. (2009). How o Make Swi ching Cos ly The Role o
Ma ke ing and Rela ionship Cha ac e is ics. Jou nal o Se ice Re-
sea ch, 12(2), 119-137. h ps://doi.o g/10.1177/1094670509335771
Pons, P., & La apy, M. (2005). Compu ing communi ies in la ge ne wo ks using
andom walks. Pape p esen ed a he In e na ional symposium on com-
pu e and in o ma ion sciences. h ps://doi.o g/10.1007/11569596_31
P ince, J., & G eens ein, S. (2014). Does se ice bundling educe chu n?
Jou nal o Economics & Managemen S a egy, 23(4), 839-875.
h ps://doi.o g/10.1111/jems.12073
P i cha d, A. (1969). S a is ical bibliog aphy o bibliome ics. Jou nal o
documen a ion, 25(4), 348-349.
Rajan Sachde a, R. R. S. (2017). E alua ing p edic ion o cus ome
chu n beha io based on a i icial bee colony algo i hm. In e na-
ional Jou nal o Enginee ing and Compu e Science, 6(1). h ps://
doi.o g/10.18535/ijecs/ 6i1.32
Ramos-Rod iguez, A. R., & Ruiz-Na a o, J. (2004). Changes in he in-
ellec ual s uc u e o s a egic managemen esea ch: A bibliome ic
s udy o he S a egic Managemen Jou nal, 1980-2000. S a egic Man-
agemen Jou nal, 25(10), 981-1004. h ps://doi.o g/10.1002/smj.397
Reichheld, F. F., & Sasse , W. E. (1990). Ze o de ec ions - Quali y comes
o se ices. Ha a d Business Re iew, 68(5), 105-111.
Rus , R. T., & Zaho ik, A. J. (1993). Cus ome sa is ac ion, cus ome
e en ion, and ma ke sha e. Jou nal o Re ailing, 69(2), 193-215.
h ps://doi.o g/10.1016/0022-4359(93)90003-2
Si ap acha, J., & Tocque , G. (2012). Cus ome expe ience, b and image
and cus ome loyal y in elecommunica ion se ices. Pape p esen ed
a he In Con Econ Bus Ma k Manag.
Small, H. (1973). Co‐ci a ion in he scien i ic li e a u e: A new measu e
o he ela ionship be ween wo documen s. Jou nal o he Ame -
ican Socie y o in o ma ion Science, 24(4), 265-269. h ps://doi.
o g/10.1002/asi.4630240406
Small, H. (1999). Visualizing science by ci a ion mapping. Jou nal o he
Ame ican Socie y o In o ma ion Science, 50(9), 799-813. h ps://doi.
o g/10.1002/(sici)1097-4571(1999)50:9<799::aid-asi9>3.0.co;2-g
Team, R. C. (2021). R: A language and en i onmen o s a is ical com-
pu ing.
Tsai, C. F., & Lu, Y. H. (2009). Cus ome chu n p edic ion by hyb id
neu al ne wo ks. Expe Sys ems wi h Applica ions, 36(10), 12547-
12553. h ps://doi.o g/10.1016/j.eswa.2009.05.032
Ullah, I., Raza, B., Malik, A. K., Im an, M., Ul Islam, S., & Kim, S. W.
(2019). A chu n p edic ion model using andom o es : Analysis o
machine lea ning echniques o chu n p edic ion and ac o iden-
i ica ion in elecom sec o . Ieee Access, 7, 60134-60149. h ps://doi.
o g/10.1109/access.2019.2914999
Van den Poel, D., & La i ie e, B. (2004). Cus ome a i ion analysis
o inancial se ices using p opo ional haza d models. Eu ope-
an Jou nal o Ope a ional Resea ch, 157(1), 196-217. h ps://doi.
o g/10.1016/s0377-2217(03)00069-9
an Eck, N. J., & Wal man, L. (2010). So wa e su ey: VOS iewe , a
compu e p og am o bibliome ic mapping. Scien ome ics, 84(2),
523-538. h ps://doi.o g/10.1007/s11192-009-0146-3
Ve beke, W., Dejaege , K., Ma ens, D., Hu , J., & Baesens, B. (2012). New
insigh s in o chu n p edic ion in he elecommunica ion sec o : A
p o i d i en da a mining app oach. Eu opean Jou nal o Ope a ional
Resea ch, 218(1), 211-229. h ps://doi.o g/10.1016/j.ejo .2011.09.031
Ve beke, W., Ma ens, D., Mues, C., & Baesens, B. (2011). Building
comp ehensible cus ome chu n p edic ion models wi h ad anced
ule induc ion echniques. Expe Sys ems wi h Applica ions, 38(3),
2354-2364. h ps://doi.o g/10.1016/j.eswa.2010.08.023
Wei, C. P., & Chiu, I. T. (2002). Tu ning elecommunica ions call de-
ails o chu n p edic ion: a da a mining app oach. Expe Sys ems
wi h Applica ions, 23(2), 103-112. h ps://doi.o g/10.1016/s0957-
4174(02)00030-1
Yan, E. J., & Ding, Y. (2011). Disco e ing au ho impac : A PageRank
pe spec i e. In o ma ion P ocessing & Managemen , 47(1), 125-134.
h ps://doi.o g/10.1016/j.ipm.2010.05.002
Zei haml, V. A., Be y, L. L., & Pa asu aman, A. (1996). The beha io al
consequences o se ice quali y. Jou nal o Ma ke ing, 60(2), 31-46.
h ps://doi.o g/10.2307/1251929
Zhang, J., Yu, Q., Zheng, F. S., Long, C., Lu, Z. X., & Duan, Z. G. (2016).
Compa ing keywo ds plus o WOS and au ho keywo ds: A case
s udy o pa ien adhe ence esea ch. Jou nal o he Associa ion o
In o ma ion Science and Technology, 67(4), 967-972. h ps://doi.
o g/10.1002/asi.23437
Zhao, D. Z., & S o mann, A. (2008). E olu ion o esea ch ac i i ies and
in ellec ual in luences in in o ma ion science 1996-2005: In oduc-
ing au ho bibliog aphic-coupling analysis. Jou nal o he Ame ican
Socie y o In o ma ion Science and Technology, 59(13), 2070-2086.
h ps://doi.o g/10.1002/asi.20910
Zupic, I., & Ca e , T. (2015). Bibliome ic me hods in managemen and
o ganiza ion. O ganiza ional Resea ch Me hods, 18(3), 429-472.
h ps://doi.o g/10.1177/1094428114562629
Managemen Le e s / Cuade nos de Ges ión 22/2 (2022) 97-121
