ABSTRACT
Clien s Seeking Red ess and i s In luence on
Rela ionship Quali y Dimensions: A Sou h
A ican Banking F aud Pe spec i e
Kholiswa Ma hiyase*
Depa men o Business
Managemen ,
Facul y o Business and Economic
Sciences,
Nelson Mandela Uni e si y,
Gqebe ha, Sou h A ica
[email p o ec ed]
* Co esponding au ho
D . Xolile An oni
Depa men o Managemen ,
Facul y o Comme ce,
Rhodes Uni e si y,
Makanda, Sou h A ica
[email p o ec ed]
P o . Chan al Roo man
Depa men o Business
Managemen ,
Facul y o Business and Economic
Sciences,
Nelson Mandela Uni e si y,
Gqebe ha, Sou h A ica
[email p o ec ed]
The Re ail and Ma ke ing Re iew
Volume 21, Issue 2, No embe 2025, Pages 51-68
Doi: h ps://doi.o g/10.5281/zenodo.17341018
Despi e he ise in digi al banking aud inciden s in Sou h A ica and i s in luence on banking clien ela ionships,
he e a e limi ed s udies conduc ed in Sou h A ica ha conside how seeking ed ess in luences ela ionship quali y
be ween banks and clien s who ha e encoun e ed banking aud. Seeking ed ess allows clien s o oice hei complain s
and imp o es he quali y o hei ela ionship wi h he bank. Clien s who seek ed ess when hey ha e encoun e ed
banking aud p o ide banks wi h he oppo uni y o co ec he mis ake and p e en e-occu ences o audulen
ansac ions in clien s' bank accoun s. This s udy in es iga es he in luence o seeking ed ess on ela ionship quali y
dimensions, based on he pe cep ions o banking clien s in Sou h A ica. A quan i a i e esea ch me hodology was
adop ed, and an online ques ionnai e was used o sample 399 Sou h A ican bank clien s who had p e iously been
a ec ed by aud. The s udy used s a is ical echniques such as con i ma o y ac o analysis (CFA) and s uc u al
equa ion modelling (SEM) o es he ela ionships be ween seeking ed ess and ela ionship quali y dimensions. The
esul s show ha seeking ed ess has a signi ican impac on ela ionship quali y dimensions, such as sa is ac ion and
loyal y. Banks should hus encou age banking clien s who ha e encoun e ed banking aud o seek ed ess o imp o e
he sa is ac ion and loyal y o banking clien s. The implica ions a e ha banks need a unc ioning complain managemen
sys em o ecei e aud complain s and p o ide sa is ac o y esponses o hei clien s. The esul s also showed ha
seeking ed ess and sa is ac ion does no di ec ly in luence loyal y.
Keywo ds: banking aud; clien s, complain beha iou ; ela ionship quali y, seeking ed ess.
Clien s Seeking Red ess … A Sou h A ican Banking F aud Pe spec i e 52
1. INTRODUCTION AND BACKGROUND
Banking aud con inues o ise in Sou h A ica despi e all he banks’ e o s o p e en such inciden s. Banking aud
escala ed by 45% in 2023, wi h inciden s ela ed o banking applica ions accoun ing o 60% o all banking aud
epo ed inciden s (SABRIC 2023). The ise in aud inciden s has also inc eased he numbe o complain s agains
a ious banks in Sou h A ica. In 2023 alone, he Ombudsman o Banking Se ices ecei ed abou 21.641 banking
aud complain cases, an 11% inc ease om he p e ious yea , 2022 (Ombudsman o Banking Se ices 2023). The
impac o banking aud ex ends beyond inancial losses, a ec ing bo h clien s and banks in a ious ways (Phi i e al.
2024:2). The e o e, banking clien s a e expec ed o complain abou banking aud so ha banks can be in o med abou
he inciden . Clien complain s a e a aluable sou ce o impo an ma ke in o ma ion ha i ms can use o co ec he
p oblem's oo cause and imp o e hei se ice deli e y o p oduc (Oluwasanmi e al. 2023:20).
Is anbulluoglu e al. (2017) explain clien complain beha iou (CCB) as a complex and mul idimensional concep .
I am and Iqbal (2023:1276) de ine CCB as an ac ion unde aken by clien s a e hey ha e expe ienced a nega i e
elemen o a p oduc o se ice deli e y. Oluwasanmi e al. (2023:21) also explain CCB as a p ocess ha cons i u es
a subse o clien s’ possible and dis inc esponses o pe cei ed dissa is ac ion a ound pu chasing p oduc s o du ing
se ice encoun e s and consuming p oduc s o se ices. Clien s who encoun e banking aud may exhibi di e en
ypes o complain beha iou (So ia e al. 2023:4). Se e al au ho s ha e ou lined di e en ypes o complain beha iou ,
including oicing a complain (which e e s o a clien exp essing hei dissa is ac ion wi h a i m); aking no ac ion and
emaining loyal; exi ing o ending he ela ionship wi h he bank; sp eading nega i e wo d o mou h; hi d-pa y
complain s and seeking ed ess (I am & Iqbal 2023; Is anbulluoglu e al. 2017; Oluwasanmi e al. 2023; So ia e al.
2023).
This s udy ocuses on clien s seeking ed ess. Ong e al. (2016:1) de ines ed ess as he p ocess o dealing wi h
complain s emana ing a e a clien has used a i m’s se ice o a e pu chasing p oduc s. Red ess is a p ocess whe e
a i m esol es o a ones o he p oblems ha may ha e occu ed a e clien s ha e used a i m’s se ice o a e he
pu chase o p oduc s (Mayombo 2014:4). Acco ding o F asque e al. (2021:1643), ed ess also p o ides clien s wi h
an oppo uni y o exp ess hei dissa is ac ion and add ess hei complain s. The e o e, seeking ed ess conce ns
p ocedu es h ough which indi iduals seek emedies o esponses o hei complain s (I am & Iqbal 2023:1310). Clien
complain s a e a aluable sou ce o impo an ma ke in o ma ion which i ms can use o co ec he oo cause o he
p oblem and o imp o e hei se ice deli e y (Oluwasanmi e al. 2023:21). Al hough complain s a e an insepa able
pa o he banking indus y complaining clien s ep esen clien s who a e willing o p o ide a bank wi h he oppo uni y
o imp o e and main ain business ansac ions (Pa ikh & Du 2022). Acco ding o Pio e al. (2024:1050), banks may
no be able o imp o e wi hou aluable eedback om complain s and may lose hei compe i i eness in he
ma ke place. This means ha complain s in he banking indus y mus be ecognised and esol ed o imp o e se ices
and p ocesses o g ow h and sus ainabili y. The e o e, complain s a e an impo an eedback mechanism o i ms o
moni o clien sa is ac ion wi h a i m’s p oduc s and se ices (Oluwasanmi e al. 2023:21). Encou aging clien s o seek
ed ess when hey encoun e aud may he e o e imp o e he ela ionship quali y be ween banks and clien s.
Rela ionship quali y dimensions include sa is ac ion, us and loyal y. Banks commi ed o deli e ing excellen clien
se ice should allow clien s o complain (Pio e al. 2024:1050; an De en e & Redda 2023:211). F asque e al.
(2021:1643) also asse ha “ i ms encou aging clien s o seek ed ess h ough specially designa ed channels can
inc ease clien sa is ac ion and loyal y”, and hus possibly imp o e he ela ionship be ween banks and hei clien s.
53 The Re ail and Ma ke ing Re iew: Vol21 Issue 2 (2025) ISSN:2708-3209
2. RESEARCH PROBLEM AND PURPOSE
Banking aud may nega i ely impac clien s’ ea lie pe cep ions o eeling secu e and p o ec ed by hei bank.
Clien s who no longe eel secu e and p o ec ed by hei bank may ul ima ely swi ch o ano he (Puen e-Ca azos e al.
2025:2). Mo eo e , gaining new clien s is mo e cos ly han main aining an exis ing clien ele (Rudd e al. 2022:70).
When i ms do no know he sou ces o dissa is ac ion among clien s, hese i ms may lose clien s and su e inancially
(Phi i e al. 2024:2). The e o e, complain s, pa icula ly seeking ed ess a e an impo an eedback mechanism o i ms
o moni o clien sa is ac ion wi h a i m’s p oduc s and se ice e ec i eness (F asque e al. 2021:1643). Complaining
clien s should no be seen as a nuisance, bu a he as hose ha a e in es ed in p o iding banks wi h an oppo uni y
o esol e ailu es and s eng hen ela ionships (Tax e al.1998).Gi en his backg ound, he pu pose o his s udy is o
in es iga e he in luence o seeking ed ess on ela ionship quali y dimensions ( us , sa is ac ion and loyal y) among
banking aud clien s. While s udies on complain beha iou exis in ela ion o se ice ailu e, se ice eco e y and
dissa is ac ion, e y ew s udies ha e examined complain s, speci ically a e aud inciden s whe e bo h inancial and
emo ional ha m a e heigh ened (Kassem, 2024; Alashwali e al. 2024; Phi i e al. 2024:2). The e a e also limi ed s udies
ha ha e in es iga ed he di ec in luence o seeking ed ess on ela ionship quali y dimensions such as us ,
sa is ac ion and loyal y. This s udy add esses ha gap by examining whe he clien s seeking ed ess ollowing a
banking aud inciden s eng hens o weakens he bank-clien ela ionship in e ms o sa is ac ion, us and loyal y.
This is pa icula ly impo an in Sou h A ica, whe e banking aud is on he ise, wi h he Ombudsman o Banking
Se ices epo ing app oxima ely 20,000 aud- ela ed complain s in 2023 alone. The e o e, his s udy ocuses on
banking aud and complain s in he Sou h A ican banking indus y. The complain beha iou p ocesses, and bank-
clien ela ionships a e ele an o his. On his basis, i is also concluded ha he s udy is based on ma ke ing li e a u e,
speci ically ela ionship ma ke ing and banking aud.
3. A LITERATURE OVERVIEW OF COMPLAINT BEHAVIOUR, SEEKING REDRESS
AND RELATIONSHIP QUALITY DIMENSIONS
A ew seminal clien complain models we e conside ed when de eloping he hypo heses o he s udy. The Day
and Landon’s (1977) model o clien complain beha iou ep esen s he di e ences be ween hose who ac ( ake
ac ion) and hose who do no ac (non-ac ion) when hey ha e a complain . The model u he di e en ia es be ween
aking public o p i a e ac ion, in o he wo ds, he di e en ways in which complain ac ion can be aken. In addi ion,
he model speci ies ac ions, such as seeking ed ess di ec ly, aking legal ac ion, complaining o a hi d-pa y o iends
and boyco ing a i m. Simila ly, he o he model ha was conside ed was he Singh’s (1988) model o clien complain
beha iou , whe e clien s' esponses a e classi ied in o h ee ca ego ies: oice esponses, p i a e esponses and hi d-
pa y esponses. Voice esponses a e di ec complain beha iou o he se ice i m, including seeking ed ess. Second
is p i a e esponses, which ela e o indi ec complain beha iou , including wo d-o -mou h communica ion. Thi d is
hi d-pa y esponses, which means aking legal ac ion o epo ing a complain o a egula o y body. The wo adop ed
models ha e some limi a ions due o hei s a ic and single pos -pu chase ocus (Mayombo 2014:4). Fo ins ance, he
Day and Landon (1977) and Singh (1988) models do no show he complain beha iou sequence clea ly. These
models only poin ou clien s' op ions when clien s encoun e dissa is ac ion and need o complain. This s udy adop ed
he wo models by conside ing clien s who ake ac ion and oice hei complain s, as ou lined by Singh (1988), no only
by oicing conce ns bu also by seeking ed ess wi h a bank. Seeking ed ess may p o ide an oppo uni y o banks o
co ec mis akes and p ese e he quali y o he bank-clien ela ionship, including aspec s such as us , sa is ac ion
Clien s Seeking Red ess … A Sou h A ican Banking F aud Pe spec i e 54
and loyal y owa ds he bank. This means ha a e clien s’ expe ience banking aud, hey may seek ed ess di ec ly
om he bank i s .
3.1 SEEKING REDRESS
Seeking ed ess occu s when clien s exp ess hei dissa is ac ion wi h an expec a ion ha hei complain s will be
esol ed o ha a i m will ix he p oblem (Ong & Teh 2016:1). F asque e al. (2021:1643) also asse ha clien s who
seek ed ess a e clien s who complain o a i m and equi e an explana ion. They expec o be compensa ed o he
loss, eques a e und o an exchange, o o he p oduc o be epai ed. Acco ding o I am and Iqbal (2023:1315),
seeking ed ess is mo e likely o be used ea ly and depends on clien s' abili y and willingness o exe cise i . Clien s
may seek ed ess di ec ly om a i m in ol ed in hei dissa is ac ion expe ience o om a hi d pa y. When a clien
seeks ed ess di ec ly om a i m, i enables he i m o acknowledge he complain and add ess he ma e . Mo eo e ,
a i m will ha e an oppo uni y o p e en he p oblems om ecu ing and p ese e he ela ionship wi h i s clien s
(Ca lson e al. 2023:1). Va ious ac o s in luence clien s’ decisions o seek o no seek ed ess. Examples include he
pe cei ed likelihood o success and clien s’ a i udes owa ds complaining migh play a signi ican ole (Blodge &
Ande son 2000:322; Tu aga e al. 2020:1076). Geo ge (2021) no es ha dissa is ied clien s choose o seek ed ess
based on he pe cei ed likelihood o success, hei a i udes owa ds complaining and pe cei ed jus ice. Simila ly,
Geo ge (2021) asse s ha clien s o en a oid seeking ed ess when hey belie e hei e o s will no yield a success ul
ou come.
3.2 RELATIONSHIP QUALITY DIMENSIONS
Dai (2025:155) explains ela ionship quali y as an o e all s a e o he ela ionship, iewed as a concep wi h se e al
dimensions. Ballan yne e al. (2003:14) a gue ha in an en i onmen whe e i ms o e simila p oduc s and se ices,
he quali y o an ongoing ela ionship becomes a way o gaining a compe i i e ad an age. Acco ding o Roy e al.
(2023), ela ionship quali y e e s o clien s’ pe cep ions o how well he ela ionship ul ils hei expec a ions,
p edic ions, goals and desi es. The mos commonly used dimensions o examine ela ionship quali y a e sa is ac ion
us and loyal y (Zie sman 2017:2). Dai (2025:155) s udied ela ionship quali y by using he ac o s o sa is ac ion, us
and loyal y om he pe spec i e o clien s. In he cu en s udy, ela ionship quali y is examined as clien s’ o e all
e alua ions o he long- e m in e ac ions o sa is ac ion, us and loyal y.
Sa is ac ion can be de ined in a ious si ua ions ha a e associa ed wi h p oduc s and se ices. Acco ding o Sabi
e al. (2014:1015), sa is ac ion occu s when a clien belie es ha a i m’s p oduc s and se ices mee hei expec a ions.
Singh e al. (2023:3458) also explain sa is ac ion as clien s’ pe cep ions ha a i m has ully, e icien ly and p omp ly
me hei expec a ions. Sa is ac ion is also he esul o a compa ison be ween a clien 's pu chase o a p oduc o
se ice and i s expec ed pe o mance e sus i s ac ual pe o mance (He zbe g e al. 2020). Sa is ac ion desc ibes a
gene ally posi i e expe ience o a pu chase, which includes he se ice and p oduc quali y(Singh e al. 2023:3458).
The sa is ac ion is based on a clien ’s judgemen and pe cep ion o whe he he bank has me o exceeded he clien ’s
expec a ions. One way o achie e s ong ela ionships is o ensu e ha clien s a e sa is ied. This is because dissa is ied
clien s will de ec , leading o he end o a ela ionship (Abdella & Ind adewa 2024:370).
T us is also one o he mos widely de ined and examined ela ionship quali y dimensions in ma ke ing li e a u e
(Dai 2025; Roy 2023; Telang & Somanchi 2016). T us la gely de e mines he es ablishmen and main enance o
ela ionships be ween clien s and i ms. U uena and Hildalgo (2015:6) explain us as a se o belie s in he compe ence
and in eg i y o he o he pa y. T us is also based on he le el o clien s’ con idence in a i m’s compe ence and
55 The Re ail and Ma ke ing Re iew: Vol21 Issue 2 (2025) ISSN:2708-3209
Seeking Red ess
Sa is ac ion
T us
Loyal y
H1
H2
H3
H4
H5
H6
pe o mance. T us has been gene ally accep ed as a c i ical elemen in posi i e ela ionships (Akhlaq & Ki an
2022:582). The s eng h and quali y o a ela ionship ely on he le el o us , and he highe he us le el, he s onge
he ela ionship will be, as us is i al in building quali y ela ionships, especially in se ice i ms o banks (Roy e al.
2023). Banks o e banking se ices and p oduc s ha a e complex and hus clien s need o ha e con idence ha he
banking p oduc s and se ices will deli e as p omised (Chen e al. 2012:7). De eloping clien s’ us is especially
essen ial whe e ulne abili y and isk o aud a e eminen (Ndubisi & Wah, 2005; Roy e al. 2023). As an example,
when clien s expe ience mone a y loss due o banking aud, hey migh no us hei banks o keep hei money and
o he inancial asse s sa e and may blame hem o he loss.
Se e al s udies ha e highligh ed a di ec link be ween clien sa is ac ion and clien loyal y (Kusumawa i & Rahayu
2020; Su ikun & Meep om 2021; Zephan 2018:2). Rosá io and Joaquim (2023:52) de ine loyal y as "a deeply held
commi men o a p oduc o se ice o a i m despi e si ua ional in luences and ma ke ing e o s ha ing he po en ial
o cause swi ching beha iou s”. Loyal y is also iewed as he con inued connec ion clien s ha e wi h hei banks
(Khadka & Maha jan 2017). Acco ding o Rosá io and Joaquim (2023:52), loyal y is ocused on he aspec o a
ela ionship wi h a i m, and a loyal clien is a pe son who makes egula pu chases, uses a i m’s p oduc s and se ices
and is ea ed as a pa ne . Loyal clien s also ha e an impac on he c ea ion o new p oduc s, ailo ed o hei indi idual
and speci ic needs. Loyal y is conside ed as he unc ion o he sha e o o al pu chases, he unc ion o buying
equency o buying pa e n o he unc ion o buying p obabili y (Kuusik 2017:5). Acco ding o Rahman and Ramli
(2016:607), loyal y has been ecognised as a key de e minan o clien s’ e en ion and plays a signi ican ole in long-
e m p o i abili y. A loyal clien will s ay wi h he same i m, is likely o engage in posi i e wo d-o -mou h communica ion,
ecommend he p oduc and in luence he decision beha iou o iends and amily membe s (Zhao e al. 2023). Loyal y
in a bank indica es clien s’ in en ion o e isi he bank and o e-use he bank’s p oduc s and se ices in he u u e
despi e a si ua ion ha may make hem wan o swi ch (Rosá io & Joaquim 2023:52). Based on he li e a u e abo e,
Figu e 1 illus a es a hypo hesised model be ween he independen a iable (Seeking ed ess), and he dependen
a iables (Sa is ac ion, T us and Loyal y).
FIGURE 1: HYPOTHESISED MODEL
Sou ce: Resea che ’s cons uc ion
Clien s Seeking Red ess … A Sou h A ican Banking F aud Pe spec i e 56
The ollowing hypo heses a e p oposed:
H1: The e is a signi ican ela ionship be ween seeking ed ess and sa is ac ion.
H2: The e is a signi ican ela ionship be ween seeking ed ess and us .
H3: The e is a signi ican ela ionship be ween seeking ed ess and loyal y.
H4: The e is a signi ican ela ionship be ween sa is ac ion and us .
H5: The e is a signi ican ela ionship be ween us and loyal y.
H6: The e is a signi ican ela ionship be ween sa is ac ion and loyal y.
4. RESEARCH DESIGN AND METHODOLOGY
A quan i a i e esea ch me hodology was adop ed. A quan i a i e me hodology was mos sui able gi en he s udy's
p ima y objec i e, which in ol ed de eloping hypo heses and es ing ela ionships be ween he selec ed a iables. The
popula ion o he s udy was he clien s o any bank ope a ing in Sou h A ica who had been p e iously a ec ed by aud
in hei bank accoun s and had epo ed his o he bank. The da a o he s udy was collec ed by a epu able esea ch
agency (CINT). CINT has a digi al esea ch da abase consis ing o mo e han 155 million pa icipan s in o e 130
coun ies, including Sou h A ica. The p o ile o he pa icipan s on he da abase includes banking and non-banking
clien s. The ques ionnai e was cap u ed on Mic oso Fo ms, and he URL link was sen o CINT o dis ibu e on behal
o he esea che s. CINT dis ibu ed he link o he su ey h ough i s online pla o m. The known popula ion o he
cu en s udy is 2.6 million Sou h A ican esponden s om he CINT da abase. Acco ding o Rahman (2023), when
he popula ion size is known, a sample size o 384 is su icien o a popula ion o 1 million. The e o e, a sample size
o 384 (minimum) esponden s a a con idence le el o 95% was conside ed su icien o he cu en s udy, o each a
maximum o 400 (Collis & Hussey 2014:198). A non-p obabili y judgmen al o pu posi e sampling was adop ed
because no sampling ame was a ailable and due o s ic bank-clien con iden iali y and he POPIA, he esea che
had no access o he bank-clien da abase. The sample size was 400 Sou h A ican bank clien s who had p e iously
been a ec ed by aud and complained o hei espec i e banks. Bo h seconda y da a and p ima y da a me hods we e
used o da a collec ion. Seconda y da a we e ga he ed by ex ensi ely e iewing a ious academic jou nals, a icles,
and ex books. Fo his s udy, he seconda y da a e iew ocused on opics, such as banking aud, complain
beha iou , complain managemen and ela ionship quali y. The p ima y da a was collec ed using a closed-ended
online ques ionnai e ocusing on Sou h A ican bank clien s who had p e iously been a ec ed by aud. Be o e p ima y
da a was collec ed, e hical issues we e conside ed. Fo his s udy, e hics app o al was ecei ed om he ele an
uni e si y’s commi ee. The su ey o he s udy included a consen o m, and he esponden s we e also gi en an
op ion o op ou o he su ey. To ensu e anonymi y, he esponden s we e no equi ed o p o ide hei pe sonal
in o ma ion, such as name, su name o ID numbe . The ques ionnai e had wo sec ions, A and B. Sec ion A used a
ca ego ical scale o collec in o ma ion abou demog aphics such as age, gende , popula ion g oup and educa ional
le el. Sec ion B used a i e-poin Like - ype scale con aining s a emen s ela ing o clien s ‘seeking ed ess’ and
‘ ela ionship quali y dimensions’ (sa is ac ion, us and loyal y). Exis ing i ems we e used bu a ew new i ems we e
also de eloped by he esea che based on li e a u e. O e all, 24 s a emen s we e posed in he ques ionnai e, 8
s a emen s ela ed o he ‘seeking ed ess’ a iable (F asque e al. 2021; Mayombo 2014; Resea che ’s cons uc ion),
s a emen s ela ing o ela ionship quali y dimensions: 5 s a emen s ela ed o sa is ac ion (Khadka & Maha jan 2017;
57 The Re ail and Ma ke ing Re iew: Vol21 Issue 2 (2025) ISSN:2708-3209
Resea che ’s cons uc ion; Rosá io & Joaquim 2023), 6 s a emen s ela ed o ‘ us ’ (Dai 2025; Ndubisi & Wah, 2005;
Resea che ’s cons uc ion; Roy e al. 2023) and 5 s a emen s ela ed o ‘loyal y’ (Akhlaq & Ki an 2022; Maxham &
Me emeye 2002; Puen e-Ca azos e al. 2025; Resea che ’s cons uc ion; Sousa & Voss 2009; U uena & Hidalgo
2015). This s udy’s ques ionnai e i ems we e de eloped by conside ing alidi y and eliabili y.
The da a was analysed using bo h desc ip i e and in e en ial s a is ical echniques. Va ious ypes o desc ip i e
analyses, such as equency and pe cen ages we e used o summa ise he clien s' demog aphic in o ma ion.
Desc ip i e analyses, including mean and s anda d de ia ion, we e used o summa ise he demog aphic a iables and
he ac o s o he s udy, namely seeking ed ess and ela ionship quali y. In e en ial s a is ics we e used o measu e
alidi y and eliabili y as well as es ablish co ela ions be ween a iables. Fu he mo e, he in e en ial s a is ical me hod
used o analyse he da a was he s uc u al equa ion model (SEM).
SEM was applied o examine he ela ionships among he a iables in he s udy. SEM is a mul i a ia e echnique
ha es s and e alua es mul i a ia e ela ionships be ween independen , media ing and dependen a iables (Hai e
al. 2014:565; Moqbel e al. 2020). SEM enables a es o complex pa e ns o ela ionships and he simul aneous es ing
o mul iple hypo heses wi hin a single comp ehensi e model, an app oach ha would o he wise equi e se e al
sepa a e analyses using o he s a is ical echniques (B yan, 2015; Sa hyana ayana & Mohanasunda am, 2024:562).
Hai e al. (2014:565) also highligh ha one o he ad an ages o using SEM is ha i allows he simul aneous use o
se e al indica o a iables pe cons uc , which leads o mo e alid conclusions on he cons uc le el. Acco ding o
Randall and Lomax (2022:136), SEM add esses measu emen e o by di ec ly modelling he e o e ms linked o
obse ed a iables. This app oach ensu es ha he es ima ed ela ionships be ween cons uc s emain accu a e and
ee om bias caused by measu emen inaccu acies, e ec i ely e lec ing ela ionships among pe ec ly eliable
a iables. SEM was he e o e used o examine he ela ionships among he a iables o he s udy.
Be o e SEM is assessed o o e all model i , a CFA is conduc ed o alida e he measu emen model. CFA
assesses whe he he obse ed a iables load signi ican ly in o hei espec i e la en , as speci ied in he model (Hai
e al. 2010:567-567). In o he wo ds, CFA is used o e alua e he s uc u e and alidi y o he measu emen model
(Hai e al. 2020:109). In CFA, eg ession weigh s a e de e mined. Reg ession weigh s (also e e ed o as ac o
loadings) ep esen he s eng h and di ec ion o he ela ionship be ween obse ed a iables and hei unde lying la en
ac o s (Hai e al. 2010:567-567; Tabachnick & Fidell, 2019:476). Once CFA is conduc ed, he SEM model is assessed
o o e all model i . Va ious indices we e used in he s udy o es he goodness o i o he p oposed hypo hesised
model. The goodness o i indica es how well he speci ied model ep oduces he obse ed co a iance ma ix among
he indica o i ems (Kline, 2023:141; Sa hyana ayana & Mohanasunda am, 2024:563). Goodness o i is e alua ed,
using a numbe o model i indices, which examines he ela ionships be ween he obse ed da a and he heo e ical
model which would be expec ed om he model (Hai e al. 2010:567; Sa hyana ayana & Mohanasunda am,
2024:563). The i s s ep in CFA is o assess con e gen alidi y, a e age ex ac ed a iance (AVE) and composi e
eliabili y, as Cheung e al. (2024:746) sugges . Con e gen alidi y is a concep o SEM ha assesses whe he di e en
measu es ha a e mean o measu e he same cons uc s co ela e well wi h each o he . I is con i med h ough high
loadings (Hai e al. 2021:78). The AVE is a measu e used o assess a cons uc ’s con e gen alidi y. I ep esen s
he mean o he squa ed loadings o he indica o s linked o la en cons uc and e lec s he cons uc ’s commonali y
(Hai e al. 2021:109). An AVE alue o a leas 0.50 is deemed accep able, indica ing ha a cons uc explains a leas
50% o mo e o he a iance in i s indica o s (B yan, 2015; Cheung e al. 2024:750). Thus, AVE measu es he
p opo ion o a iance cap u ed by a la en cons uc ela i e o ha a ibu ed o measu emen e o (Hai e al.
2021:77). I summa ises he o e all con e gen alidi y o a se o i ems. The high loadings o 0.50 and mo e con i med
Clien s Seeking Red ess … A Sou h A ican Banking F aud Pe spec i e 58
con e gen alidi y in he s udy. Once alidi y is checked, he nex s ep is o assess eliabili y. Composi e eliabili y
indica es he ex en o which a se o i ems consis en ly measu es he same la en cons uc (Hai e al. 2021:77). Unlike
C onbach’s alpha, composi e eliabili y does no assume equal indica o loadings. A high composi e eliabili y (abo e
0.70) sugges s ha he i ems a e well aligned in he measu ing ins umen . Composi e eliabili y uses ac o loadings
o con e gen alidi y o examine he in e nal consis ency o he cons uc s (Cheung e al. 2024:750; Kline, 2023:141).
5. 5 EMPIRICAL RESULTS
5.1 DEMOGRAPHIC DATA
The agency conduc ed he ec ui men p ocess o esponden s, based on he esea che s’ inclusion c i e ia. Using
he ques ion link, CINT in i ed sui able esponden s (Sou h A ican bank clien s) o pa icipa e in he s udy h ough
email and hei mobile app (OpinionAPP). Responden s used he enc yp ed link, p o ided by CINT, o access he online
ques ionnai e. Be o e comple ing he ques ionnai e, a consen le e and in oduc o y sec ion was p o ided, de ailing
he objec i e o he s udy. The link o he ques ionnai e was open o all Sou h A icans bank clien s in he pla o m. A
o al o 400 bank clien s in Sou h A ica comple ed he ques ionnai e. O he 400 comple ed ques ionnai es, 399 (one
had missing da a) we e use ul o da a analysis. Sec ion A o he ques ionnai e consis ed o demog aphic ques ions
and ga he ed da a, such as he gende , age, popula ion g oup and educa ional le el o he esponden s. Desc ip i e
s a is ics we e used o calcula e and summa ise he demog aphic da a o he esponden s. Table 1 p esen s he
demog aphic da a o he esponden s.
TABLE 1: DEMOGRAPHIC DATA OF RESPONDENTS
Gende
F equency
Pe cen (Rounded o )
Male
141
35.3
Female
258
64.7
To al
399
100
Age
<20 yea s old
9
2.3
20-29 yea s old
164
41.1
30-39 yea s old
137
34.3
40-49 yea s old
70
17.5
50-59 yea s old
11
2.8
60+ yea s old
8
2.0
To al
399
100
Popula ion
Asian
20
5.0
Black
283
70.9
Colou ed
30
7.5
Whi e
64
16.0
O he s
0
0.0
No willing o say
2
0.5
To al
399
100
59 The Re ail and Ma ke ing Re iew: Vol21 Issue 2 (2025) ISSN:2708-3209
Educa ional le el
F equency
Pe cen (Rounded o )
Ma ic ce i ica e (G ade 12)
92
23.1
Diploma
54
13.5
Pos g adua e diploma
12
3.0
Bachelo 's deg ee
158
39.6
Honou s deg ee
31
7.8
Mas e ’s deg ee
45
11.3
Doc o a e
7
1.8
To al
399
100
Sou ce: S a is ical esul s
As shown in Table 1, he majo i y o esponden s we e emales (64.7%), and 35.3% we e males. A la ge sha e o
he sample consis ed o younge bank esponden s wi h hose aged 20 and 29 yea s making up 41.1%. Responden s
younge han 20 yea s old cons i u ed 2.3% o he sample. Those aged 30 and 39 yea s made up 34.3%, while
esponden s be ween 40 and 49 yea s accoun ed o 17.5%. The 50 and 59 age g oup ep esen ed app oxima ely
2.8% and esponden s aged 60 yea s and abo e comp ised app oxima ely 2.0% o he sample. In e ms o he
popula ion g oup, Black esponden s we e he majo i y a 70.9%, ollowed by he Whi e popula ion g oup (16%) and
he Colou ed popula ion g oup (7.5%). The Asian popula ion g oup ep esen ed 5.0% o he sample, and a small
ac ion o esponden s (0.5%) chose no o disclose hei popula ion g oup. Las ly, he majo i y o esponden s selec ed
a bachelo ’s deg ee as hei highes quali ica ion (39.6%), ollowed by Ma ic ce i ica e (G ade 12) (23.1%), Diploma
(13.5%), Mas e s (11.3%), Honou s deg ee (7.8%), Pos g adua e diploma (3.0%), and Doc o a e deg ee (1.8%).
5.2 VALIDITY AND RELIABILITY
The alidi y and eliabili y o he measu ing ins umen we e assessed using CFA.
TABLE 2: VALIDITY AND RELIABILITY RESULTS
I ems
S a emen
Loading
P- alues
AVE
CR
SR7
I eques ed he bank o esol e he ma e .
0.922
***
0.692
0.878
SR3
I eques ed he bank o in es iga e he audulen ansac ion.
0.880
***
SR1
I complained di ec ly o he bank.
0.868
***
SR8
I eques ed he bank o ensu e he e will be no e-occu ence o he same p oblem.
0.850
***
SR2
I eques ed an explana ion om he bank.
0.843
***
SR5
I eques ed he bank o eco e my s olen money.
0.823
***
SR4
I eques ed a e und om he bank.
0.725
***
SR6
I eques ed he bank o imp o e he secu i y ela ed o digi al banking acili ies.
0.720
***
Sa 1
I am sa is ied wi h he bank.
0.930
***
0.792
0.948
Sa 2
The se ices o he bank mee my expec a ions.
0.912
***
Sa 4
I am sa is ied wi h he bank’s p oduc s and se ices.
0.895
***
Sa 5
O e all, my expe ience wi h he bank has been good.
0.892
***
Sa 3
The digi al banking op ions o e ed by he bank, o example in e ne o mobile banking,
mee my expec a ions.
0.817
***
T u6
I ha e con idence in how he bank deli e s i s se ices.
0.931
***
0.768
0.934
T u5
I belie e ha he bank deli e s i s se ices as expec ed.
0.930
***
T u4
I belie e ha he bank’s p oduc s pe o m as expec ed.
0.906
***
T u3
I us he bank wi h keeping my money sa e.
0.902
***
T u1
I us he bank's digi al banking acili ies.
0.811
***
T u2
I us he bank's s a .
0.764
***
Clien s Seeking Red ess … A Sou h A ican Banking F aud Pe spec i e 66
Mayombo, M.B. 2014. The in luence o cus ome complain beha iou , i m esponses and se ice quali y on
cus ome loyal y o mobile elephone subsc ibe s in Uganda. Uni e si y o Nai obi, Kenya.
Menon, V., G o e , S., Gup a, S., Indu, P.V., Chacko, D. & Vidhukuma , K. 2025. A p ime on eliabili y es ing o a
a ing scale. Indian Jou nal o Psychia y, 67(7): 725-729.
Moqbel, M., Gudu u, R. & Ha un, A. 2020. Tes ing media ion ia indi ec e ec s in PLS-SEM: A social ne wo king si e
illus a ion. Da a Analysis Pe spec i es Jou nal, 1(3): 1–6.
Ndubisi, N.O. & Wah, C.K. 2005. Fac o ial and disc iminan analyses o he unde pinnings o ela ionship ma ke ing
and cus ome sa is ac ion. In e na ional Jou nal o Bank Ma ke ing, 23(7): 542–57.
Oluwasanmi, O., Adeyemi, E.A. & Akanji, E.0. 2023. Cus ome complain managemen , se ice quali y and ma ke ing
pe o mance in he se ice indus y. Ad ances in Managemen , 22(1): 20-39.
Ombudsman o Banking Se ices. 2023. Banking aud [Online]. A ailable om: h ps://n osa.co.za/ombudsman-
sounds-ala m-on- i ual-banking-ca d- aud/ [Accessed 12 Ap il 2024].
Ong, C. E. & Teh, D. 2016. Red ess p ocedu es expec ed by consume s du ing a business- o-consume e-
comme ce dispu e. Elec onic Comme ce Resea ch and Applica ions, 19(4):1–13.
Pa ikh, P. & Du , C.S. 2022. Impac s o s a demog aphics on complain managemen beha iou in i e-s a ho els
in Dubai. In e na ional Hospi ali y Re iew, 36(2): 199-219.
Phi i, J., La hengwa, T. & Segooa, M.A. 2024. Online banking aud de ec ion: A compa a i e s udy o cases om
Sou h A ica and Spain. Sou h A ican Jou nal o In o ma ion Managemen , 26(1):1763.
Pio, P. G. C., Sigahi, T., Rampasso, I. S., Sa olo, E. G., Se a im, M. P., Quelhas, O. L. G., Leal Filho, W. & Anholon,
R. 2024. Complain managemen : Compa ison be ween adi ional and digi al banks based on ISO 9001
equi emen s. In e na ional Jou nal o P oduc i i y and Pe o mance Managemen , 73(4): 1050–1070.
Puen e-Ca azos, C.D., Ca azos-A oyo, J. & Puen e-Díaz, R. 2025. The impac o pe cei ed inancial secu i y in
neobanks on cus ome e en ion and he cos o swi ching o a adi ional bank: he media ing ole o us in
neobanks. B azilian Jou nal o Business Managemen , 27(1):2024-0079.
Rahman, M.M. 2023. Sample Size De e mina ion o Su ey Resea ch and Non-P obabili y Sampling Techniques.
Jou nal o Economics, Business and En ep eneu ship, 2(2): 55–60.
Rahman, M.A. & Ramli, M.F. 2016. The in luence o ela ionship quali y on cus ome loyal y in he dual-banking
sys em in he no he n s a es o peninsula Malaysia. P ocedia-Social and Beha io al Sciences, 219(2):606-613.
Randall, E.S. & Lomax, R.G. 2022. 3 d edi ion. A Beginne 's Guide o S uc u al Equa ion Modelling. Rou lage p ess,
London.
Rosá io, A. & Casaca, J.A. 2023. Rela ionship ma ke ing and cus ome e en ion-a sys ema ic li e a u e
e iew. S udies in Business and Economics, 18(3): 44-66.
67 The Re ail and Ma ke ing Re iew: Vol21 Issue 2 (2025) ISSN:2708-3209
Roy, S.K., Singh, G., G une , R.L., Dey, B.L., Shabnam, S., Muhammad, S.S. & Quaddus, M. 2023. Rela ionship
quali y in cus ome -se ice obo in e ac ions in Indus y 5.0: An analysis o alue ecipes. In o ma ion Sys ems
F on ie s, 1-20.
Rudd, H.D., Huo, H. & Xu, G. 2022. Imp o ed chu n causal analysis h ough es ained high-dimensional ea u e
space e ec s in inancial ins i u ions. Human-Cen ic In elligen Sys ems, 2(3): 70-80.
Sabi , R.I., Gha oo , O., Akh a , N., Ha eez, I. & Rehman, A.U. 2014. Fac o s a ec ing cus ome sa is ac ion in
banking sec o o Pakis an. In e na ional Re iew o Managemen and Business Resea ch, 3(2): 1014-1025.
SABRIC. 2023. Annual c ime s a s [Online], A ailable om: h ps://www.sab ic.co.za/media-and-news/p ess-
eleases/ [Accessed 16 Ma ch 2024].
Sa hyana ayana, S. & Mohanasunda am, T. 2024. Fi indices in s uc u al equa ion modeling and con i ma o y ac o
analysis: epo ing guidelines. Asian Jou nal o Economics, Business and Accoun ing, 24(7): 561-577.
Shan ikuma, S. 2018. Validi y, eliabili y and gene alisabili y [Online]. A ailable om:
h ps://www.heal hknowledge.o g.uk/con en / alidi y- eliabili y-and-gene alisabili y [Accessed 19 Ma ch 2024].
Singh, J. 1988. Consume complain in en ions and beha io : De ini ional and axonomical issues. Jou nal o
Ma ke ing, 52(1): 93-107.
Singh, V., Sha ma, M.P., Jayap iya, K., Kuma , B.K., Chande , M.A.R.N. & Kuma , B.R. 2023. Se ice quali y,
cus ome sa is ac ion and cus ome loyal y: A comp ehensi e li e a u e e iew. Jou nal o Su ey in Fishe ies
Sciences, 10(45): 3457-3464.
Smi h, B. 1998. Buye ‐selle ela ionships: Bonds, ela ionship managemen , and sex‐ ype. Canadian Jou nal o
Adminis a i e Sciences, 15(1): 76-92.
So ia, R.D., A i , A.Z. & A uan, D.T.H. 2023. The e ec o a ibu ions and ailu e se e i y on consume complain
beha io s in sha ing economy. Cogen Business & Managemen , 10(3): 2275848.
Sousa, R. & Voss, C.A. 2009. The e ec s o se ice ailu es and eco e y on cus ome loyal y in e‐se ices: An
empi ical in es iga ion. In e na ional Jou nal o Ope a ions and P oduc ion Managemen , 29(8): 834-864.
Su ikun, C. & Meep om, S. 2021. Examining he e ec o pe cei ed quali y o au hen ic sou eni p oduc , pe cei ed
alue, and sa is ac ion on cus ome loyal y. Cogen Business and Managemen , 8(1): 1976468.
Tabachnick, B.G. & Fidell, L.S. 2019. Using mul i a ia e s a is ics. 7 h edi ion. Pea son.
Tanko, D., Tunce , T., Dogan, S. & Akbal, E. 2022. Reg ession me hods o social media da a analysis. Mugla
Jou nal o Science and Technology, 8(1):31-40.
Tax, S.S., B own, S.W. & Chand asheka an, M. 1998. Cus ome e alua ions o se ice complain expe iences:
Implica ions o ela ionship ma ke ing. Jou nal o Ma ke ing, 62(2):60-76.
Clien s Seeking Red ess … A Sou h A ican Banking F aud Pe spec i e 68
Telang, R. & Somanchi, S. 2016. Secu i y, audulen ansac ions and cus ome loyal y: A ield s udy.Ca negie
Mellon Uni e si y, USA.
Tu aga, N.K., Pa ibandla, J.K., Tu aga, M. & Banda u, S.R. 2020. Fac o s in luencing o complain by i s ime
cus ome s and loyal cus ome s in DTH se ices. J C i Re , 7(12):1076-1080.
U uena, A. & Hidalgo, A. 2015. Jus ice and emo ions in se ice eco e y: A complain in B2C e-comme ce.
In e na ional Jou nal o Enginee ing Managemen and Economics, 5(1-2):3-18.
an De en e , M. & Redda, E. H. 2023. Cus ome loyal y and us in Sou h A ican e ail banking. Banks and Bank
Sys ems, 19(2): 211–222.
Xia, Y. & Yang, Y. 2019. RMSEA, CFI, and TLI in s uc u al equa ion modeling wi h o de ed ca ego ical da a: The
s o y hey ell depends on he es ima ion me hods. Beha io esea ch me hods, 51(1): 409-428.
Zephan, N. 2018. Rela ionship be ween cus ome sa is ac ion and cus ome loyal y. Case s udy: Hil on ho el
Yaoundé, Came oon. (Unpublished disse a ion), Cen ia Uni e si y o Applied Science, Came oon.
Zhao, S., Jiang, Z., Li, Y. & Meng, J. 2023. How does Wo d-o -mou h Ma ke ing A ec Cus ome Loyal y. Ad ances
in Economics, Managemen and Poli ical Sciences, 10(1): 51-57.
Zie sman, P.J. 2017. The ela ionship be ween ela ionship quali y cons uc s, o e all sa is ac ion and loyal y: An
eme ging ma ke pe spec i e o SME banking cus ome s. (Unpublished disse a ion), Uni e si y o P e o ia, Sou h
A ica.
