The Taguchi approach to large-scale experimental designs: A powerful and efficient tool for advancing marketing theory and practice

Mo e , Jo dan W.; Fennell, Pa ick; Ha meling, Colleen M.; Sheehan, Daniel; Bleie ,
Alexande
A icle — Published Ve sion
The Taguchi app oach o la ge-scale expe imen al
designs: A powe ul and e icien ool o ad ancing
ma ke ing heo y and p ac ice
Jou nal o he Academy o Ma ke ing Science
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Mo e , Jo dan W.; Fennell, Pa ick; Ha meling, Colleen M.; Sheehan, Daniel;
Bleie , Alexande (2024) : The Taguchi app oach o la ge-scale expe imen al designs: A powe ul
and e icien ool o ad ancing ma ke ing heo y and p ac ice, Jou nal o he Academy o Ma ke ing
Science, ISSN 1552-7824, Sp inge US, New Yo k, NY, Vol. 53, Iss. 3, pp. 949-954,
h ps://doi.o g/10.1007/s11747-024-01059-0
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/323655
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by/4.0/
Vol.:(0123456789)
Jou nal o he Academy o Ma ke ing Science (2025) 53:949–954
h ps://doi.o g/10.1007/s11747-024-01059-0
METHODOLOGICAL PAPER
The Taguchi app oach ola ge‑scale expe imen al designs: Apowe ul
ande icien ool o ad ancing ma ke ing heo y andp ac ice
Jo danW.Mo e 1· Pa ickFennell2· ColleenM.Ha meling3 · DanielSheehan1· Alexande Bleie 4
Recei ed: 1 Janua y 2024 / Accep ed: 27 Sep embe 2024 / Published online: 5 No embe 2024
© The Au ho s 2024, co ec ed publica ion 2024
Abs ac
Cu en esea ch o en elies on na owly ocused expe imen al me hods ha add ess jus a ew independen a iables o
co ela ional designs, despi e calls o u u e esea ch o ake big-pic u e pe spec i es ha o e eal-wo ld applicabili y and
causal e idence. This dispa i y likely e lec s he cons ain s imposed by he need o ex ensi e esou ces o conduc b oad,
causal examina ions. To b idge his gap, he cu en a icle p esen s he Taguchi app oach o la ge-scale expe imen al design,
which emains no ably unde u ilized in ma ke ing esea ch despi e being well-es ablished in o he ields. I s e ec i eness
s ems om he obus ca alog o expe imen al design ub ics ha can inco po a e many di e en independen a iables
sys ema ically and e icien ly. The causal and e icien expe imen al op ion o b oad scopes o in es iga ion emb aces he
embeddedness o independen a iables and hus can help build ma ke ing heo y and ad ance p ac ice. This a icle de ails
he undamen als o he Taguchi app oach, i s ela i e ad an ages, and a h ee-s ep implemen a ion p ocess.
Keywo ds Expe imen al design· Hypo hesis es ing· Taguchi me hodology· F ac ional ac o ial design· Ma ke ing heo y
The mode n ma ke ing landscape is complex. Ye adi ional
academic esea ch designs end o abs ac eali y o p ag-
ma ically educe i s complexi y. Fo example, expe imen s
o e s ong e idence o causali y, bu 98% o expe imen al
s udies published in op ma ke ing jou nals in 2023 manipu-
la ed only one o h ee independen a iables.1 This na ow
ocus o e s high in e nal alidi y o es a limi ed se o
hypo heses bu o en ails o accoun o he complex web
o ac o s in luencing eal-wo ld ou comes, educing ex e -
nal alidi y. Fu he , s e ile lab se ings equi e omi ing o
holding cons an in luen ial a iables, which isks o e - o
unde -es ima ing ocal e ec s occu ing ou side o expe i-
men al acuums and obscu ing ecommenda ions o con-
sume s, i ms, and socie y. Thus, cu en scien i ic con en-
ions and o e eliance on na ow esea ch scopes may hinde
heo e ical and p ac ical ad ancemen s (Kinde mann e al.,
2024). As a complemen a y me hod, we p esen he Taguchi
app oach—a causal and e icien expe imen al op ion o
b oad scopes o in es iga ion ha emb aces he embedded-
ness o independen a iables in eal-wo ld en i onmen s.
I allows esea che s o p agma ically and p ecisely s udy
complex concep s while es ablishing causali y, a ea o en
challenging o accomplish wi h adi ional echniques.
Jo dan W. Mo e and Pa ick Fennell con ibu ed equally.
Ma k Hous on se ed as edi o o his a icle.
* Colleen M. Ha meling
[email p o ec ed]
Jo dan W. Mo e
jwmo e @uky.edu
Pa ick Fennell
p [email protected]
Daniel Sheehan
[email protected]
Alexande Bleie
[email p o ec ed]
1 Ga on College o Business andEconomics, Uni e si y
o Ken ucky, Lexing on, KY40506, USA
2 Coles College o Business, Kennesaw S a e Uni e si y,
Kennesaw, GA30144, USA
3 D . Pe sis E. Rockwood School o Ma ke ing, College o Business,
Flo ida S a e Uni e si y, Tallahassee, FL32306, USA
4 F ank u School o Finance & Managemen ,
60322F ank u amMain, Ge many
1 We e iewed esea ch published in he Jou nal o Consume
Resea ch, Jou nal o Ma ke ing, Jou nal o Ma ke ing Resea ch,
Jou nal o he Academy o Ma ke ing Science, and Ma ke ing Sci-
ence. We also ound ha 97% o expe imen al s udies used single-
ac o o ull- ac o ial designs and 3% used a conjoin o ac ional
ac o ial app oach.
950 Jou nal o he Academy o Ma ke ing Science (2025) 53:949–954
Panel A:
Panel B:
S imuli
Numbe
Independen Va iables and
Manipula ion Le els
–
S imulus #1 S imulus #12
Panel C:
–
Roy (1990) and Mini ab’s in e ac i e design ool walks
esea che s h ough adap ing Taguchi’s ca alog designs o
Le el1 Le el2
A
Absen P esen
B
SolidPa e ned
C
Rec angleCi cle
D
O ganicNa u al
E
Passi e oice Ac i e oice
F
Absen P esen
G
4oz. 11 oz.
H
Absen P esen
I
Absen P esen
J
Absen P esen
K
Absen P esen
Manipula ions
Independen
Va iables
S imuli
Numbe
Independen Va iables and
Manipula ion Le els
Fig. 1 P oduc package design illus a i e example
951Jou nal o he Academy o Ma ke ing Science (2025) 53:949–954
Fundamen als andad an ages
o  heTaguchi app oach inma ke ing
esea ch
The Taguchi app oach o igina ed in enginee ing, whe e
i was de eloped o isola e and es independen a iables
wi hin highly complex sys ems and ul ima ely iden i y op i-
mal con igu a ions while ensu ing esou ce e iciency (Tagu-
chi, 1986). Ag icul u e and medical esea ch also use his
app oach o ackle ques ions ela ed o p oduc quali y con-
ol, a ie y selec ion, and p ocess op imiza ion. Despi e i s
sui abili y o b oad-scope in es iga ions, Taguchi-designed
s udies ha e only ecen ly eme ged in ma ke ing (e.g.,
Bleie e al., 2019). The powe o Taguchi’s app oach lies
in i s unique, accessible, and obus ca alog o expe imen-
al design ub ics, which se e as empla es o esea che s.
These ub ics speci y he minimum equi ed numbe and
speci ic con igu a ions o expe imen al condi ions needed
o es ima e he main e ec s o nume ous independen a i-
ables on a dependen a iable, equi ing only a ac ion o
he condi ions needed in ull- ac o ial designs. The ub ics
a e ac ional ac o ial designs ha p esc ibe he manipula-
ion le els o each independen a iable o each condi ion
and a e o hogonally balanced so ha (1) each manipula-
ion le el o each independen a iable and (2) each possible
o de ed pai o independen a iables a each le el appea
he same numbe o imes in he o e all design. This ca e-
ully cons uc ed balance enables e icien and eliable es i-
ma ions o main e ec s and a selec numbe o in e ac ion
e ec s while educing impac s o non- heo ized in e ac ion
e ec s (Roy, 1990). This app oach can be applied o a ious
con ex s (e.g., social media communica ions, sales aining,
loyal y p og ams, ood consump ion).2 As an illus a ion,
conside p oduc packaging design a ma ke ing p oblem
wi h 11 coexis ing independen a iables (e.g., logo shape,
size) ha migh a ec a dependen a iable, such as pu chase
in en ions (Fig.1, Panel A). Wi h his example, we discuss
ou ad an ages o he Taguchi app oach, highligh ing bene-
i s o e o he me hods (e.g., ull- ac o ial expe imen s, con-
join analyses, obse a ional da a analyses, me a-analyses).
Fi s , he Taguchi app oach o e s signi ican esou ce
sa ings compa ed wi h o he app oaches. Full- ac o ial
expe imen s a e bes sui ed o in es iga ing he e ec s
o a small se o independen a iables; Taguchi-designed
s udies a e especially adep o b oad in es iga ions due o
hei educed numbe o expe imen al condi ions. In ou
example, Taguchi’s ca alog speci ies ha he main e ec s
o 11 package design elemen s manipula ed a wo le els
can be in es iga ed wi h 12 expe imen al condi ions (e.g.,
L12 design ub ic; Fig.1, Panel B). This equi es access
o oughly 2,400 pa icipan s, equa ing o cos s o a ound
$6,000.3 In con as , a ull- ac o ial, be ween-subjec s expe -
imen would equi e 2,048 expe imen al condi ions (211),
access o o e 400,000 pa icipan s—2.5 imes he numbe
a ailable on leading pla o ms—and a budge exceeding
$1,000,000. Because each pa icipan is only exposed o one
s imulus, a Taguchi-designed s udy is also economical in
e ms o pa icipan a en ion and s amina, especially ela i e
o adi ional conjoin app oaches ha use wi hin-subjec s
designs and expose pa icipan s o mul iple s imuli (e.g., 12
p oduc package designs). Cu en ecommenda ions on con-
join s udies e en limi he numbe o independen a iables
explo ed o six o se en. Thus, he esou ce e iciencies o
he Taguchi app oach allow heo y and in ui ion o d i e he
selec ion o a iables wi h educed biases ela ed o pa ici-
pan p ac ice and a igue, enabling in es iga i e scopes ha
we e p e iously di icul o achie e wi hou comp omising
causali y.
Second, he Taguchi app oach enhances esea che s’
abili ies o demons a e causali y by es ing maximally
di e se, expe imen ally manipula ed con igu a ions wi h
andom pa icipan assignmen . In con as , seconda y da a
se s a e limi ed o obse able a iables collec ed in he pas ,
which a e u he cons ained by he no ms and con en ions
o he speci ic con ex (e.g., packaging no ms wi hin a pa -
icula indus y o i m and wha he i m deemed wo hy o
acking). The e o e, he con igu a ion di e si y achie able
h ough he Taguchi app oach can help esea che s unco e
ela ionships ha migh emain hidden in seconda y da a-
se s due o hese cons ain s. Taguchi-designed s udies a e
hus adep a modeling eal-wo ld complexi y. Manipula ing
co-occu ing d i e s o he dependen a iable, a he han
holding hem cons an ac oss condi ions (as is he case in a-
di ional expe imen s), may p o ide mo e accu a e es ima es
o each independen a iable’s con ibu ion o a iance in
he dependen a iable. Taguchi-designed s udies mee he
s ingen condi ions equi ed o es ablish causali y and sup-
po he in es iga ion o he impac s o many independen
a iables in andem, a he han in isola ion h ough sin-
gle o sequen ial na ow, ull- ac o ial s udies. This allows
esea che s o subs an ia e b oad cons uc s (e.g., package
design expe ience) while isola ing he independen a iables
ha d i e obse ed changes (e.g., package design elemen s).
Thi d, esea che s can causally in es iga e selec , heo-
ized wo-way in e ac ions wi hou gene a ing non- heo ized
wo-way o highe -o de in e ac ion e ec s. Taguchi’s
2 Fo addi ional examples o Taguchi-designed expe imen s, see
h ps:// inyu l. com/ bd 4 5c.
3 The sample size is based on 200 pa icipan s pe expe imen al con-
di ion, wi h cos s a $2 pe pa icipan plus a 25% P oli ic o 40%
MTu k ee.
952 Jou nal o he Academy o Ma ke ing Science (2025) 53:949–954
designs a e ca e ully balanced so ha he impac s o non-
heo ized in e ac ion e ec s (e.g., 3-way, 5-way, e c.) a e
uni o mly dis ibu ed ac oss all condi ions. The heo ized
main and in e ac ion e ec s hus can be clea ly iden i ied
and es ima ed. Conside ing Taguchi’s ca alog,4 esea che s
can explo e up o 20 o he 55 possible wo-way in e ac-
ions among he 11 package design elemen s in ou exam-
ple, using anywhe e om 16 o 32 expe imen al condi ions
(e.g., L16 ca alog design allows up o 4 in e ac ions, and L32
allows up o 20). A ull- ac o ial design o his size would
gene a e 55 wo-way in e ac ion e ec s and 1,981 highe -
o de in e ac ions. These, ypically no heo ized, highe -
o de in e ac ions can al e and complica e he in e p e a ion
o he main e ec s. Fu he , adi ional conjoin s udies do
no ypically accommoda e any in e ac ions, because hey
equi e bo h mo e sophis ica ed modeling echniques and
mo e da a o es ima ions.
Fou h, wi h he Taguchi app oach, esea che s can con-
solida e hough while being o wa d-looking. Designing
no el b oad-scope in es iga ions is easible using he Tagu-
chi app oach, conside ing i s manageable da a collec ion
and esponden equi emen s. This g an s esea che s he
oppo uni y o consolida e di e se heo e ical pe spec i es
o build heo y using p e iously unconside ed independen
a iables and ela ionships. Al hough me a-analyses can
consolida e a chi al hough , hey equi e ma u e domains
wi h su icien p e iously published esea ch. Al e na i ely,
he Taguchi app oach can be applied in eme ging and no el
domains because i en ails new da a collec ion. Fo ou
p oduc packaging example, esea che s could in es iga e
di e en se s o independen a iables ha a ious heo ies
p io i ize, like cogni i e o a ec i e in o ma ion p ocessing
(e.g., a ounde na a i e, ac i e e sus passi e desc ip ion)
o sensing (e.g., g aphics, logo shape), in a single s udy.
They also could in es iga e independen a iables ha ha e
no been collec ed o es ed be o e (e.g., he p esence o a
QR code).
Implemen ing heTaguchi app oach
oexpe imen al design
The Taguchi app oach o expe imen al design can be imple-
men ed in he lab o he ield, wi h a sys ema ic h ee-s ep
p ocess. Be o e beginning his p ocess, esea che s will
iden i y hei dependen a iable(s) o in e es along wi h
a po en ially la ge se o independen a iables and hei
manipula ion le els by e iewing ele an heo y and
p ac ice, as well as quali a i e esea ch (e.g., expe o cus-
ome in e iews). The independen a iables mus be non-
nes ed and con ollable so ha hey can be manipula ed inde-
penden ly o one ano he in he expe imen (e.g., g aphic
colo ulness would be nes ed wi hin he p esence [ e sus
absence] o g aphics).
S ep 1: Selec heapp op ia e Taguchi expe imen al
design ub ic
Conside ing he numbe o independen a iables iden-
i ied o in es iga ion (e.g., 11) and hei manipula ion
le els (e.g., 2), esea che s selec an expe imen al design
ub ic om Taguchi’s ca alog. This ca alog ea u es a ious
ub ics ha accommoda e anywhe e om 4 (e.g., L4) o 26
(e.g., L54) di e en independen a iables, each wi h 2–8
manipula ion le els, he eby suppo ing a wide a ie y o
esea ch con ex s ac oss di e en domains. These ub ics
speci y he numbe o equi ed s imuli (e.g., p oduc pack-
age con igu a ions). In ou p oduc packaging scena io, he
L12 design allows he simul aneous in es iga ion o he 11
package design elemen s a wo di e en manipula ion le els
wi h only 12 expe imen al s imuli. Fu he , hese ub ics ell
esea che s how o con igu e each s imulus, indica ing he
manipula ion le el o each independen a iable in each
s imulus (Fig.1, Panel B). These ub ics a e o hogonally
balanced, meaning ha each manipula ion le el o each
package design elemen and each possible o de ed pai o
elemen s a each le el appea s he same numbe o imes in
he o e all design. Fo ins ance, logo shape Le el 1 ( ec-
angle) appea s 6 imes, and Le el 2 (ci cle) also appea s
6 imes. Logo shape Le el 1 ( ec angle) wi h backg ound
Le el 2 (pa e ned) appea s 6 imes, and he e e se appea s
6 imes (ci cle logo shape wi h solid backg ound). Tools
o design selec ion a e a ailable in he Mini ab s a is ical
so wa e, JMP by SAS, o XLSTAT.5
Resea che s can also in es iga e a limi ed se o wo-way
in e ac ion e ec s be ween he manipula ed a iables, wi h
ca e ul conside a ion o he independen a iables and hei
column assignmen s in he ub ic (Roy, 1990; see Mini ab’s
in e ac i e design ool). In ou example, he heo y could
p edic in e ac ions be ween he backg ound (e.g., solid, pa -
e ned) and g aphics (e.g., p esen , absen ), as well as he
backg ound and logo shape (e.g., ec angle, ci cle). Wi h
he L16 ub ic, we can in es iga e he main e ec s o he 11
package design elemen s and hese wo in e ac ion e ec s
wi h 16 expe imen al condi ions (Compa e Fig.1, Panel B
and Panel C).
4 You can ind Taguchi’s ca alog on Mini ab’s websi e: h ps:// www.
mini ab. com.
5 h ps:// www. mini ab. com; h ps:// www. jmp. com; h ps:// www.
xls a . com

953Jou nal o he Academy o Ma ke ing Science (2025) 53:949–954
S ep 2: Execu e andanalyze heTaguchi‑designed
expe imen
A e selec ing he app op ia e design ub ic, s imuli a e
designed and p oduced acco dingly (e.g., 12 s imuli; Fig.1,
Panel B), hen andomly p esen ed o pa icipan s in a
be ween-subjec s design. A e he expe imen , analyses o
a iance can p o ide an ini ial de e mina ion o he ela i e
impac o each independen a iable on he dependen a i-
able (Roy, 1990). In ou example, we de e mine he ela i e
impo ance o each o he 11 package design a iables o
inc easing pu chase in en ions. Resea che s migh also pe -
o m Tukey’s hones ly signi ican di e ence pos hoc es o
mul iple compa isons. Resea che s can apply o he analy-
sis me hods, depending on hei concep ual models (e.g.,
s uc u al equa ion modeling o es pa allel media ion o
second-s age mode a ion; Bleie e al., 2019).
S ep 3: Follow upwi ha e i ica ion s udy
Taguchi-designed expe imen s p o ide comp ehensi e,
a iable-le el examina ions o a phenomenon bu should
be e i ied wi h a single- ac o con igu a ion-le el s udy
(Roy, 1990). Findings om he Taguchi-designed s udy
will iden i y which independen a iables signi ican ly a ec
he dependen a iable and he di ec ion o hese e ec s.
Resea che s should hen alida e hese esul s by compa ing
a s imulus con igu a ion con aining one o all he iden i ied
in luen ial a iables a hei op imal le els (i.e., op imal con-
di ion) agains a baseline con igu a ion wi h di e en le els
(i.e., subop imal condi ion). Resul s ha yield simila and
p edic ed changes in he dependen a iable co obo a e he
conclusions o he Taguchi s udy.
Fo example, conside ha ou illus a i e Taguchi-
designed s udy e ealed ha a ci cle logo shape (Le el 2),
small package size (Le el 1), o ganic o igin claim (Le el
1), and QR code (Le el 2) a e he signi ican a iables and
manipula ion le els ha p oduce he g ea es pu chase in en-
ions. The op imal con igu a ion condi ion would ea u e a
s imulus wi h hese a iables and le els. The compa ison
al e na i e would be any subop imal con igu a ion, which
esea che s can selec acco ding o hei goals and con ex .
Resea che s could use a s imulus ha mimics cu en p ac-
ices o a s imulus wi h he same ou signi ican a iables
bu a hei sub-op imum le els: a ec angle logo shape
(Le el 1), la ge package size (Le el 2), na u al o igin claim
(Le el 2), and no QR code (Le el 1). Non-signi ican a i-
ables om he Taguchi s udy could be held cons an . This
simple design alida es he Taguchi s udy indings and p o-
ides addi ional con idence ha he esul s a e no a ec ed
by unan icipa ed ela ionships among he independen a i-
ables. Resea che s can also use addi ional s udies o u he
unde s and he phenomenon (e.g., explo e media o s and
mode a o s, ex end gene alizabili y o o he si ua ions and
con ex s).
Ex ension o  heTaguchi app oach
oexpe imen al design: Signal‑ o‑noise
a ios
Due o i s o igins in enginee ing and emphasis on i m-con-
ollable ac o s, Taguchi-designed expe imen s can assess
and minimize sensi i i y o a ia ions in noise. Noise ac-
o s a e di icul o expensi e o con ol in p ac ice bu can
be expe imen ally manipula ed (e.g., shel placemen , cus-
ome s’ pe cei ed ime p essu e while shopping) and a e
expec ed o in luence he dependen a iable o in e es .
In he social sciences, many ac o s can be con olled, and
be ween-subjec s designs wi h pa icipan andomiza ion a e
o en used; since a ia ion and noise should be andomly dis-
ibu ed ac oss condi ions, including hem may no be nec-
essa y. S ill, some esea che s may include noise ac o s in
hei expe imen al designs o ensu e he indings a e obus
ac oss ac o s beyond hei esea ch scope o o add ess
in e ac ion e ec s be ween he ocal independen a iables
and noise ac o s. Resea che s hen es ima e signal- o-noise
a ios, which indica e he ela i e e ec s o independen a -
iables ac oss a ying le els o noise. S onge e ec s in he
p esence o noise sugges mo e obus e ec s. Fo example,
esea che s migh aim o design he mos e ec i e p oduc
packaging o s imula e pu chase in en ions ega dless o
e ail shel placemen . By conduc ing a Taguchi-designed
s udy in he lab o he ield mul iple imes ac oss di e en
shel loca ions, obus e ec s can be iden i ied.
Conclusion andimplica ions
In conside ing he ole o science in e e yday li e, i is
impo an o ecognize he colloquial use o he e idence
p esen ed in esea ch. Such e idence is equen ly cons ued
as objec i e ac s and e lec ions o eali y ha demand
a en ion, wa an conside a ion, and ul ima ely ca alyze
change. Ye , he acuum ha science o en c ea es ou o
necessi y in he pu sui o causal e idence s ands in s a k
con as o he complex, mul i ace ed eali ies in luencing
any gi en ou come. This disc epancy can lead o e o s
when applying indings. The Taguchi app oach is a powe -
ul expe imen al ool ha p agma ically emula es eal-wo ld
complexi y in i s b oad expe imen al designs. By in eg a -
ing di e se, heo e ically- ele an independen a iables—
whe he p e iously iden i ied in li e a u e o p ac ice, o
no el a iables ye o be collec ed o s udied— he Taguchi
app oach no only has he po en ial o consolida e exis ing
pe spec i es bu also o in oduce new ones. I s bene i s,
954 Jou nal o he Academy o Ma ke ing Science (2025) 53:949–954
howe e , go beyond p agma ism and consolida ion. Na -
ow scopes o in es iga ion o en accompany a single o
limi ed se o hypo heses, which can lead o esea che
a achmen , an endowmen e ec , o con i ma ion bias—all
po en ial h ea s o he ield. In de ense o igo ous scien i ic
me hods, Pla (1964, p. 350) o e s a clea an ido e: he
“me hod o mul iple hypo heses …[which] di e s om he
simple wo king hypo hesis in ha i dis ibu es he e o
and di ides he a ec ions.” The Taguchi app oach suppo s
his by encou aging he iendly co-exis ence o mul iple
compe ing hypo heses and p o iding a easible means o es
hem simul aneously. I he eby se es as a powe ul ool o
acili a ing scien i ic disco e y and ad ancemen .
Acknowledgemen s We wan o acknowledge and hank D . Na alie
Chisam o he iendly e iews and eedback.
Au ho s' con ibu ions All au ho s con ibu ed o he concep ualiza ion.
Funding No applicable.
Da a a ailabili y No applicable.
Decla a ions
E hical app o al No applicable.
Compe ing in e es No applicable.
Open Access This a icle is licensed unde a C ea i e Commons A i-
bu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a-
ion, dis ibu ion and ep oduc ion in any medium o o ma , as long
as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce,
p o ide a link o he C ea i e Commons licence, and indica e i changes
we e made. The images o o he hi d pa y ma e ial in his a icle a e
included in he a icle’s C ea i e Commons licence, unless indica ed
o he wise in a c edi line o he ma e ial. I ma e ial is no included in
he a icle’s C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will
need o ob ain pe mission di ec ly om he copy igh holde . To iew a
copy o his licence, isi h p:// c ea i eco mmons. o g/ licen ses/ by/4. 0/.
Re e ences
Bleie , A., Ha meling, C. M., & Palma ie , R. W. (2019). C ea ing
e ec i e online cus ome expe iences. Jou nal o Ma ke ing,
83(2), 98–119.
Kinde mann, B., Wen zel, D., An ons, D., & Salge, T. O. (2024). Con-
cep ual con ibu ions in ma ke ing schola ship: Pa e ns, mecha-
nisms, and ebalancing op ions. Jou nal o Ma ke ing, 88(3), 29–49.
Pla , J. R. (1964). S ong in e ence. Science, 146(3642), 347–353.
Roy, R. (1990). A p ime on he Taguchi me hod. Socie y o Manu ac-
u ing Enginee s.
Taguchi, G. (1986). In oduc ion o quali y enginee ing: design-
ing quali y in o p oduc s and p ocesses. Asian P oduc i i y
O ganiza ion.
Publishe 's No e Sp inge Na u e emains neu al wi h ega d o
ju isdic ional claims in published maps and ins i u ional a ilia ions.