Kristal R. Ray, Utah State University
Paul W. Fombelle, Northeastern University
Sterling A. Bone, Utah State University
Michael K. Brady, Florida State University
Scott A. Thompson, University of Georgia
Cliffs of Dissatisfaction: The Effect of Introducing Technology- Based Innovations on Service Employees & Customers
1. Kristal R. Ray, Utah State University
Paul W. Fombelle, Northeastern University
Sterling A. Bone, Utah State University
Michael K. Brady, Florida State University
Scott A.Thompson, University of Georgia
Cliffs of
Dissatisfaction:
The Effect of
Introducing
Technology-
Based
Innovations on
Service
Employees &
Customers
1
2. DRIVING INNOVATION: A TOP PRIORITY
„ With
a
market
capitaliza/on
of
$350
billion,
Google
a;ributes
its
success
to
con/nuous
innova/on.
„ Apple’s
tremendous
run
up
in
value
from
$2
billion
in
1997
to
more
than
$600
billion
in
2012
was
fueled
almost
en/rely
by
innova/on
(iPod,
iPhone,
iPad).
„ Amazon
Prime,
the
idea
of
a
soMware
engineer,
is
responsible
for
20%
of
Amazon’s
revenue.
2
4. „ Exis/ng
innova/on
research
focuses
on
features
of
the
product
or
service
itself
„ The
core
product
(Dahan
and
Srinivasan
2000)
and
product
a;ributes
(Green
and
Srinivasan
1990)
„ Product
variants
and
shared
variants
(Ho
and
Tang
1998;
Gupta
and
Krishnan
1999)
„ Supply
chain
(Clark
1989;
Dyer
1997)
and
product
design
(Finger
and
Dixon
1989)
„ Product
tes/ng
and
launch
(Hendricks
and
Singhal
1997;
Mahajan
and
Wind
1988)
„ No
research
looks
at
who
is
responsible
for
launching
the
innova6on
or
how
their
reac6ons
to
it
impact
customers
PRODUCT AND SERVICE INNOVATION
4
5. „ Customer-‐employee
interac/on
is
required
in
the
delivery
of
the
service
(e.g.,
Crosby,
Evans,
and
Cowles
1990;
File
and
Prince
1993)
„ Both
the
service
employee
and
the
customer
contribute
to
the
emo/onal
aspects
of
the
service
encounter.
Employee
a_tudes
may
translate
to
customer
sa/sfac/on
(Bailey, Gremler, McCollough 2001)
„ Understanding
the
employee
linkages
of
the
Service
Profit
Chain
are
cri/cal
to
understanding
the
spillover
on
customer
experience
and
firm
outcomes
(Homburg,
Wieseke, and Hoyer 2009; Heskett, Sasser, and Schlesinger 2003 )
FRONTLINE SERVICE EMPLOYEES
5
6. „ Innova/on
implementa/on
is
“the
cri/cal
gateway
between
the
decision
to
adopt
the
innova/on
and
the
rou/ne
use
of
the
innova/on”
(Klein
and
Sora
1996,
p.
1057)
„ The
failure
of
an
innova/on
is
not
the
ineffec/veness
of
the
innova/on
but
the
ineffec/veness
of
the
implementa/on
process
(Klein
and
Sora
1996)
„ It
is
es/mated
that
50%
or
more
of
a;empts
to
implement
major
technological
innova/ons
end
in
failure
(Aiman-‐Smith
and
Green
2003,
Baer
and
Frese
2003;
Repenning
and
Sterman
2002)
„ Of
the
$2.7
trillion
that
companies
invest
in
technology
each
year,
more
than
$500
billion
is
wasted
due
in
large
part
to
implementa/on
failure
(Klein
and
Knight
2005)
INNOVATION IMPLEMENTATION
6
8. „ With origins in the computer industry, the beta test is
the first stage of consumer product testing which
follows in-house usage testing, called the alpha test
(Pitta and Franzak 1996)
„ Beta testing is the leading approach to gathering user
input in NPD (Barczak, Griffin, and Kahn, 2009)
„ Most research focuses on getting the customer
involved in beta testing and innovation.
„ Research to further understand the user–producer
innovation dynamic is missing (Bogers,Afuah, and Bastian, 2010)
BETA TESTING
8
10. RESEARCH QUESTIONS
„ Innovation Timing: Is it more effective to rollout
innovations with employees prior to or
simultaneously with the customer implementation?
„ Spillover Effect: Does the rollout strategy of one
innovation influence the firm’s other innovations?
„ How does the (dis)satisfaction of an innovation
influence employee recommendation intentions of a
different innovation?
10
13. DATA AND ANALYSIS
Search Technology Dataset:
o 4 wave panel (T1,T2,T3,T4)
• Wave 1 – pre-employee rollout
• Wave 2 – employee rollout without customer
• Waves 3 & 4 – both employee & customer
o 2,586 employees
o Dependent variables: satisfaction with search product and
likelihood to recommend search product
o Fixed effects model with robust standard errors
13
14. DATA AND ANALYSIS
Community Platform Dataset:
o 2 wave panel (T1,T2)
o 1,268 employees
o Dependent variables: satisfaction with community product
and likelihood to recommend community product
o Fixed effects model with robust standard errors
14
15. March
2013
April
2013
May
2013
June
2013
July
2013
Aug.
2013
DATA COLLECTION TIMELINE
Employee Search
Implementation
Employee: Pre
Implementation
Survey-
Employee: Post
Implementation
Survey
Sept.
2013
Oct.
2013
Nov.
2013
Dec.
2013
Jan.
2014
Feb.
2014
Customer
Search
Implementation
Customer: Pre
Implementation
Survey
Customer: Pre
Implementation
Survey
Customer:
Pre-
Implementation
Survey
Customer:
Post Search
/ Pre Community
Implementation
Survey
Employee: Post-Search /
Pre Community Survey
Customer:
Post Community
Implementation
Survey
Community
Implementation
Employee:
Post
Community
Survey
15
20. SEARCH RESULTS: SUMMARY
o Employee release occurred prior to customer release
o Satisfaction drops after employee implementation
o Satisfaction only recovers after customer introduction
o Recommendation driven by search satisfaction
o QUESTIONS REMAINING:
o Is the employee cliff a result of a learning curve?
o What will be the effect of deploying more radical innovation?
20
21. STUDY 2: LEARNING EFFECT?
„ Alternative hypothesis: Employee cliff is a result of the need to
learn the new innovation and not a result of the timing of customer
onboarding. If so, the data pattern would look like:
5.00
5.50
6.00
6.50
7.00
7.50
Wave 1 Wave 2 Wave 3 Wave 4
Search Satisfaction
Recommend Search
Community Satisfaction
Recommend Community
21
22. STUDY 2: RESULTS
6.07
5.92
6.41 6.39
5.00
5.50
6.00
6.50
7.00
7.50
Wave 1 Wave 2 Wave 3 Wave 4
Search Satisfaction
Recommend Search
Community Satisfaction
Recommend Community
„ The cliff is not attributed to a learning
effect but is a result of the timing of
customer onboarding
22
23. SPILLOVER EFFECT
„ Prior research has only investigated the effect of one innovation
strategy.
„ QUESTIONS:
„ Does the implementation strategy of one innovation influence the firm’s other
innovation implementations?
„ How does the (dis)satisfaction of an innovation influence employee recommendation
intentions of a later innovation?
23
24. SPILLOVER EFFECT
1
2
3
4
5
6
7
8
9
10
10 9 8 7 6 5 4 3 2 1
RecommendCommunity
Search Satisfaction
The Effect of Search Satisfaction
(Innovation 1) on Employee
Recommendations of Community
(Innovation 2)
24
25. TESTING FOR SPILLOVER:
COMMUNITY RESULTS: RECOMMEND
Robust
Recommend | Coef. Std. Err. t Sig
-------------+----------------------------------------------------------------
Wave 2 dummy | .063 .081 0.79 0.431
Community satisfaction | .640 .077 8.27 0.000
Generally recommend Firm | .202 .058 3.51 0.000
Tired of changes | -.0623 .034 -1.85 0.065
Search Product Satisfaction | .116 .062 1.88 0.061
Community expertise | .115 .040 2.93 0.003
Importance of ease of use | .006 .004 1.64 0.101
Rated ease of use | .005 .003 1.92 0.055
Importance of features | -.015 .008 -1.94 0.053
_cons | -.227 .709 -0.32 0.748
-------------+----------------------------------------------------------------
R-sq: 0.6718
25
26. ESTIMATING SPILLOVER EFFECT ACROSS 2
SEPARATE INNOVATIONS
5.2
5.4
5.6
5.8
6
6.2
6.4
6.6
6.8
1 2 3 4
PredictedValues
Wave
The Effect of Search Satisfaction (actual) on
Recommend Community (predicted)
Search Satisfaction
Recommend Community
26
27. COMMUNITY RESULTS: SUMMARY
o Employee release coincided with customer release
o No drop in satisfaction or recommendation after employee release
o Covariates parallel those for the Search product
o Satisfaction with the Search product influences likelihood to recommend
the Community product
o Higher satisfaction with the Search product leads to a higher likelihood to
recommend the Community product
o Conversely, lower satisfaction with the Search product leads to a lower
likelihood to recommend the Community product
o Reveals an implementation spillover effect: : implementation strategy for one
product influences the implementation of another through satisfaction
27
29. MANAGERIAL IMPLICATIONS
„ Timing
of
an
innova/on
rollout
is
cri/cal
to
employees
as
well
as
customers.
„ Ge_ng
an
innova/on
to
employees
earlier
is
not
always
be;er.
„ Gives
them
/me
to
think
about
all
the
ways
a
new
innova/on
could
go
wrong.
„ We
seem
to
have
a
“much
ado
about
nothing
effect.”
29
30. MANAGERIAL IMPLICATIONS
„ We
know
a
lot
about
how
frontline
employees
impact
customers
but
not
the
other
way
around.
„ Our
findings
show
that
interac/ons
with
customers
seem
to
have
a
calming
effect
on
frontline
employees
who
otherwise
may
worry
about
how
an
innova/on
may
go
wrong.
30
31. MANAGERIAL IMPLICATIONS
„ We
know
about
how
an
innova/on
launch
in
one
company
impacts
innova/ons
from
other
companies.
„ We
know
li;le
about
how
an
innova/on
launch
impacts
other
innova/on
launches
in
the
same
company.
„ This
can
only
be
seen
if
one
looks
at
mul/ple
launches
in
one
company.
„ Results
show
that
frontline
service
employee
sa/sfac/on
with
the
launch
of
one
innova/on
impacts
later
innova/ons.
31
32. MANAGERIAL IMPLICATIONS
„ On
a
broader
level,
this
research
supports
efforts
to
understand
frontline
employees
in
addi/on
to
customers.
„ We
need
to
hear
the
voice
of
the
customer
and
the
voice
of
the
employee,
and
consider
how
and
when
one
impacts
the
other.
32
34. FUTURE RESEARCH
„ We
need
to
know
more
about
why
the
employee
shiM
occurs.
„ It
doesn’t
seem
like
customer
a_tudes
„ Seems
to
be
a
lack
of
complaints
„ We
need
to
know
more
about
the
launch
/ming
“sweet
spot”
„ What
is
the
op/mal
/ming
for
launching
an
innova/on
on
employees?
34
35. FUTURE RESEARCH
„ Dig
further
into
the
dyadic
rela/onship
between
individual
employees
and
their
customers.
„ Service
request
data
are
available
„ Analyses
are
underway
35