Agile Enterprise Architecture Management
An Analysis on the Application of Agile Principles
Matheus Hauder
1
, Sascha Roth
1
, Christopher Schulz
2
and Florian Matthes
1
1
Software Engineering for Business Information Systems (SEBIS), Technical University Munich,
Boltzmannstraße 3, 85748 Garching, Germay
2
Syracom, Parkring 4, 85748 Garching, Germany
{matheus.hauder, roth, matthes}@tum.de, christopher.schulz@syracom.de
Keywords: Enterprise Architecture Management, Survey, Agile, Enterprise Architecture Framework.
Abstract: Enterprise Architecture (EA) management has proven to be an efficient instrument to align business and IT
from a holistic perspective. Many organizations have established a permanent EA management function
responsible for modeling, analyzing, and defining the current and future EA state as well as the roadmap.
Similar as in software development, EA management initiatives face challenges that delay results,
complicate the collaboration, and deteriorate the overall work quality. While in software development, agile
principles and values reflected in tangible methods like Scrum and Extreme Programming are increasingly
adopted by organizations, there is little known whether these practices have already made their way into EA
management. Based on three research questions, this paper sheds light on the status-quo of agile principles
applied to EA management. We present results of an online survey among 105 industry experts working for
more than 10 industry sectors across 22 different countries.
1 INTRODUCTION
Globalization, frequently altering market conditions,
and the pressing need to reduce operating costs force
organizations to carry out complex business
transformations at a regular interval. However,
performed without a holistic and explicit picture of
the organization, these transformations are likely to
fail (Ross et al., 2006). An Enterprise Architecture
(EA) serves as a common means to look at an entire
organization as a whole. It captures both, business
aspects (e.g., business processes, business objects)
and IT aspects (e.g., interfaces, networks, devices)
as well as their interrelations (Buschle et al., 2012).
Being applied by an increasing number of
enterprises, the corresponding discipline EA
management fosters the mutual alignment of
business and IT (Weill and Ross, 2009).
EA management deals with capturing, modeling,
analyzing, and defining the current, planned, and
future architecture in conjunction with the roadmap
leading from the as-is to the target state (The Open
Group, 2011). However, EA management faces
various challenges ranging for instance from the late
return on investment to the delayed valuation of the
disciplines by concerned stakeholders (cf. e.g.
(Hauder et al., 2013), (Lucke et al., 2010) and
(Lucke et al., 2012).
When looking on the domain of software
development, researchers likewise to practitioners
propose the adherence to so-called agile values
helping to address these types of challenges
(Schwaber, 2004). Key to these values are agile
principles like the avoidance of waste (Gloger,
2010), an early stakeholder involvement (Beck et al.,
2001), and gathering feedback at an ongoing basis
(Highsmith and Cockburn, 2001).
In many cases these principles are based on lean
production practices initially applied by the Japanese
car manufacturer Toyota (Deming, 2000), (Holweg,
2007). As of today, the benefits of agile principles to
software development are still discussed
controversially (Reifer et al., 2003).
Several similarities between software
development - centering rather on single systems -
and EA management - focusing on the holistic
management of systems of systems - can be drawn.
Both disciplines have to handle frequently changing
requirements while ensuring a close collaboration
among the multi-disciplinary stakeholders. Focusing
on the latter, researchers have already proposed to
apply agile practices known from the development
38
Hauder M., Roth S., Schulz C. and Matthes F.
Agile Enterprise Architecture ManagementAn Analysis on the Application of Agile Principles.
DOI: 10.5220/0005424100380046
In Proceedings of the Fourth International Symposium on Business Modeling and Software Design (BMSD 2014), pages 38-46
ISBN: 978-989-758-032-1
Copyright
c
2014 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
of softw
a
Give
n
initially
Matthes
aware
o
Progra
m
(FDD),
a
their d
a
standpoi
endeavo
r
apply th
e
agile p
r
today’s
Open G
r
Yet,
EA ma
n
ground
conclud
e
‘Whi
engi
n
desi
g
man
a
The res
illustrat
e
the rese
review
domains
online
s
manage
m
question
applicat
i
with the
organiz
a
The
findings
literatur
e
explain
softwar
e
EA ma
n
our rese
a
setup of
experts
i
in Secti
o
remarks
2 A
G
E
N
To iden
t
values i
n
approac
h
(Webste
r
a
re (Ambler,
2
n
that in m
a
promoted
and Schulz,
o
f agile pra
m
ming (XP),
a
nd might ap
p
a
y-to-day b
u
nt, we wi
t
r
s of our i
n
e
se agile pra
c
r
escriptions
a
EA manage
m
oup, 2011).
literature do
c
n
agement is
s
is
m
issing.
A
e
to the follo
w
ch agile prin
c
n
eering dom
a
g
n of an o
r
a
gement prac
t
earch ap
p
ro
a
e
d in Figure
1
arch questio
n
to identify
. Based on t
h
survey to
e
m
ent. In or
d
n
s and gain
a
i
on of the p
r
specific EA
a
tions (Haude
r
following se
c
we made w
h
e
looking for
how agile p
r
e
developme
n
n
agement. In
a
a
rch hypothe
s
f
an online su
r
i
n the field.
W
o
n 5 and 6 be
on futu
r
e res
e
G
ILE PRI
N
TERPRI
t
ify tangible
m
n
EA manag
e
h
as recomm
e
r
and Watso
n
2
010a), (Buc
k
a
ny cases E
A
through IT
2013), EA i
n
ctices, e.g.
Feature Dri
v
p
ly their acco
r
u
siness. Fro
m
t
ness that
E
n
dustry part
n
c
tices. In addi
a
re implicitl
y
m
ent frame
w
c
umenting th
s
carce; in pa
r
A
gainst this
w
ing research
o
c
iples known
f
a
in
s
hould b
e
r
ganization-
sp
t
ice?’
a
ch and the
1
: After defin
i
n
s, we cond
u
agile princi
p
h
ese principle
e
valuate thei
r
d
er to ans
w
a
deeper und
e
r
inciples, we
management
r
et al., 2013).
c
tion (Sectio
n
h
en perusing
agile pointer
s
r
inciples and
n
t world cou
l
a
ddition to t
h
s
es, Section 4
r
vey we con
d
W
e analyze an
fore concludi
n
e
arch.
NCIPLE
S
SE ARC
H
m
aterial on a
g
e
ment, we ap
p
e
nded by We
b
n
, 2002). Duri
n
k
l et al., 2011
)
A
manageme
n
(Hauder,
R
n
itiatives are
Scrum, Ext
r
v
en Develop
m
r
ding principl
e
m
an empi
r
E
A manage
m
n
ers increasi
n
t
ion, we diag
n
incorporate
d
w
orks, e.g.,
(
e agile natur
r
ticular empi
r
background,
o
bjective:
f
rom the soft
w
e
applied fo
r
p
ecific agile
deliverables
i
ng the scope
u
cted a liter
a
p
les from
o
s we designe
d
r
usage in
w
er the rese
e
rstanding o
n
correlated
t
challenges o
f
n
2) provide
s
EA manage
m
s
. In Section
3
values fro
m
d be adopte
d
h
e introductio
outlines the
m
d
ucted among
d
discuss the
n
g the paper
w
S
IN
H
ITECTU
R
g
ile principles
p
lied a struct
u
b
ster and W
a
n
g April 201
3
)
.
n
t is
R
oth,
well
r
eme
m
ent
e
s in
r
ical
m
ent
n
gly
n
ose
d
in
(
The
r
e of
r
ical
we
w
are
r
the
EA
are
and
a
ture
o
ther
d
an
EA
e
arch
n
the
t
hem
f
the
s
the
m
ent
3
we
m
the
d
by
o
n of
m
ain
105
data
with
R
E
and
u
red
a
tson
3
we
per
u
p
ro
Go
o
an
d
we
fol
l
ma
n
tra
n
so
u
me
t
fol
l
giv
e
b
e
b
foc
u
20
1
ma
n
p
ra
c
the
agi
l
an
20
0
fur
t
co
m
20
1
p
ra
c
qu
a
co
u
Or
a
fro
m
u
sed diffe
r
ceedings, an
d
o
gle Scholar,
d
the library
o
carried out
e
l
owing Englis
n
agement’ a
n
n
slations. Aft
e
u
rces (title,
a
t
hod of her
m
l
owing sourc
e
e
n their focus
Figure 1: R
e
Ambler acce
n
b
usiness driv
u
sed on pro
d
1
0a). Based
o
n
agement i
s
c
titioner poi
n
managemen
t
l
e, among ot
h
iterative an
d
0
9). In the lat
e
t
her, proposi
n
m
plemented
b
1
0b). The fin
d
c
tical work
e
a
ntitative scal
e
Bob Rhubart
u
ld be turne
d
a
cle represent
a
m
architects,
r
ent IS j
d
books usin
g
IEEExplore,
C
o
f our researc
e
lectronic ful
h
keywords:
n
d ’agile’ as
e
r a first ana
l
a
bstract, outl
m
eneutic tex
t
e
s have been
on the topic.
e
search approa
c
n
tuates that
E
e
n, evolution
a
d
ucing valua
b
o
n an examin
a
s
typically
t
s out six pie
t
of enterpri
s
ers, simplicit
y
d
increment
a
e
st of his rep
o
n
g an agile
b
y several ke
y
ings publishe
d
e
xperience wi
t
e
.
describes ho
w
more agile
a
tive highligh
t
developers,
a
j
ournals, c
o
g
the Web o
f
Citeseer, Spr
i
c
h institution.
l
l-text search
e
enterprise ar
c
well as thei
r
l
ysis of the
o
l
ine) adherin
g
t
comprehen
identified a
s
c
h and delivera
b
E
A managem
e
ary, collabor
a
b
le artifacts
n
ation of pro
b
coping
w
e
ces of advic
e
se architectu
y
, focus on p
e
a
l approach
o
rts, Ambler
g
architectur
e
y
techniques
d by Ambler
a
i
th no evalua
t
w an EA ma
n
(Rhubart, 2
0
h
ts the necess
a
a
nd other sta
k
o
nference
f
Science,
i
ngerLink
Thereby,
e
s on the
c
hitecture
r
German
o
verall 53
g
to the
sion, the
s
relevant
b
les.
e
nt has to
a
tive, and
(Ambler,
b
lems EA
w
ith, the
e
to make
r
es more
e
ople, and
(Ambler,
g
oes even
e
process
(Ambler,
a
ll root in
t
ion on a
n
agement
0
10). The
a
ry buy-in
k
eholders
Agile Enterprise Architecture Management - An Analysis on the Application of Agile Principles
39
at all levels of the organization. Next to the
importance of conversation in particular with the
developing teams, the manager considers the
involvement of enterprise architects at the project
level as very crucial. Again, all suggestions are
based on in-the-field work based on a single
company (employee) perspective.
Friedrichsen and Schrewe see typical EA
management problems (e.g. losing sight of
fundamentals, becoming a slave of the EA
management framework) as a reason to introduce
agile values (Friedrichsen and Schrewe, 2010). The
consultants advise to launch an EA management
initiative with clear goals and a limited scope while
always keeping potential risks in mind. In their eyes,
frameworks and tools have to be considered as
toolboxes that ensure to reach the stated goals more
efficiently.
While Eric Landes recommends applying
concrete techniques like retrospectives and lessons
learned action items, iterative cycles, as well as
automated acceptance criteria in the emergent design
of an architecture (Landes, 2012), Scott Nelson
assumes two distinct viewpoints when discussing the
similarities and differences of managing enterprises
architectures vs. developing software in an agile
manner (Nelson, 2012).
As another industry expert and active blogger,
Gabhart advises to avoid big bang EA management
projects attempting to “boil the ocean”, thus are too
big in scope (Gabhart, 2013). Instead of that, the
author proposes to start off small, building up an EA
management capability in an incremental and
iterative 4-step process. Lastly, the staff member
Gattadahalli of the former IT Company EDS shares
the knowledge of an agile management of EAs in
terms of seven critical success factors (Gattadahalli,
2004).
After having introduced EA management to the
reader of their book, Bente, Bombach, and Langade
proposes six so-called building blocks helping to
render the discipline more agile and lean (Bente et
al., 2012). Benefiting from examined sources paired
with their professional experience, Bente et al.,
describe how to streamline the architecture
processes, setup an agile EA project, and foster
collaboration and participation. Even though their
explanations are backed by several fictitious
examples, no quantitative results are provided that
would prove the adoption of agile practices in EA
management.
To respond to the problems often encountered in
EA management, Shirazi et al. propose a framework
rendering the discipline more agile (Shirazi and
Rouhani, 2009). Named Agile Enterprise
Architecture Framework (AEAF), the artifact
consists of seven models and eleven interactions
both based on agile principles and values. Even if
the authors do not indicate any relations, the five
viewpoints and six project aspects also included in
AEAF resemble the Zachman framework (Zachman,
1987).
Although AEAF touches on several agile aspects
like regular feedback or focus on cooperation, the
research group’s paper neither proves the empirical
relevance of an agile EA management nor it validate
the framework work in practice.
Rooted in lean principles, information
technology architectures, and systems engineering
methods, Comm and Mathaisel propose the Lean
Enterprise Architecture (LEA), a three-phase
structure to organize the activities for the
transformation of the enterprise to agility (Comm
and Mathaisel, 2010). The researchers combine their
framework with concepts from the Lean Enterprise
Transformation Engineering while also
incorporating lean principles and practices in the
resulting process. However, their work does not
detail on these principles or explains how an agile
enterprise should evolve its EA.
As one of the most popular approaches, The
Open Group Architecture Framework (TOGAF) 9.1
does not explicitly recommend to manage an EA in
an agile style (The Open Group, 2011). In turn, a
more agile organization is considered as a surplus
brought along by a “good” enterprise architecture.
Notwithstanding, with concepts like iterations to
develop a comprehensive architecture landscape and
architecture, to manage changes to the
organization’s architecture capability, as well as
appropriate stakeholder management the EA
framework TOGAF promotes important agile
principles.
The striving for agile principles and values
enhancing the efficiency of EA management is
mainly found in practitioners’ circles. While only a
small number of experts emphasize the misfit of
both disciplines, e.g. (Nicholette, 2007), the majority
of industry authors consider agile means as being
well suited for EA management (Banerjee, 2011).
As of January 2014, few academic publications and
frameworks embrace or even mention to apply an
agile management means for EAs. Studied sources
are very new, indicating that the mind-set of an agile
EA management is still nascent. No contribution
was found that investigated on the current status quo
of agile practices in industry.
Fourth International Symposium on Business Modeling and Software Design
40
Figure 2:
current E
A
3 A
P
D
E
T
O
A
R
The ap
p
p
rincipl
e
p
art of
t
p
rincipl
e
evaluate
d
manage
m
the ap
p
manage
m
Whe
n
managi
n
early a
n
maintai
n
changin
g
2005).
T
enterpri
s
delivera
b
increme
n
Design of an
A
frameworks
w
P
PLYING
E
VELOP
M
O
ENTER
P
R
CHITE
C
p
lication of
e
s to EA is ill
u
t
he figure s
h
e
s that are
d
to design
m
ent function
p
lication of
m
ent.
n
focusing o
n
n
g EAs, the A
n
d constant
n
ing a respo
n
g
requiremen
t
T
ranslated int
o
s
e architects
b
les as ea
r
n
tal and iter
organization-s
p
w
hich are exte
n
AGILE S
M
ENT P
R
P
RISE
C
TURE
agile softw
u
strated in Fi
g
h
ows the fra
m
synthesized
an organiz
a
. In the follo
w
these pri
n
n
the workin
g
gile Manifest
delivery
o
n
sive attitude
t
s (Beck et
a
o
an EA ma
n
should stri
v
r
ly as poss
i
ative approa
c
p
ecific agile E
A
n
ded by agile p
r
OFT
W
A
R
R
INCIPL
E
a
re develop
m
g
ure 2. The l
o
m
ework and
a
and const
a
a
tion-specific
w
ing, we illus
t
n
ciples for
g
style applie
d
o
recommen
d
o
f results
w
with regar
d
a
l., 2001), (C
n
agement con
t
v
e to ship
t
i
ble, pursue
c
h, and em
b
A
managemen
t
rinciples, cf. (
R
R
E
E
S
m
ent
o
wer
a
gile
a
ntly
EA
trate
EA
d
for
d
s an
w
hile
d
s to
ohn,
n
text,
t
heir
an
b
race
ch
a
Si
m
EA
mo
s
ov
e
an
co
n
en
o
Sp
e
a
m
ex
p
20
1
20
1
the
onl
y
en
v
int
e
the
b
e
Ro
b
the
sh
o
t
practice base
d
R
oth et al., 201
4
a
nges regardi
n
m
ilar to their
s
management
s
t important
t
e
r completene
s
As goes the
A
EA manag
e
n
stant pace tr
y
o
ugh leeway
e
aking of fle
x
m
odus opera
n
p
eriment and
1
2). In confo
r
1
0) as well a
s
EA manage
m
y
upon st
a
v
ironment w
i
e
rference dur
i
other side o
f
eager to give
b
ertson, 200
6
EA manag
e
o
uld be incorp
o
d
on theoretica
l
4
).
n
g
t
heir wor
k
oftware deve
l
team should
a
t
asks first wi
t
s
s and quality
A
gile Manife
s
m
ent team
s
y
ing to avoid
for
r
eflectio
n
x
ibility, agile
n
di where m
e
try out ne
w
r
ming to the
p
s
the one-
p
ie
c
m
ent team sho
a
keholders’
i
th little or
n
g the work
f
the spectru
m
regular feed
b
6
) on the res
u
e
ment team.
o
rated into th
e
l
concepts ofte
n
k
ing style an
l
oping count
e
always take c
th a valuatio
n
y
(Stal, 2012).
s
to (Beck et
a
should adva
n
overtime wh
i
n
s and retro
s
literature rec
o
e
mbers are a
l
w
things (
C
p
ull-
p
rincipl
e
c
e flow (Fish
e
o
uld create de
l
demand w
i
no distrac
t
(Schwaber,
2
m
, stakehold
e
b
ack (Ross,
W
u
lts delivere
d
However,
t
e
work of the
n
found in
d
results.
e
rparts, an
are of the
n
of time
a
l., 2001),
n
ce in a
i
le having
s
pectives.
o
mmends
l
lowed to
C
oldewey,
e
(Gloger,
er
, 2000),
l
iverables
i
thin an
t
ion and
2
004). On
rs should
W
eill and
d
through
t
he latter
team.
Agile Enterprise Architecture Management - An Analysis on the Application of Agile Principles
41
In the sense of working software and simplicity
(Beck et al., 2001), (Highsmith and Cockburn,
2001), EA management results should be as usable,
simple, and accessible as possible for EA
management stakeholders. Benefiting from each
individual deliverable the EA management team
releases, stakeholders should be satisfied with the
outcome and value the EA management team
creates. As called for in agile literature (Highsmith,
2002), (Gloger, 2010), EA management results
should be of the highest quality, crafted in a way that
they only respond to the stakeholders’ demand with
a level of done that is understood and agreed upon.
Centering on the actors performing the work,
agile sources emphasize a cross-organizational team
whose members are specialized to perform various
tasks (Gloger, 2010) in a self-organized manner
(Beck et al., 2001). From an educational perspective
(Coldewey, 2012), the EA management team
members should have special skills and training in
multiple organizational areas (e.g., infrastructure,
processes, application) while being capable to
manage the sequence order their tasks are eventually
completed.
Both, high education and expertise permit the
team to speak the same language as stakeholders and
information providers on a daily basis. In line with
the fifth agile principles (Beck et al., 2001), the EA
management team leader has to create a positive
work environment while catering to the team’s self-
organization. Besides an intrinsic motivation (Beck
et al., 2001), and work satisfaction, each EA
management team member should have a notion of
his/her colleagues’ duties and results. Looking on
the overall organizational structure (Fisher, 2000),
EAM tasks should be accomplished through small
sub-teams in which roles and responsibilities are
clearly defined and understood. Finally, the team
requires strong diplomacy and negotiation skills
employed when interacting with stakeholders and
EA information providers.
4 RESEARCH QUESTIONS,
SURVEY DESIGN, AND
EMPRICAL BASIS
In above outlined literature the application of agile
principles for EA management has been widely
suggested by authors in the community. However, to
date neither a comprehensive list of practical applied
agile principles in EA management is published nor
an empirical validation thereof exists.
Since in many cases EA management is initially
promoted through IT (Hauder et al., 2013) which
adopts agile principles more or less eagerly, we
formulate the first research question as follows:
Research Question 1: What are frequently applied
agile principles for EA management in practice?
Our second research question aims at validating
observations, e.g. on the shift towards an
incremental and iterative work fashion for certain
EA management challenges. Not only this enhances
the scoping during the launch of EA initiatives,
incremental and iterative developed products might
provide stakeholders with early results and, thus,
lead to an increased buy-in.
Research Question 2: Which agile principles are
used in enterprises for certain EA management
challenges?
Typically EA management puts focus on a long term
plan how organization should evolve, while agile
practices promote the constant change of ongoing
projects. Since both approaches appear contradictory
at the first look, we formulate the third research
question as follows:
Research Question 3: What are challenges for the
design of an organization-specific agile EA
management practice?
To evaluate these three questions on an empirical
basis, we compiled an online questionnaire using 3-
point Likert scale questions. The contained questions
were based on the collection of agile principles we
explained above. To optimize the questionnaire’s
design, we conducted a pre-test with three
independent non-related researchers who were
requested to complete our survey.
Figure 3: Industry sector of organizations (n and %).
IT Consulting 25
26%
Finance 18 19%
Public Service 8
8%
Manufacturing 6
6%
Telecommunicati
ons 6 6%
Education 5 5%
Management
consulting 4 4%
Transportation 3
3%
Health 2 2%
Other 20 21%
Fourth International Symposium on Business Modeling and Software Design
42
The final version of the questionnaire was
available for 21 days. To receive relevant
information we targeted participants working in EA
management or related fields. Using e-mail, we sent
over 1100 survey invitations to industry experts we
collaborated with during the last 8 years.
Figure 4: Job titles of participants (n and %).
In addition, the survey has been published in the
two online forums Xing and LinkedIn, announcing
them as topics related to EA or strategic IT
management. We received input from 178 survey
participants, filtered duplicate answers, and ended
up with 105 completed answers for the evaluation,
i.e. a dropout quote of ~41%.
As the survey was conducted primarily in
Germany, 61 (~58%) participants are employed in
Europe. 18 (~17%) work in the USA and 26 (~25%)
are employed in other countries having less than 10
responding participants. Figure 3 illustrates the
distribution of the industry sectors of the
participating organizations. IT consulting is the
largest sector, whereas all consultancies were
requested to answer on behalf of one particular EA
management engagement. IT consulting is followed
by the Finance and Public sectors.
Figure 4 depicts the participants of the online
survey divided by job title. The largest groups
consist of Enterprise Architects followed by IT
Architects and Consultants. Among the participants
are also Business Architects and members of the
management board. In average, questioned
organizations have an experience of 5 years in EA
management.
5 AGILE PRINCIPLES FOR
ENTERPRISE ARCHITECTURE
MANAGEMENT IN PRACTICE
In the following three subsections the research
questions are evaluated based on our empirical data
set. The second research question is evaluated by
applying the Pearson’s chi-square test to validate the
dimensions in our data set.
5.1 Application of Agile Principles
The first research question deals with the application
of agile principles for EA management in practice.
Figure 5 illustrates the practical adoption of agile
principles in EA management ordered by frequency.
As depicted, organizations adhere to agile principles
with a different degree of intensity, confirming our
assumption that the applicability of agile principles
varies for EA management. For instance, while most
organizations perform retrospectives within their EA
management team, only few value time over quality.
Most EA management initiatives apply an iterative
(~79%) and incremental (~87%) approach. About
93% of the organizations apply EA management in a
self-organized manner. Moreover, ~75% say that
they act cross-functionally.
While the overwhelming majority of
organizations apply several agile principles for the
introduction and operation of their EA management
initiatives in practice, some principles are less
frequently traceable. In particular some of these less
frequent agile principles are related with the quality
and completeness of the developed EA products.
Only ~42% of the participating organizations
apply time over completeness and only ~25% rate
time over quality for the developed EA products.
Next to agile principles related to quality and
completeness of the developed EA products, actual
stakeholder demands and utilization of the produced
EA products by these stakeholders are only applied
by the minority of the organizations in our dataset.
With ~38% only a small number organizations are
truly concerned whether these EA products are
actually used by stakeholders.
Enterprise
Architect 54 51%
IT Architect 15
14%
Consultant 12
11%
Business
Architect 6 6%
CxO 6 6%
IT Operations 3
3%
Software
Engineer 1 1%
Other 8 8%
Agile Enterprise Architecture Management - An Analysis on the Application of Agile Principles
43
Figure 5: Applied agile principles for EA management in practice (n=105).
5.2 Agile Principles and Enterprise
Architecture Challenges
We answer the second research question by
correlating EA management challenges from our
empirical basis (cf. Hauder et al., 2013) with the
agile principles illustrated in Figure 5. Due to space
limitations, we only illustrate the statistical
correlations for three major EA management
challenges with agile principles using Pearson’s chi-
square test.
The challenge late valuation of EA management
through stakeholders appears in ~51% of the
participating organizations. According to our
statistical test these organizations apply the principle
adherence to one-piece flow with p = .047 (p .05).
In addition, the principle focus on requirements
resulted in a goodness of fit test of p = .00004 (p
.05). Further agile principles that correlate with this
Fourth International Symposium on Business Modeling and Software Design
44
challenge are advancement with an indefinite &
constant pace p = .002 (p .05), stakeholders
provide feedback to EA management team p = .0002
(p .05), agreed level of done p = .009 (p .05),
useable for stakeholders p = .042 (p .05), and as
simple and accessible as possible p = .005 (p .05).
All other agile principles were not statistically
dependent on this challenge for the given relevance.
Around ~38% of the organizations are struggling
with outdated EA results. This means that
architecture descriptions are often outdated before
they are complete and often understood as a project
rather than a continuous process. The agile
principles characterized by defined roles &
responsibilities correlates with p = .004 (p .05),
members knows their colleagues’ duties with p =
.0001 (p .05), focus on high quality p = .005 (p
.05), satisfied with its work p = .001 (p .05),
adherence to one-piece flow p = .00001 (p .05),
incorporation of reflections & retrospectives p =
.001 (p .05), agreed level of done p = .0001 (p
.05), and usable for stakeholders p = .001 (p .05).
Reluctant information providers are a challenge
for ~65% of the organizations. This is a very critical
problem since enterprise architects heavily rely on
the information and knowledge provided by
stakeholders. The agile principle satisfied with its
work correlates with p = .043 (p .05), focus on
requirements p = .00001 (p .05), application of the
pull principle p = .009 (p .05), embracement of
changes p = .030 (p .05), valuation of time over
quality p = .004 (p .05), as simple and accessible
as possible p = .00001 (p .05), and exactly respond
to stakeholders’ demand p = .003 (p .05) correlate
with this challenge.
5.3 Designing an Agile Enterprise
Architecture Management Practice
Designing an agile EA management practice is a
challenging issue. While EA management
frameworks typically work towards a long range
vision of the organization or a business case, agile
practices incorporate findings from ongoing projects
immediately in the process. To put it in another way,
both approaches appear contradictory due to their
top-down and planning (EA management)
respectively bottom-up and emergent course of
action.
Regarding the challenges EA management
initiatives in organizations are faced with neither of
these approaches can solve all challenges on his
own. Integrating both approaches within one agile
EA management practice that is tailored to the
specific demand of the organizational context would
be desirable. The findings presented in this paper
provide an initial empirical basis for further research
on an agile EA management practice. This
compromises the development of agile EA
management roles, activities, and deliverables.
6 CONCLUSIONS
In this paper we provided an empirical foundation
for agile principles applied to EA management by
today’s organizations. Due to the survey design, the
asked industry experts could only confirm or reject
the application of an agile principle for EA
management. Details about their actual
implementation are yet to be revealed. As of today,
this might be challenging, given the scarce literature
on agile EA management and only the implicit
adoption through EA frameworks. Regarding our
survey results, a potential bias might originate from
the lack of a common understanding on how to
operationalize agile principles in EA management.
Further research could examine the impact of
agile principles on the success and benefits of EA
management initiatives. Thereby, the efforts should
account for different organizational factors like the
size of the business, structure, EA management
experience, industry, and tool support. Further
studies could also focus on the correlation (and later
causalities) between challenges encountered in EA
management and possible mitigation through agile
principles.
REFERENCES
Ambler, S. W. (2009): Agile Enterprise Architecture.
http://www.agiledata.org/essays/enterpriseArchitecture
.html. Last opened: 27/08/2013.
Ambler, S. W. (2010a): Agile and Enterprise Architecture.
https://www.ibm.com/developerworks/mydeveloperw
orks/blogs/ambler/entry/agile_and_enterprise_architec
ture?lang=en. Last opened: 27/08/2013.
Ambler, S. W. (2010b): Agile Architecture: Strategies for
Scaling Agile Development. http://www.agile
modeling.com/essays/agileArchitecture.htm. Last
opened: 27/08/2013.
Banerjee, U. (2011): Agile development and Enterprise
Architecture practice - Can they coexist. Technology
Trend Analysis. http://setandbma.wordpress.com/
2011/04/11/agile-development-and-enterprise-
architecture-practice-can-they-coexist. Last opened:
27/08/2013.
Beck, K., Beedle, M., Bennekum, A. van, Cockburn, A.,
Agile Enterprise Architecture Management - An Analysis on the Application of Agile Principles
45
Cunningham, W., Fowler, M., Grenning, J., et al.
(2001): Manifesto for Agile Software Development.
Agile Alliance. http://agilemanifesto.org. Last opened:
27/08/2013.
Bente, S., Bombach, U., & Langade, S. (2012):
Collaborative Enterprise Architecture: Enriching EA
with Lean, Agile, and Enterprise 2.0 Practices. 1
st
ed.
Morgan Kaufmann, Burlington.
Buckl, S., Matthes, F., Monahov, I., Roth, S., Schulz, C.,
& Schweda, C. M. (2011): Towards an Agile Design
of the Enterprise Architecture Management Function.
6
th
International Workshop on Trends in Enterprise
Architecture Research (TEAR). Helsinki.
Buschle, M., Grunow, S., Matthes, F., Ekstedt, M.,
Hauder, M., & Roth, S. (2012): Automating enterprise
architecture documentation using an enterprise service
bus. In Proceedings of the 18
th
Americas Conference
on Information Systems. Washington.
Cohn M. (2005): Agile Estimating and Planning, Prentice
Hall PTR, Upper Saddle River.
Coldewey, J. 2012: Was heißt hier eigentlich “Agil”?
Kennzeichen agiler Organisationen. In
ObjektSpektrum 05/2012.
Comm, C. L., & Mathaisel, D. F. X. (2010): A Lean
Enterprise Architecture for Business Process Re-
engineering and Re-marketing. 12
th
International
Conference on Enterprise Information Systems (pp.
497–500). Madeira.
Deming, W. E. (2000): Out of the Crisis. MIT press,
Cambridge.
Fisher, K. (2000): Leading Self-Directing Work Teams.
McGraw-Hill, New York.
Friedrichsen, U., & Schrewe, I. (2010): Leichtgewichtige
Unternehmensarchitekturen – Wie Agilität bei der
Einführung eines EA Management helfen kann. In
OBJEKTspektrum, EAM/2010.
Gabhart, K. (2013): Generating Value through
Information Architecture. http://archvalue.com/agile-
enterprise-architecture (last opened: 27/08/2013)
Gattadahalli, S. (2004): Agile Enterprise Architecture
(AEA) - 7 Steps to Success. London.
Gloger, B. (2010): Scrum. In Informatik-Spektrum, 33(2):
195–200.
Hauder, M., Roth, S., Matthes, F., & Schulz, C. (2013):
Organizational factors influencing enterprise
architecture management challenges. 21
st
European
Conference on Information Systems (ECIS). Utrecht.
Highsmith J. (2002): Agile Software Development
Ecosystems. Pearson Education, Indianapolis.
Highsmith, J., & Cockburn, A. (2001): Agile software
development: the business of innovation. In Computer,
34(9): 120–127.
Holweg, Matthias (2007): The genealogy of lean
production. In Journal of Operations Management 25
(2): 420–437.
Landes, E. (2012): Agile Software Development Concepts
for Enterprise Architects. http://www.devx.com/
architect/Article/47842. Last opened: 27/08/2013.
Lucke, C., Krell, S., & Lechner, U. (2010): Critical Issues
in Enterprise Architecting - A Literature Review. In
AMCIS 2010 Proceedings (pp. 1-11). Association for
Information Systems.
Lucke, C., Bürger, M., Diefenbach, T., Freter, J., &
Lechner, U. (2012): Categories of Enterprise
Architecting Issues - An Empirical Investigation based
on Expert Interviews. In D. C. Mattfeld & S. Robra-
Bissantz (Eds.), Multikonferenz Wirtschaftsinformatik
(pp. 999-1010). GITO mbH Verlag, Berlin.
Nelson, S. (2012): Making Enterprise Architecture Work
in Agile Environments. http://www.devx.com/
architect/Article/47871. Last opened: 27/08/2013.
Nicholette, D. (2007): Enterprise Architecture and Agile.
Musings of a Software Development Manager.
http://edgibbs.com/2007/10/04/enterprise-architecture-
and-agile. Last opened: 27/08/2013.
Reifer, D. J., Maurer, F., & Erdogmus, H. (2003): Scaling
agile methods. Software, IEEE, 20(4): 12-14.
Roth, S., Zec, M. & Matthes, F. (2014): Enterprise
Architecture Visualization Tool Survey 2014.
Technical Report, Technische Universität München.
Ross, J. W., Weill, P., & Robertson, D. (2006): Enterprise
architecture as strategy: Creating a foundation for
business execution. Harvard Business Press, Boston.
Webster, J., & Watson, R. T. (2002): Analyzing the Past to
Prepare for the Future: Writing a Literature Review.
MIS Quarterly, 26(2): 13–23.
Weill, P., & Ross, J. W. (2009): IT Savvy: What top
executives must know to go from pain to gain. Harvard
Business Press, Boston.
Rhubart, B. (2010): Agile Enterprise Architecture. Oracle
Magazine. http://www.oracle.com/technetwork/issue-
archive/2010/10-nov/o60architect-175580.html. Last
opened: 27/08/2013.
Schwaber, K. (2004): Agile Project Management with
Scrum. 1
st
ed., Microsoft Press, Redmond,
Washington.
Shirazi, H. M., Rouhani, B. D., & Shirazi, M. M. (2009):
A Framework for Agile Enterprise Architecture. In
International Journal of Intelligent Information
Technology Application, 2(4): 182–186.
Stal, M. (2012): Softwarearchitektur und Agilität - Freund
oder Feind?. OOP 2012. München.
The Open Group (2011): TOGAF® Version 9.1, Van
Haren Publishing, Zaltbommel.
Zachman, J. A. (1987): A framework for information
systems architecture. In IBM systems journal, 26(3):
276-292.
Fourth International Symposium on Business Modeling and Software Design
46