Improving Computer-Support for

Collaborative Business Model Design and Exploration

Marin Zec

, Peter D

urr

, Alexander W. Schneider

and Florian Matthes

Fakult

at f

ur Informatik, Technische Universit

at M

unchen, Boltzmannstr. 3, 85748 Garching, Germany

Applied Social Sciences, Munich University of Applied Sciences, Am Stadtpark 20, 81243 M

unchen, Germany

{marin.zec, alexander.schneider, matthes}@tum.de, peter.duerr@hm.edu

Keywords:

Business Model Innovation, Business Model Canvas, Business Model Development, Business Model

Exploration, Morphological Analysis, Computer-Aided Business Model Design, Computer-Aided Business

Model Generation, Group Dynamics

Abstract:

Finding a viable and sustainable business model is a major challenge not only for startup companies. Estab-

lished companies are re-thinking their existing business models and explore new business opportunities. The

Business Model Canvas is currently one of the most popular frameworks for business model innovation. While

computer-aided design (CAD) tools are well-established in mechanical engineering, business model design is

still mostly done using pen-and-paper methods. In this paper, we (1) discuss beneﬁts and shortcomings of

the Business Model Canvas approach, (2) show how it can borrow techniques from General Morphological

Analysis to overcome shortcomings, and (3) derive three key requirements for future collaborative CAD tools

for business model design. Our analysis contributes to an understanding of how software support can improve

collaborative design and evaluation of business models.

1 INTRODUCTION

Business model innovation plays an increasingly im-

portant role for both startup as well as mature com-

panies due to increasing competition (Mitchell and

Coles, 2003; Mitchell and Coles, 2004). On one hand,

the primary organizational goal of startup companies

is to generate a viable business model – sometimes re-

sulting in disruption of whole markets. On the other

hand, established companies primarily aim to execute

their business model as efﬁciently as possible. As a

result, they tend to struggle with rapid and/or pro-

found market changes.

However, many established companies realize the

strategic importance of business model innovation for

the sustainability of their organization (e.g. Ama-

zon Web Services, a collection of cloud computing

services offered by Amazon.com in addition to their

primary e-commerce business). Consequently, an

increasing number of companies strive for continu-

ous business model innovation. They pursue vari-

ous strategies to do so, such as promoting and estab-

lishing intrapreneurship, corporate venturing or cre-

ation of corporate venture capital units. Examples in-

clude Google Ventures, Siemens Venture Capital and

Unilever Ventures. Still business model innovation re-

mains to be a complex problem for both startups and

more mature companies (Chesbrough, 2010).

Business model innovation is typically conducted

in teamwork since expertise in different domains such

as marketing, engineering and accounting is needed

for holistic business model design. However, each

expert tends to maintain his/her domain-speciﬁc men-

tal model and terminology. Boundary objects make

collaboration across different groups possible since

they provide a common point of reference for discus-

sion and collaboration. Individuals/groups with dif-

ferent background can interpret them differently with-

out surrendering the shared boundary object’s iden-

tity (Carlile, 2002). The Business Model Canvas

(BMC) (Osterwalder and Pigneur, 2010), an artifact

designed to facilitate business model design by pro-

viding a problem structure and focus of thought (Ep-

pler et al., 2011), can serve as boundary object for

business model innovation.

The BMC was at ﬁrst proposed as a pen-and-paper

or whiteboard tool for business model design work-

shops. However, Osterwalder et al. (2013) call for

CAD software for business modeling since they ex-

pect it to yield beneﬁts for strategic planning similar

to the beneﬁts and features CAD software brought to

engineering or architecture, e.g. “[. . . ] speed, rapid

Zec M., DÃijrr P., Schneider A. and Matthes F.

Improving Computer-Support for Collaborative Business Model Design and Exploration.

DOI: 10.5220/0005424000290037

In Proceedings of the Fourth International Symposium on Business Modeling and Software Design (BMSD 2014), pages 29-37

ISBN: 978-989-758-032-1

prototyping, quicker visualization, better collabora-

tion, simulation, and better planning [. . . ]”.

In this paper, we discuss the strengths and short-

comings of state-of-the art BMC software (e.g. facil-

itation capabilities). We introduce General Morpho-

logical Analysis as a powerful technique for complex

problem modeling and argue, that the design of CAD

software for business model design should borrow

useful concepts from computer-aided General Mor-

phological Analysis. The contribution of this paper

is the identiﬁcation of three key requirements for col-

laborative CAD software for business model design.

2 BUSINESS MODEL CANVAS

The BMC gained popularity in practice since it pro-

vides a simple framework and, as a result, facilitates

structured discussion about a hypothesized or actual

business model. The BMC promotes visual thinking

and a holistic perspective on a business model. Visual

thinking is a way to develop and clarify ideas about

designs and acts as a catalyst for new ideas (Bux-

ton, 2007). Furthermore, part of the power of visuals

is their ability to serve as a visible external memory

(Baddeley, 1998). Particularly whiteboards are an ef-

fective medium for visual thinking due to their free-

form nature (Walny et al., 2012).

The BMC is a visual one-page template for de-

scribing a business model (Osterwalder and Pigneur,

2010). The BMC comprises four perspectives on a

business model: Customers, Offer, Infrastructure, and

Financial Viability. Each perspective subsumes one or

more business model building blocks; (1) Customer:

Customer Segments, Customer Relationships, Chan-

nels, (2) Offer: Value Propositions, (3) Infrastructure:

Key Activities, Key Resources, Key Partners, and (4)

Financial Viability: Cost Structure, Revenue Streams.

The BMC is comprised of nine building blocks in to-

tal (see Figure 1).

The BMC is based on the Business Model On-

tology developed in (Osterwalder, 2004). The key

idea of the BMC is to provide a simple and com-

mon visual framework for communicating and de-

veloping business models. The BMC reached wide

adoption among practitioners such as business devel-

opment units, startup companies or seed accelerators.

There is no mandatory sequence according to

which the building blocks of the canvas have to be

worked out (Fritscher and Pigneur, 2010). Since the

building blocks are interrelated, various elements of

the business model have to be modiﬁed during mul-

tiple iterations anyway. However, many practitioners

start with identifying customer segments (Who are we

creating value for?) or the value proposition (Which

value are we creating?) and iterate over all remaining

blocks.

Irrespective of which starting point was chosen,

the business model designers scrape through the re-

maining building blocks in a sequence that seems rea-

sonable to them. They jump back and forth to reﬁne

the business model until they are satisﬁed. During the

process, alternative versions can be sketched out on

separate canvases or on the same canvas using differ-

ent colors. The ﬁnished BMC represents a business

model hypothesis and serves as a basis for subsequent

steps in the business model innovation process such as

validation or implementation.

In a workshop setting, the BMC can be printed out

on a poster or sketched on a whiteboard to provide a

shared display. The simple structure and the visual

arrangement of the building blocks provide a shared

language and point of reference for group discussions.

A major advantage of the pen-and-paper or white-

board approach is that elements (i.e. ideas) can be

added, edited, moved around and, if necessary, re-

moved spontaneously. Thus, the current state of the

group discussion is captured at any time and group

members do not forget ideas in the heat of the dis-

cussion. The use of color coding, drawings and rear-

rangement of elements helps maintain a clear organi-

zation of the canvas elements.

2.1 Software Support for BMC

A major drawback of the pen-and-paper as well as

whiteboard approach is that the BMC cannot easily

be shared with interested stakeholders. In addition,

the more elements the canvas contains the harder it

is to keep track of them since it gets more and more

difﬁcult to maintain clarity.

One way to share a physical BMC is to take a dig-

ital picture of it and send the picture to the relevant

stakeholder. Another way is to (re)create the canvas

using text-processing/graphics/slide-based presenta-

tion software or embedding the picture in an elec-

tronic document. In any case, creating a BMC using

generic software tends to be rather time-consuming

(Fritscher and Pigneur, 2010).

Despite a lack of thorough scientiﬁc investiga-

tion on the effectiveness of BMC mapping soft-

ware (Eppler et al., 2011), various BMC mod-

eling tools emerged for computer-aided business

model design to facilitate easier sharing while pro-

viding speciﬁc modeling facilities well adapted for

the BMC. CAD software for business model design

based on the BMC ranges from rather lightweight

browser-based canvas tools such as BM|DESIGN|ER

Fourth International Symposium on Business Modeling and Software Design

Figure 1: The Business Model Canvas (Osterwalder and Pigneur, 2010).

Figure 2: Screenshot of Business Model Fiddle

(Steenkamp, 2012).

(Fritscher and Pigneur, 2010), Business Model Fiddle

– Fig. 2(Steenkamp, 2012), Canvanizer (Proud Sourc-

ing GmbH, 2011) and Strategyzer (Business Model

Foundry GmbH, 2013) over meta-model-based wiz-

ards (Hauksson and Johannesson, 2014) to rich desk-

top software suites, e.g. BiZZdesign Architect (BiZ-

Zdesign, 2004).

Eppler et al. (2011) compared the use of generic

slide-based presentation software to BMC mapping

software. Their study found a positive effect of the

BMC on perceived collaboration and a negative ef-

fect on perceived creativity. The authors conclude

“that artefacts can have considerable power in shap-

ing group interactions and idea generation in the con-

text of business model innovation” and call for further

research on visual artefacts used to facilitate business

model innovation.

2.2 Beneﬁts of BMC Software

Particularly collaborative, web-based BMC soft-

ware offers two major advantages over pen-and-

paper/whiteboard sessions and general-purpose soft-

ware:

1. Support for Distributed Teams. Web appli-

cations facilitate collaboration across time, lo-

cation and organizational boundaries. Since

teams are often distributed in terms of at least

one of these dimensions, software-based business

modeling saves costs in comparison to pen-and-

paper/whiteboard workshops.

2. Easier and More Flexible Customization, Re-

use and Sharing. Canvas elements can be easily

customized in terms of coloring, typography and

position. Various media can be embedded (e.g.

images, video or spreadsheets). Canvases can be

saved in various formats, re-opened for further re-

ﬁnement and easily shared with stakeholders.

We analyzed three popular web-based BMC mod-

eling tools: Business Model Fiddle, Canvanizer, and

Strategyzer. All tool sticks to the one-page layout of

the paper template for their main view. The tools

differ slightly in the way how building blocks can

be ﬁlled with content and how elements can be cus-

tomized. Some tools offer features which extend the

original idea of the BMC. For instance, Strategyzer

provides a ﬁnancial estimator. Strategyzer and Can-

vanizer offer real-time collaboration, i.e. concurrent

editing and chat support. Business Model Fiddle

allows many customization of the canvas (e.g. re-

naming of building blocks) and the creation of snap-

shots to record changes. In addition, designers can

sketch on uploaded images and assign them to build-

ing blocks.

The main beneﬁts of CAD for the BMC approach

is the support for collaboration across time, location

Improving Computer-Support for Collaborative Business Model Design and Exploration

and organizational boundaries as well as reuse. The

analyzed tools mimic a paper/whiteboard and exploit

beneﬁts of information technology. However, there is

little facilitation support since in both, on-site meet-

ings and asynchronous distributed modeling sessions,

many methodological questions such as ﬁnding a rea-

sonable starting point have to be answered by the

modelers. None of the analyzed tools guides busi-

ness modelers through the process. While sticking to

the one-page layout, they slightly differ in the way

how building blocks can be ﬁlled with content and el-

ements be customized.

2.3 BMC and Facilitation

The role of the BMC in business model design is sim-

ilar to the notion of grammatical design which con-

stitutes boundaries within which designers can ﬁnd

creative solutions if they deviate from standard solu-

tions (Brown and Cagan, 1996). The BMC frames

the discussion and predetermines the general struc-

ture of the boundary object. BMC designers tend to

“think ‘within’ the given domains of the template”

(Eppler et al., 2011). They are supposed to be creative

within said domains. However, there is little guid-

ance on how to make use of one’s full creative poten-

tial. In practice, groups often ideate through sponta-

neous associations between concepts. Thus, the busi-

ness model design process that supplements the BMC

is intentionally informal and generic. It consists of

ﬁve phases which do not necessarily have to be gone

through in a linear manner (Osterwalder and Pigneur,

2010): (1) Mobilize (i.e. preparation), (2) Understand

(i.e. research and analyze the context), (3) Design

(i.e. generation of business model hypotheses using

the canvas), (4) Implement (i.e. implementing the best

hypothesis in the ﬁeld), and (5) Manage (i.e. moni-

tor the market and update the business model accord-

ingly). The process model does not prescribe what to

do in a speciﬁc phase. Instead, the authors refer to

various other tools and methods.

Business model generation demands a broad range

of skills, knowledge and experience as well as cre-

ativity. Thus, business model generation is often con-

ducted in teamwork. There is a widespread belief

that the performance of interactive groups is higher

than the performance of nominal groups (i.e. the

aggregate performance of the same number of non-

interacting individuals). However, this does not nec-

essarily hold true. For instance, various studies on

group performance in brainstorming have shown that

nominal groups outperform interactive groups, e.g.

(Diehl and Stroebe, 1987; Mullen et al., 1991). While

teams tend to generate more ideas if they follow Os-

born’s brainstorming rules (Osborn, 1957) than when

they do not stick to those rules, they are still not as ef-

fective as nominal groups. Explanations for this phe-

nomenon include social loaﬁng (Paulus et al., 1993),

social anxiety (Camacho and Paulus, 1995) and pro-

duction block (Diehl and Stroebe, 1987).

While the BMC deserves merit for providing a

common language for group discussion, teams have

to be careful. Pitfalls in group brainstorming are only

one example illustrating the intricacy of group dy-

namics in collaborative settings.

Teamwork can yield group process gains as well.

For instance, team members might become more mo-

tivated when facing social competition. They might

also improve their skills due to knowledge transfer.

And even coordination gains are possible (e.g. con-

ductor of an orchestra).

Generally, process gains as well as losses can oc-

cur in terms of motivation, capabilities and coordina-

tion. A skilled and experienced facilitator promotes

process gains and mitigates process losses. For in-

stance, they play devil’s advocate to encourage al-

ternative or counter-intuitive thoughts and, conse-

quently, avoid groupthink and shared information bias

(Baker, 2010). However, teams often do not involve a

skilled facilitator because it is considered too expen-

sive, they think there is no need for a facilitator (since

negative effects are implicit and thus hard to notice)

or there is simply no facilitator available.

In many situations, collaborative web-based CAD

software is expected to be used without a dedicated

facilitator, mostly to save time and money. There-

fore, we argue that collaborative BMC software has

to compensate for the lack of a skilled facilitator as

much as possible. We expect CAD software for busi-

ness design to enforce meta-model constraints and

validate the model (e.g. impose mappings between

value propositions and customer segments). This

way, the BMC modeling tool can provide, for in-

stance, helpful directions in the case of meta-model

violations. Hauksson et al. (2014) implemented

a desktop-based wizard for business model design

which enforces BMC meta-model constraints.

While it is unlikely that software can fully replace

a human facilitator, there are some simple yet power-

ful techniques to improve collaboration and ideation

in groups. For instance, gamiﬁcation techniques can

help mitigate motivation losses. Allowing anonymous

contributions might reduce social anxiety or evalu-

ation apprehension and, as a result, increase diver-

sity of ideas since ideas can be expressed uninhib-

itedly. Separation between divergent thinking (de-

ferring judgment, creating as many ideas as possi-

ble) and convergent thinking (think and evaluate ideas

Fourth International Symposium on Business Modeling and Software Design

analytically), shown to improve creativity processes

and results (Cropley, 2006), can be supported by im-

plementing different views and means of interaction

for each of those phases. In addition, differentiat-

ing between single and team phases might increase

the quantity of ideas: during single phases contribu-

tions from other team members should be hidden to

avoid production block and mutual inﬂuence, during

team phases contributions from others should be visi-

ble such that they can serve as an inspiration for addi-

tional ideas and reﬁnements. While we do not pro-

vide an exhaustive list of design recommendations,

we argue for the integration of insights from social

psychology and creativity research into the design of

future CAD tools for business modeling.

2.4 Discussion

In general, the BMC captures one speciﬁc business

model. While it is possible to sketch more business

models on the same canvas at the same time, the can-

vas tends to get confusing. Therefore, in practice,

each business model is usually sketched out on a sep-

arate canvas.

The BMC facilitates a structured discussion about

a business model but it does not guide its users

to explore the formal solution space systematically.

Rather, users tend to start with specifying one build-

ing block and then scrape from building block to

building block. However, this heuristic approach

might miss innovative and viable solutions. One

method to circumvent this disadvantage and to ex-

plore the full solution space is Morphological Anal-

ysis which is presented in the next section.

Another issue that arises in collaborative settings

which is not addressed by the BMC is group dynam-

ics. Social psychology literature has shown how vari-

ous negative effects such as social loaﬁng tend to oc-

cur in group settings. As a result, team performance

and productivity might degrade considerably.

3 USEFUL CONCEPTS FROM

MORPHOLOGICAL ANALYSIS

FOR BMC SOFTWARE DESIGN

In this section, we introduce General Morphological

Analysis (GMA), a generic problem modeling tech-

nique, and associated software tools. We argue that

they feature useful concepts which should be con-

sidered to be adopted by CAD software for business

modeling. In a broader sense, Morphological Analy-

sis (MA) is concerned with the study of form, struc-

Figure 3: The general structure of a morphological ﬁeld.

One speciﬁc formal conﬁguration is highlighted in gray.

ture and interconnections between structural elements

(Shurig, 1986). Various disciplines such as linguistics

or biology conduct subject-speciﬁc variants of MA.

Swiss Astronomer Fritz Zwicky developed a generic

type of MA commonly referred to as General Mor-

phological Analysis (GMA) (Ritchey, 1998). GMA is

a method which aims to facilitate system or problem

understanding and structuring. GMA is particularly

suited for multi-dimensional, non-quantiﬁable prob-

lems for which mathematical or causal modeling is

not applicable or appropriate (Ritchey, 1998). The

general idea of GMA is to derive a non-quantiﬁed

model of the system or problem under examination

by identifying its key structural components. In the

following, we will focus on GMA for problem struc-

turing and solving.

3.1 Method

The initial step of GMA is to clarify the problem

statement such that there is a shared understanding

of the problem. The result of this step is a set of as-

pects, or parameters, that seem to be the most relevant

characteristics of the problem at hand.

GMA distinguishes between parameters (also:

components or dimensions) and parameter values.

The ﬁrst step of GMA is to break down the prob-

lem into subcomponents. Ideally, the subcomponents

are mutually exclusive and collectively exhaustive. In

practice, there might be some overlap between pa-

rameters. However, the overlap should be as small

as possible. Team members have to discuss and col-

laboratively devise a set of parameters which captures

all key aspects of the problem at hand. This step fos-

ters a shared understanding of the problem because

the group needs to reach a consensus about what the

key aspects of the issue are. For business model de-

sign, the BMC provides an established problem struc-

ture (i.e. the nine building blocks).

Once an adequate set of parameters is found, the

range of parameter values has to be speciﬁed for each

parameter. Parameter values can be qualitative or

quantitative. The level of abstraction is speciﬁc to the

concrete problem and purpose of the analysis. What

is important in this step of GMA is to investigate each

parameter independently. This approach promotes di-

Improving Computer-Support for Collaborative Business Model Design and Exploration

Figure 4: A screenshot from Parmenides EIDOS showing a morphological ﬁeld which represents the strategic option space

for an eBike Mobility Service.

vergent thinking and openness for counter-intuitive

values. The focus at this point is on formal properties.

Evaluation of feasibility should be deferred. We argue

that this is a major advantage over the BMC method

because individuals are tempted to think about differ-

ent building blocks at the same time because they in-

stinctively try to interrelate different building blocks

(e.g. ”‘if we want to sell our product to young people

we have to build a mobile platform as a channel”’).

While this approach can make sense, innovative ideas

might not be taken into consideration because the dis-

cussion is centering around familiar business model

patterns. Thus, business modelers constrain their cre-

ativity and, consequently, might miss particularly in-

novative business model designs.

Parameters and their value ranges constitute a

morphological ﬁeld (MF) which is usually repre-

sented by a matrix (see Figure 3). The MF matrix

is a dense representation of the formal conﬁguration

space. The formal conﬁguration space is the set of

all morphological conﬁgurations (i.e. parameter value

combinations containing one and only one value per

parameter). Depending on the perspective and pur-

pose of the analysis, the formal conﬁguration space is

sometimes referred to as formal problem space or for-

mal solution space. To sum up, GMA is a structured

problem modeling method which enables its users to

establish the space of all formal solutions, systemati-

cally discuss the contained formal conﬁgurations and

identify the best solution.

GMA prescribes clear separation between diver-

gent thinking and convergent thinking. By contrast, in

BMC workshops, participants tend to mix both styles

of thinking too frequently and, as a result, do not sys-

tematically explore the space of bounded creativity.

3.2 Example

GMA can be used for various purposes such as sce-

nario analysis, product innovation or strategy devel-

opment. Figure 4 shows an example for a MF rep-

resenting the strategic option space for an eBike Mo-

bility Service in terms of BMC terminology. Each

parameter (i.e. “building block”, gray background)

can take a value from its parameter range (depicted

in yellow or white, respectively). A formal business

model is given by a speciﬁc conﬁguration from the

morphological ﬁeld (i.e. solution space). For in-

stance, a speciﬁc formal business model is given by

(“First in Technology”, “Pragmatists: Short-distance

Travelers”, “One-Stop Information Shop”, “Flagship

Stores”, “Vehicle Leasing”, “Public Transport Com-

panies”, “Research & Development”, “Promotion”,

“Patents”). Not all possible conﬁgurations represent

viable business models since some business model el-

ements might be incompatible. However, using GMA

the complete set of formal solutions can be identiﬁed

and systematically evaluated. GMA software can help

Fourth International Symposium on Business Modeling and Software Design

Figure 5: Consistency matrix in MA/Carma. Adopted from

(Ritchey, 2005).

reduce the formal solution space to a viable solution

space using various techniques such as cross consis-

tency assessment or clustering.

3.3 Sofware Support for GMA

Conducting a GMA by hand has one major challenge:

the size of the solution space grows exponentially

with each additional parameter. As a result, it is of-

ten not possible to evaluate all formal solutions of

the MF (e.g. a 5-parameter MF with 6 possible val-

ues for each parameter yields 6

= 7776 formal so-

lutions. The example in Figure 4 contains 480.000

formal business models.). Therefore, in workshops

without software support, the solution space cannot

be analyzed exhaustively. Rather, only a small subset

of conﬁgurations is selected for deeper analysis ac-

cording to subjective preferences and within objective

constraints (e.g. time limit). GMA software addresses

the challenge of large solution spaces by providing

means for reduction of the formal conﬁguration space

to a practical solution space. Examples for software

which supports GMA and solution space reduction

include Parmenides EIDOS (Parmenides Foundation,

2014) and MA/Carma (Ritchey, 2005).

MA/Carma allows the creation of an inference

model. First, the user has to specify the consis-

tency (or compatibility) of each pair of parameter val-

ues. Given such a consistency matrix (see Figure 5),

MA/Carma generates an interactive inference model

(see Figure 6). The user can declare arbitrary param-

eter values to be exogenous (colored red or medium

Figure 6: An interactive inference model constructed by

MA/Carma. Adopted from (Ritchey, 2005).

gray, respectively) and the software calculates which

values are still viable for the remaining parameters

(colored blue or dark gray, respectively). A detailed

description can be found in (Ritchey, 2005). Par-

menides EIDOS supports a similar technique: users

have to specify a numeric consistency value for each

pair of parameters. Then, a consistency value for

each conﬁguration is calculated by the software. As

a result, in their subsequent analysis, users can focus

on conﬁgurations which yield the highest consistency

values. In contrast to MA/Carma, Parmenides EIDOS

supports clustering of similar conﬁgurations to iden-

tify more abstract patterns among viable solutions.

3.4 Discussion

GMA is a simple problem structuring method for in-

dividuals and groups. It can be used for various prob-

lems such as product innovation, strategy develop-

ment or scenario analysis. MA is particularly useful if

the problem at hand cannot be adequately expressed

in a mathematical model, many different stakeholders

are involved and various viewpoints have to be con-

sidered.

MA aims at constructing a formal solution space.

Formal solutions in the solution space might be in-

consistent and/or impractical. Solution space reduc-

tion techniques such as cross consistency assessment

in combination with complementary software support

help rule out inconsistent conﬁgurations and reduce

the solution space signiﬁcantly. This way, the solution

space can be explored exhaustively yet efﬁciently.

To the best of our knowledge, there is no GMA

software that is inherently collaborative and supports

distributed teams. While GMA software such as Par-

menides EIDOS or MA/Carma is used by facilitators

to provide a shared display in workshops, their user

interface and interaction design is tailored to individ-

ual users.

Moreover, we have not found any GMA software

Improving Computer-Support for Collaborative Business Model Design and Exploration

that accounts for psychological aspects and group dy-

namics. However, we argue that BMC software de-

sign can borrow from GMA and respective software.

They key advantage of the software-based GMA ap-

proach to modeling of complex problems such as

business model innovation is the systematic construc-

tion of the formal solution space and leveraging soft-

ware to ﬁnd viable solutions within the practical solu-

tion space.

A key feature of MA/Carma is the construction

of an inference model of the solution space. Design-

ers can interact with the model and analyze the de-

pendencies of strategic decisions by treating speciﬁc

parameter values as ﬁxed input. Given a consistency

matrix, the software automatically calculates which

options remain to be viable for all other parameters.

We argue that such an inference model provides more

insights into the business modeling space than the de-

piction of the BMC. The BMC is a descriptive model

that only represents one particular business model (or

multiple descriptive models if colors are used.).

4 KEY REQUIREMENTS FOR

FUTURE BMC SOFTWARE

In the previous sections, we identiﬁed three key re-

quirements for future CAD software for business

model design: (1) the ability to perform an interactive

“what-if” analysis (inference capability), (2) method-

ical guidance throughout the design process (facilita-

tion capability) to mitigate negative effects of group

dynamics, and (3) the ability to collaborate across

time, location and organizational boundaries (support

of distributed teams).

The inference capability enables designers to ex-

periment with business models by declaring partic-

ular business model elements as exogenous and an-

alyzing the implications for the remaining building

blocks. The facilitation capability offers guidance

during the design process in order to increase cre-

ativity of the designers while limiting negative effects

stemming from group dynamics. In addition, bor-

rowing techniques from GMA software, future BMC

software might provide a concrete process model for

systematic generation of business models within the

conceptual boundaries of the BMC while still allow-

ing and promoting creativity (i.e. parameter (value)

deﬁnition, cross consistency assessment and conﬁg-

uration space reduction). The support of distributed

teams refers to the ability to carry out the business

model design process cost-efﬁciently in a distributed

setting.

Collaborative, web-based CAD software support-

ing the BMC is readily available. Thus, there is BMC

support for distributed teams. However, those tools

lack inference and facilitation capabilities.

Therefore, we propose to extend BMC software

by borrowing concepts from GMA and respective

software. Neither BMC methodology nor GMA ad-

dress potential group process losses such as motiva-

tion loss (e.g. social loaﬁng) or skill impairment (e.g.

social anxiety). Thus, groups might not exploit their

full cognitive and creative potential.

We argue that methods as well as tools for col-

laborative business model generation have to address

negative effects of teamwork and try to mitigate them.

Skilled human facilitators might accomplish this task.

However, assessing the skill of a human facilitator is

hard. In addition, in some contexts there is no possi-

bility to hire an experienced facilitator or it might be

considered too expensive.

The absence of a facilitator can be compensated in

BMC software by implementing various facilitation

and creativity techniques. For instance, creativity can

be increased by clearly separating between divergent

and convergent thinking phases. Alternating between

individual and team ideation helps decreasing nega-

tive impact of anchoring and groupthink. Anonymity

can help reduce social anxiety. On the other hand,

anonymity might increase social loaﬁng since contri-

butions cannot be attributed to individuals and, thus,

individuals might ”‘hide”’ behind the team. Facilita-

tion capabilities have to be well-conceived and evalu-

ated.

5 CONCLUSION AND FUTURE

WORK

We identiﬁed beneﬁts and shortcomings of state-of-

the art CAD software for the BMC methodology.

We agree with Osterwalder, Fritscher et al. that

CAD software for business model design is likely

to improve strategic planning, particularly for dis-

tributed business model design teams which face in-

creasing competition (Osterwalder and Pigneur, 2013;

Fritscher and Pigneur, 2014). The BMC deserves

merit for providing a common (visual) language for

business model design. However, we think that CAD

support for business model design is still only in its in-

fancy and that various questions need to be answered

about how to design effective next-generation CAD

software for collaborative business model design. We

expect future BMC software to leverage the potential

of software support and provide additional value that

goes beyond mere digitization of the BMC approach.

Our next step is to reﬁne the general requirements

Fourth International Symposium on Business Modeling and Software Design

identiﬁed in this paper. Then, we will build a pro-

totype which implements the ideas discussed above

(e.g. inference and facilitation capabilities) and vali-

date our hypotheses.

REFERENCES

Baddeley, A. (1998). Recent developments in working

memory. Current opinion in neurobiology, 8(2):234–

238.

Baker, D. F. (2010). Enhancing group decision making: An

exercise to reduce shared information bias. Journal of

Management Education, 34(2):249–279.

BiZZdesign (2004). BiZZdesign Architect. http://

www.bizzdesign.com. [Online; accessed April 15,

2014].

Brown, K. N. and Cagan, J. (1996). Grammatical design

and bounded creativity. Carnegie Mellon University,

Department of Mechanical Engineering.

Business Model Foundry GmbH (2013). Strategyzer.

http://www.strategyzer.com. [Online; accessed April

15, 2014].

Buxton, B. (2007). The Anatomy of Sketching. Morgan

Kaufmann, San Francisco and CA and USA.

Camacho, L. M. and Paulus, P. B. (1995). The role of so-

cial anxiousness in group brainstorming. Journal of

Personality and Social Psychology, (68):1071–1080.

Carlile, P. R. (2002). A pragmatic view of knowledge and

boundaries: Boundary objects in new product devel-

opment. Organization Science, 13(4):442–455.

Chesbrough, H. (2010). Business model innovation: Op-

portunities and barriers. Long Range Planning , 43(2-

3):354–363.

Cropley, A. (2006). In praise of convergent thinking. Cre-

ativity Research Journal, 18(3):391–404.

Diehl, M. and Stroebe, W. (1987). Productivity loss in

brainstorming groups: Toward the solution of a rid-

dle. Journal of Personality and Social Psychology,

53(3):497–509.

Eppler, M. J., Hoffmann, F., and Bresciani, S. (2011). New

business models through collaborative idea genera-

tion. International Journal of Innovation Manage-

ment, 15(6):1323–1341.

Fritscher, B. and Pigneur, Y. (2010). Supporting business

model modelling: A compromise between creativity

and constraints. In Task Models and Diagrams for

User Interface Design, 8th International Workshop,

Revised Selected Papers, pages 28–43, Brussels and

Belgium.

Fritscher, B. and Pigneur, Y. (2014). Computer aided busi-

ness model design: Analysis of key features adopted

by users. In 47th Hawaii International Conference on

System Science, pages 3929–3938, Waikoloa and HI

and USA.

Hauksson, H. and Johannesson, P. (2014). Metamodeling

for business model design. facilitating development

and communication of business model canvas (bmc)

models with an omg standards-based metamodel. In

The 8th International Workshop on Value Modeling

and Business Ontology (VMBO 2014), Berlin and

Germany.

Mitchell, D. W. and Coles, C. B. (2003). The ultimate com-

petitive advantage of continuing business model inno-

vation. Journal of Business Strategy, 24(5):15–21.

Mitchell, D. W. and Coles, C. B. (2004). Establishing a con-

tinuing business model innovation process. Journal of

Business Strategy, 25(3):39–49.

Mullen, B., Johnson, C., and Salas, E. (1991). Productivity

loss in brainstorming groups: A meta-analytic integra-

tion. Basic and Applied Social Psychology, (12):3–23.

Osborn, A. F. (1957). Applied imagination: Principles

and Procedures of Creative Problem Solving. Charles

Scribner’s Sons, New York and NY and USA.

Osterwalder, A. (2004). The business model ontology:

A proposition in a design science approach, volume

HEC 173. Institut d’Informatique et Organisation,

Lausanne and Switzerland.

Osterwalder, A. and Pigneur, Y. (2010). Business Model

Generation: a handbook for visionaries, game chang-

ers, and challengers. Jon Wiley & Sons.

Osterwalder, A. and Pigneur, Y. (2013). Designing business

models and similar strategic objects: The contribution

of is. Journal of the Association for Information Sys-

tems, 14(5):237–244.

Parmenides Foundation (2014). Parmenides EIDOS.

https://www.parmenides-foundation.org/application/.

[Online; accessed April 15, 2014].

Paulus, P. B., Dzindolet, M. T., Poletes, G., and Cama-

cho, L. M. (1993). Perception of performance in

group brainstorming: The illusion of group produc-

tivity. Personality and Social Psychology Bulletin,

(19):78–89.

Proud Sourcing GmbH (2011). Canvanizer.

http://www.canvanizer.com. [Online; accessed

April 15, 2014].

Ritchey, T. (1998). Fritz Zwicky, Morphology and Policy

Analysis. In 16th Euro Conference on Operational

Analysis, Brussels.

Ritchey, T. (2005). MA/Carma: Advanced Com-

puter Support for General Morphological Analysis.

http://www.swemorph.com/macarma.html. [Online;

accessed April 15, 2014].

Shurig, R. (1986). Morphology: A knowledge tool. Systems

Research, 3:9–19.

Steenkamp, J. (2012). Business Model Fiddle.

http://www.bmﬁddle.com. [Online; accessed

April 15, 2014].

Walny, J., Carpendale, S., Henry Riche, N., Venolia, G.,

and Fawcett, P. (2012). Visual thinking in action:

Visualizations as used on whiteboards. IEEE Trans-

actions on Visualizations and Computer Graphics,

18(12):2779–2788.

Improving Computer-Support for Collaborative Business Model Design and Exploration