USING RICH INFORMATION TO PLAN MOBILE FINANCIAL SERVICES

APPLICATIONS WITH MAXIMUM POSITIVE IMPACT: A CASE STUDY.

Ken Peffers Tuure Tuunanen

ISMT Department Information Systems Science
Helsinki School of Economics
HK University of Science and Technology Helsinki, Finland
Clearwater Bay, Kowloon Tel. +358 9431 39283
Hong Kong Mobile +358 40 544 5591
Tel: +852 9525 1041 Fax: +358 40 644 5591
Fax (US) +1 603 696 1289 Email: [Link]@[Link]
Email: k@[Link]
 USING RICH INFORMATION TO PLAN MOBILE FINANCIAL SERVICES

APPLICATIONS WITH MAXIMUM POSITIVE IMPACT: A CASE STUDY.

For next-generation wireless devices, firms planning financial services

applications face a familiar high-tech problem, “How do you know what customers
will want and use of technology that has never been used before?” Many innovations
work fine in the lab and factory only to fall flat with customers. Here we propose a
method to help managers better understand what new applications and features users
will value most and why. We apply the method in a case study involving the
development of financial service applications for mobile devices.
Researchers have long argued for the importance of information richness to
innovation management. In IS planning this has been addressed, not entirely
satisfactorily, with a variety of methods based on bottom-up or top-down approaches.
Top-down approaches assume that managers know everything about what features are
needed, while bottom-up approaches are faulted for lack of strategic focus.
We develop an IS planning method, Critical Success Chains (CSC), that
allows analysts to make economic use of rich information gathered from a widely
representative group of participants to create rich models of system features that the
users think are important to the firm and the reasons why. The models are used in
technical workshops to generate ideas for new systems that will satisfy the
relationships uncovered in the models.
We used CSC in a project to develop ideas for mobile financial services
applications at Digia, a Finland-based R & D firm. We selected a group of 32
participants, including experts and potential early adopting users, contacting each to
elicit an application idea to use as stimuli. In individual interviews with each, we
asked participants to rank order several of the stimuli. Then we asked why the
participant preferred particular ideas and the specific features the participant would
expect to be part of an application. This data was recorded as linked chains connecting
features with consequential performance and with the perceived value of such
performance. The data was analyzed across participants to create network models of
features, performance, and values.
We conducted an ideation workshop with Digia engineers and executives to
generate ideas for new systems based on the CSC models. Workshop participants
created back-of-the-envelope level ideas for three new applications, “personal
transaction assistant,” “my financial advisor,” and “mobile wallet.” A fourth CSC
didn’t result in new ideas because workshop participants felt that Digia wouldn’t be a
part of any resulting value chain. In addition, workshop participants developed
network value models to show graphically the flow of information, value, and revenue
among parties involved in producing and using the applications.
 INTRODUCTION

For next-generation wireless devices, firms planning financial services

applications face a familiar high-tech problem, “How do you know what customers

will want and use of technology that has never been used before?” Many innovations

work fine in the lab and factory only to fall flat with customers as managers and

engineers develop applications without knowing what customers want or are willing

to pay for.

The problem of innovative applications

In the past innovations required many years before users widely adopted them.

For example, in the US, 10 % penetration among households required 48 years for

television, 28 years for the Internet, and 20 years for mobile telephones (Powell

2001). Of course, not all, or even most, innovations are eventually successful. Many

innovative applications were never successfully adopted. Picture telephones,

appliances that display the pictures of callers to each other, have been technically

feasible since the mid-1920’s, were incorporated into the lore of Dick Tracy cartoons

in the 1930’s and actually developed by AT&T in the mid-1960’s, but have never been

successfully rolled out. In the 1960’s observers thought that, by the beginning of the

21st century, most retailing would be done by machine. Now that the millennium has

come, vending machines are relegated to niche products and situations.

The case of mobile commerce

Observers are concerned about whether mobile commerce applications can be

rolled out successfully, that is, whether firms will be able to develop and roll out

applications that customers value sufficiently so that they are willing to pay more for

them than they cost to produce and deliver. There is much cause for concern. Network

0
operators have paid huge license fees for the rights to build third generation mobile

networks, capable of carrying data with sufficient bandwidth to make the use of

WWW-based applications feasible. Two dramatic recent business failures, that of the

Iridium network, which cost $7 billion to set up, but only garnered 100,000 customers

(Rockman 2001) and may now be worthless, and the Metricom Ricochet network, a

128 kb/s wireless service in the U.S. that shut down in August 2001 (Fitchard 2001),

suggest that, aside from a small number of technophiles, users intend to be very

choosy about which innovative technologies and innovations they adopt.

Senior executives are concerned that they don’t have a clue about what m-

commerce applications customers might be willing pay for (Quotient

Communications 2001). So far, Marketing research suggests, the ‘killer application’

that will make m-commerce successful has not yet been identified (Nelson 2001). In

Japan, where 3G services were test marketed in 2001, customers, many of whom were

gadget enthusiasts, seemed to be blasé about the services, suggesting that they would

only be interested if it were much cheaper. Said one, “…there’s nothing new on the

service that I really feel I must have (Bickers 2001).” Other research has reported that

customers are primarily focusing on paying less for current services (Hadden 2001),

rather than looking eagerly forward to new applications.

The very modest success of WAP may be both and indicator and a cause of

some of the disinterest among customers. Many customers see m-commerce in light

of the limitations of WAP services and appliances: poor resolution, slow processes,

and few interesting services (Latour 2002). The result of these limitations has been

poor WAP penetration and, according to a JD Power study, public interest in mobile

internet services that has actually decreased recently (Broersma 2002). In the area of

1
financial services, WAP applications have been “a frustrating experience” for the

customer (Sangani 2002).

All of this points to the need for new business models that can capture

valuable applications for which customers are willing to pay more than the cost.

How to determine what innovative applications customer will value the most?

Research in IT planning has recognized this issue. Researchers, e.g., Segars

and Gover (1999), have identified information systems planning (ISP) characteristics

that lead to its success, including widespread participation among employees across

the firm, to incorporate the ideas of many, and a focus on projects that have the most

potential to be important for the firm.

For IT investments to help the firm to achieve its strategic objectives (Earl,

1993), ISP should be consistent with the firm’s explicit or implied strategy of the firm

(Tillquist 2000). For ISP to be strategic, it must have a "top-down" perspective, i.e., be

driven by information from senior managers (Shank, Boynton, and Zmud 1985).

Senior managers don’t have a monopoly on understanding what customers want and

are willing to pay for, however, particularly when it comes to applications and

products which have never been used before. In this case it is important that planning

involve extensive participation. Implicit in the value placed on top-down planning is

an idea that only senior managers can have a strategic focus (Earl 1993). Others in the

organization, in the community and among customers, however, usually know much

about what is important for the firm to succeed with new products. Where innovative

applications are contemplated, participation from outside the firm may be crucial, as

the effectiveness of such systems is often determined by current and future supplier

and customer needs and expectations (Lockamy and Smith 1997). Exclusive use of

top-down planning ignores this grass-roots knowledge (Premkumar and King 1994)

2
and effectively shuts the firm off from the effective use of the knowledge of many

people in and around the organization, a big mistake in an era where quick, proactive

adaptation to new environments is required (Cornell 1997).

Additionally, researchers have recognized the importance of understanding,

based on collection, interpretation, and communication of rich information, to

strategic decision-making, rather than reliance solely on the deterministic use of

normative firm models (Boland, Tenkasi, and Te’eni 1994).

Critical Success Chains

Critical success chains (CSC) (Peffers and Gengler 2002) is a method of data

gathering and analysis for ISP that seeks to make use of the knowledge of many

people in and around the organization to develop feasible ideas for new systems that

are potentially important to the firm.

Use of CSC, it is claimed, has the potential to positively affect the portfolio of

IS projects available to the firm. The CSC method can result in better information

from wider participation, provides rich information important for planning, helps keep

focus on what is important for the firm, is likely to result in a more complete portfolio

of ideas for systems across the strategic grid, and may help with user buy-in for

eventual development.

We used CSC at Digia, Inc. to develop a portfolio of financial services

applications to consider implementing across a multi-organizational value chain that

might include banks, mobile network operators, wireless appliance manufacturers,

retailers, and information intermediaries (infomediaries).

3
 DIGIA CASE STUDY

Charge from the firm and study scope

Digia, Inc. is a research and development firm, founded in 1996, specializing

in innovative software applications for wireless communication. Its current strategic

focus is on the creation of personal communication technologies and applications for

the next generation wireless information devices (WID), such as Smartphones and

Communicators. Digia enables WID manufacturers, i.e., the largest mobile handset

manufacturers, mobile operators and related firms to shorten substantially their

product development cycles. Digia is a private corporation, with substantial

ownership by founders, Pekka Sivonen, Mika Malin and Jarkko Virtanen, and by

firms, such as Cisco Systems, General Electric, Intel, Sony and Sonera, and venture

capital firms. Digia’s 2000 revenue was about USD 5.5 million.

In fall 2000 Digia chairman, Pekka Sivonen, approached us with a request to

“give [him] financial cocktails for mobile commerce.” He wanted a portfolio of

potential applications for Digia to develop to meet the need for financial services

delivered by the next generation wireless devices. It was imperative to insure that the

applications developed by Digia were “killer cocktails,” i.e., so well embraced by

customers that they helped to insure the acceptance of the next generation wireless

devices in a way similar to the way VisiCalc helped to insure the acceptance of the PC

by business. The scope for the applications was loosely defined as somehow involved

with money and financial services.

Participant selection

Since development, implementation and use of the desired applications would

involve people up and down the value chain, it seemed desirable to obtain input from

4
a representative group of involved people from outside the company, including

bankers, risk investors, mobile telecommunications operators, researchers, and

potential end users. Most of these people would be from outside the firm.

To elicit participation we developed two lists of potential participants, experts

and end users. For potential expert participants, the analyst worked with Digia’s

chairman and a staff assistant to develop a list that included a cross-section of

Finland’s most relevantly knowledgeable scientists, professionals, and managers.

For end-users we decided that it would be desirable to elicit participation from

sophisticated communications technology users, people who would be likely early

adopters of attractive new mobile commerce applications. We defined demographic

characteristics for the participants according to the segmentation used in mobile

value-added services study (Nokia 1999). For the study, the business mobile user

segment, presented in table 1, was considered the most appropriate.

Table 1. The business mobile user segment, based on (Nokia 1999).

THE BUSINESS USER SEGMENT OF MOBILE VALUE ADDED SERVICES

Description Aware and have used mobile value added services (VAS)
Wealth Not restrained by monetary issues
Education Highly educated, working on the managerial level or as an expert
Technology orientation Internet and information technology littermates
Mobility Very mobile people, participating actively in sports and outdoor activities
SMS usage Moderate SMS-users (1-7 mobile originated SMS/week)

To avoid excessive analyst influence in the selection of participants, several

faculty members at the Helsinki School of Economics, not including the analysts,

were asked to nominate professional, managerial, executive, and other sophisticated

end-users for participation. The resulting combined list of nominated participants

contained 40 names.

5
Data collection

We attempted to contact each of the nominated participants by telephone to

elicit participation in the study, schedule an appointment, and elicit a system idea for

use as stimulus. Every one of the nominated participants that could be reached by

telephone agreed to participate. Eight people couldn’t be reached, attributable to a

combination of successful gate-keeping activities and the influence of the approaching

Christmas season.

The resulting 32 participants included 18 outside experts and 14 potential end-

users, including just one Digia employee. The experts included 3 IS development

managers, 4 academic researchers, and 12 vice presidential level (minimum)

executives from m-commerce related industries. The potential end-users included 5

managers and executives from a variety of industries, 4 professionals, and 2 university

students. All of the participants were from the Helsinki, Finland vicinity and all were

mobile telephone users.

Table 2 shows a profile of the study group. All participants can be categorized

as moderate to heavy SMS messaging users. Several of the participants told us that

they send more than 100 messages per week. Internet use for the group was estimated

to be 2.8 hours per day on average. All except one participant were familiar with

mobile value-added services and most of them used services at least occasionally, up

to 5 times per week. The average age was 33 years and the participants estimated that

they spent 25% of their working time outside of their office on average. About half of

the participants had sold or bought stocks during the past year.

6
Table 2. Demographic profile for all participants in Digia study.

Age 20-29 yrs: 38 % 30-39 yrs: 44 % 40- yrs: 18 %

Education Bachelor: 31 % Masters: 56 % Doctorate: 13 %
Sex Male: 84 % Female: 16 %
Use of Mobile VAS 0-5 per week: 50 % 5-8 per week: 22 % over 8 per week: 25 %
SMS messages/week 0-19 per week: 38 % 20-49 per week: 38 % 50+ per week: 24 %
Internet Banking used regularly: 91 % Stock trading with Internet banking 47 %
Mobility (% of time 0-29%: 22 % 30-69%: 50 % 70%+ : 28 %
working out of office)

To elicit ideas for use as stimuli, while setting interview appointments, we

asked each participant to name the first thing that came to mind when they thought

about financial services that could be provided with the next generation mobile

devices. All but 2 of the participants provided ideas. Since the responses seemed to

group naturally into four sets, we rewrote them as four bland descriptions, from which

participants could infer specific features.

During five weeks beginning in December 2000 and ending in January 2001,

we interviewed each participant individually. The holiday season created some

scheduling problems and for this reason the interviewing period was prolonged. The

interviews averaged 50 minutes and ranged from 30 to 90 minutes. In each interview

we spent the first 5 to 10 minutes on warm-up conversation and discussion of the

interview objectives. Then we showed the participant the four system descriptions and

asked to him/her rank order the best two. Generally the participants volunteered ideas

about system features. When necessary, we also showed participants prototype

pictures of next generation mobile devices to establish the right context, emphasizing

that we were not limited to current text-based and black-and-white devices with low-

rate data transmission capabilities.

7
 Next we asked the participant a series of questions to collect chains of

attributes, CSF consequences, and personal objectives according to the laddering

method. Starting with the sample application that the participant ranked most

important, we asked, “Why would this application be important to you?” Then, in

response to the answer, “Why would that be important to you?” We continued in this

manner until the participant couldn’t continue; he/she had reached an ultimate

personal value. Then, referring to the participant’s first answer, we asked, “What

would it be about this application that would make you think that it would do that?”

We continued in this vein until the participant couldn’t give another response; he/she

had reached a concrete feature that he/she assumed would be part of the application.

The majority, all but five, of the interviews were done at the participant’s work

premises. Participants seemed much more at ease in interviews at their own work

premises, especially when it was in a conference room, away from interruptions,

rather than in the participant’s office. The interviews were tape-recorded.

After the interviewing, the chains were extracted from the tapes and were

written down. This process lasted four weeks for the 32 tapes. Later this analyst

learned that in previous studies the chains had been recorded on paper during the

interview, a much more efficient method (Peffers and Gengler 2000, 2001). Recording

the chains on paper during the interview helps to structure the interview and gives

clues to the participant about the purpose of the seemingly very tedious line of

questions. However, the interviews were well structured, so aside from two interviews

where it was challenging, transcribing the chains was a straightforward task. In total

147 chains were recorded averaging 4.59 per participant. An example chain is shown

in figure 1.

8
 Personal values Able to function
properly
Life is in order

Easy to search
information
No receipts to
store
No need to sign Performance impacts
anything

Less things to lose

No separate wallet

Pay micropayments
instantly

It is always with you Expected system feature

Figure 1. Example chain collected from a participant in the Digia study.

Analysis

We began the analysis by clustering the unique individual statements into

consistent constructs across participants. Each participant responded in the individual

interviews with unique statements. Consequently, the 147 chains contained more than

1000 individual statements. This clustering was accomplished by means of a

qualitative hierarchical clustering technique wherein two analysts sequentially

browsed through the list of statements and attached common labels to pairs of

statements that had very similar meanings. The analysts continued to take turns

clustering the statements and labels until further clustering would have resulted in

substantial lost meaning. At this point the more than 1000 statements had been

clustered into 114 concepts, representing system features, performance consequences

and personal values.

Next we mapped the chains into a matrix where the 147 rows represented the

chains and the 114 columns represented the concepts. In a cell, a one indicated that

9
the concept was present in the chain and a zero indicated that it was not present. We

clustered the chains, using Ward’s method to minimize the variance of the constructs

contained in each cluster, aggregating the chains into socially constructed CSC

models.

We examined resulting clusters, focusing on solutions with three to 10

clusters. The clustered chains constituted “implication matrices” from which we could

develop network models. After producing rough models for most of the cluster

solutions, we settled on a five cluster solution that seemed to be coherent and

meaningful. We mapped each cluster into a network model where nodes represent the

constructs and links represent the links connecting the constructs in the chains. The

size of each node and the number in the node represent the number of chains in which

the concept was mentioned. The thickness of the links indicates the number of chain

links it represents. The models were tidied up by eliminating redundant links, i.e., link

A—C if links A—B and B—C exist. The resulting CSC network models are shown in

figures 2—6.

10
Figure 2. CSC map 1, Digia Inc.

Figure 3. CSC map 2, Digia Inc.

11
Figure 4. CSC map 3, Digia Inc.

Figure 5. CSC map 4, Digia, Inc.

12
Figure 6. CSC map 5, Digia, Inc.
Pre-workshop preparation

We wanted to provide more value for our clients, in terms of rich information

to support project selection and development, than was provided by the basic CSC

method. For this, we turned to two concepts developed by Timmers (1999). We chose

to use a network business model structure described by Timmers (1999) that

concentrates on the processes and interactions of various players. This modeling

method provides a graphical representation of the flows of information, value and

revenue among the various parties producing, delivering and using the applications.

We considered this important because as the mobile commerce value chain is an

important perspective from which to evaluate the business possibilities of an

innovative application.

To aid in the development of the network business model, we developed

“interaction models for each of the application ideas as suggested by Timmers (1999).

13
This model helps us to analyze the interactions among parties to the business model.

We expected this to help us to develop finished network business models from the

rough models drawn in the workshops.

Ideation workshop

The ideation workshop was held at Digia offices at the beginning of April

2001. After discussion with the analyst, the firm chose the participants to include both

business and technical R&D people. Six people participated from the firm, including

the chairman of the board, the Nokia Key Account Director, two business

development managers and two engineering managers.

The workshop was scheduled for five hours on a single day, with a fancy lunch

provided to participants to enhance enthusiasm for the tasks. It began with a 15-

minute introduction to the CSC method and the study and a discussion of the purpose

of the meeting. Participants were instructed that the objective was to examine and

discuss each CSC map as a group, then come up with a project idea that would

address the desired consequences and values expressed in the models, including a

name, short description, architecture, a list of supply chain players, our customer

segment(s), benefits for players and customers, our profit model, and risks involved,

all at a “back-of-the-envelope’ level of detail. In addition, they were to produce a

rough graphical network business model that showed the relationships among

customers and supply chain participants, including the flow of information, value and

revenue. Asked if they could call for information from colleagues in the building, they

were told that they were “on [their] own and should use teamwork and their own

expertise to solve problems.”

Beginning with CSC map 1 (figure 2), the analyst explained the map and

participants began a very animated discussion, with every member participating and

14
participants taking turns leading discussion at the flip chart, taking notes, and writing

out and drawing models of the project. The researcher’s role was to direct the flow of

the process and give it limits if needed. During the discussion there were several

questions about the individual statements behind the CSC map labels. Discussion of

the first model took 2.5 hours, but after that participants seemed to have learned the

process and discussions progressed more quickly. The working model for the

workshop turned out to be very feasible. The participants actually drew business

models on the flap board without the researcher asking for it. Example can be seen in

Figure 7. In addition, they did not emphasize the technical architectures too much and

they were discussed on a very broad level for most of the maps.

Figure 7. The business model drawn in the workshop from CSC map 2.

Participants responded well to all of the CSC maps. Workshop participants

were able to develop a business idea for each of the maps, except that for map 3

(figure 4), there was some hesitation. Then, after a short discussion, someone in the

group said, “there is no game for us here. Let’s move on.” Digia wouldn’t be a part of

any value chain developed from this model, so there was no point in pursuing it,

although it might have been interesting to another firm, e.g., in the banking industry.

15
Group members decided that maps 4 and 5 could be satisfied with a single new

business idea and treated them as different perspectives of the same model.

In general, the atmosphere during the session was very enthusiastic and the

participants highly motivated. Several expressed interest in continuing the process

after the study. Said the Digia CEO, “this really positively over exceeded my

expectations of the results we could gain from the workshop. This is just what we

would need for our product development council.”

Post-Workshop Analysis

After the workshop we began assembling the interaction models and the

redrawing the business models according to our notes and the material provided by

the workshop. The objective was to structure the results to make them more clear to

potential readers who hadn’t been at the workshop. We intentionally tried not to

influence the results themselves and made only slight modifications to the business

models produced in the workshop to include implied, but omitted items. This process

was very easy and straightforward. Assembling the interaction models of the

applications really helped to understand the processes behind the rough drawn

business models and helped to clarify them. In below are the results of the post-

workshop analysis. These were presented to the Digia’s CEO at the end of April 2001.

16
Table 3. ‘Back-of-the-envelope’ description of the Personal Transaction Assistant, an application
idea developed from CSC map 1 (figure 2) at the Digia ideation workshop.

PERSONAL TRANSACTION ASSISTANT

Description Self-controlled and profiled access for permission marketing. (EPOC
client software for PDAs, communicators and smart phones)
Architecture Security - SSL, IPsec (VPN); Synchronising of data - Similar to SyncML;
Authentication of user - SmartTrust and others, finger print and eye's iris
identification systems; Transaction middleware - similar to Mbroker by
[Link]; Client/Server environment
Supply chain players Bank, Network operator, shop, customer, fulfilment service, Infomediary
(Info broker), Government (Population Register Centre)
Our customers for this Manufacturers, operators, Portals, E-auctions
system
Our profit model License fees per device
Benefits for players and Customers: thrift, individualism, paperless receipt, reverse-markets
customer's benefits.
Operators: increasing network traffic, security services (authentication
solutions).
Shop: could concentrate on one product leading to efficiency benefits,
better circulation speed, lower risk.
Fulfilment services: increased turnover.
Infomediaries: business opportunity, fees from participating companies
and transactions
Risks Getting the players together. How the concept can be implemented with
big players.
Miscellaneous Could the device manufacturer be the pusher of this concept?

PERSONAL TRANSACTION ASSISTANT

The first application idea was named Personal Transaction Assistant and the

‘back-of-the-envelope’ idea resulting from the workshop is shown in table 3. The

application is a Symbian OS based client software for PDAs, communicators and

smart phones that control the information flow between infomediaries and customer.

It is self-controlled and creates a profile that is offered to infomediaries according to

the reverse-market theory (Hagel and Armstrong 1997). The application would also

take care of the negotiation between the consumer and infomediaries. The architecture

for this application was discussed on a broad level and it was considered that security,

synchronizing of data, authentication of users and transaction middleware are in key

positions if a decision is made to continue the product development process with this

idea. Our analysis of the workshop discussions and the rough network business model

17
resulted in the interaction model, shown in Table 4. The end-consumer authentication

process between shop, bank and trusted third party was perceived as particularly

important. The resulting network business model is presented in figure 8 and it

demonstrates the interactions between the players and how revenue streams were

expected to be generated.

Table 4 Interaction model for Personal Transaction Assistant

V{(a),(b)} Interaction(s) Involved processes
Profile updating, Product requests and buying,
Customer : Infomediary 1:1, 1:Many
Agent negotiation
Customer : Shop 1:1 Customer service, sales
Payments, Receipts and Banking services in
Customer : Bank 1:1
general
1:1,1:Many Information brokering; agent negotiations, profile
Infomediary : Customer Many:1 updating, benefits offered to 'profiles', trading of
Many:Many transactions
Infomediary : Shop 1:1 Information brokering, fees to shops
Customer service, sales, delivery of receipts and
Shop : Customer 1:1
guarantee certificates
Shop : Fulfilment service 1:1 Logistics
Authentications and credit standing checkups,
Shop : Bank 1:1
service fees
Authentications and credit standing checkups,
Bank : Shop 1:1
payment middleware
Bank : Trusted 3rd party 1:1 Authentication process
Bank : Customer 1:1 Bank services; receipts, payment middleware
Trusted 3rd party : Bank 1:1 Authentication process, service fees
Trusted 3rd party : Government 1:1 Authentication process
Fulfillment service : Customer 1:1 Authentication process, service fees (if any)

The report to the CEO also discussed the possible benefits and risks for the

players. Benefits for the customers would be individualism and thrift. In addition, the

convenience benefits sought from electronic receipts and reverse-market benefits

should be taken into account. The network operators would participate in the value

chain by enabling data traffic between customers and infomediaries, this providing the

infrastructure and authentication services in form of a trusted third party. The main

benefits to them would be increased network traffic and revenues from trusted third

party services. Network operators might be interested in becoming an infomediaries

themselves to better control the value chain and gain more

18
Personal Transaction Assistant
Shipping of product Service
Fulfilment fee
se rvice Send!

Mobile Payment Customer

Custom er se
rvice & sa les
Shop

Product request

B uy
r
Prof ile updating

!
de
Or

Re ceipt
credit standing check up
Authentication and
M obile Network operator

Sales revenues
Information
Infomediary supply

- Information
Participation / transaction
brokering fees

Authentication
Trusted 3rd party*2 process

= information flow ? Bank

= revenue flow
= logistic flow
= negotiation process between agents Government
*1 *1 Population Register Centre
= benefits/discounts from shops *2 Owned by Mobile Network Operator (CSC M ap 6)

Figure 8. The business model of Personal Transaction Assistant.

revenue. For the retailer expected benefits would come from economies of scale and

additional efficiency that result from concentrating on fewer products. In addition, the

risks for the retailer might decrease, as they would no longer be required to keep

physical inventory. The biggest risks were thought to be in the co-operation between

the different value chain participants. Could one of the leading device manufacturers

be a ‘champion’ for this concept and pull all them together?

MY FINANCIAL ADVISOR

From CSC map 2 participants proposed the My Financial Advisor (MFA)

applications. MFA is a combination of applications that create a trouble-free and

intelligent way of receiving financial information in real time. It would be a portfolio

of analyzed, customized, and raw data. An analyst’s role could include a portfolio

manager, who refines the data manually and creates recommendations for customers.

19
Value accrues to the end-user in the form of valuable information, available anytime,

anywhere and to the analyst and infomediary in the form of fee revenue.

Table 5. ‘Back-of-the-envelope’ description of the My Financial Advisor, an application idea

developed from CSC map 2 (figure 3) at the Digia ideation workshop.

PERSONAL TRANSACTION ASSISTANT

Name My Financial Advisor
An easy and intelligent way of receiving real time financial information.
Description
An 'aggressive' screensaver that pushes information.
Always on connection (GPRS, i-MODE etc.), Multiple user interfaces /
access methods, speak synthesizer, Browser, storing of information,
Architecture
chronological order of information (Archiving the information),
structuring of the information, Lotus Domino client/server environment
Players Stock Exchange, Customer, Infomediary, Analyst, Network Operator
Customers for the developer Analysts for developer, For Analysts day traders
End-customers - Real time information, better decisions, Stock Exchange
Benefits for players and - more volume for trading, Infomediary - Revenues from Analyst,
customer's) Network Operator - more network traffic, Analyst - increased revenues,
better customer relationship
Revenues and market size 2x People travelling in business class, Nokia Communicator users
Profit model for the
License sales to analyst(s)
developer

Table 5 presents the results of the workshop and the business model is

described in figure 9. The model includes an infomediary that acts as an information

broker, but the emphasis is on the analyst role that refines the data into information

and distributes it to end-customers. The analyst role could include a ‘portfolio

managers’ portion that processes the data manually and creates recommendations that

are then distributed to the customers. The other two participants in the value chain

were considered stock exchanges and various companies that provide company

specific information to the market place. Also in this application group, the operator

role was regarded to be in the background providing only the infrastructure for the

data communication between the analyst and customers.

For this particular application idea there was a lively discussion about the

architecture. One of the most important things would be the ‘always on’ connection

type that GPRS and i-MODE could provide. The user interface raised questions if any

text-based push is adequate enough for this purpose and the need for speak

20
synthesizer was expressed. This way the application could speak out the important

news and alert the user. This feature was regarded as interesting because according to

the participants, people can usually filter better information they hear than what they

read. In addition, the application should have various user interfaces according to the

devices used and the multi channel concept was presented often. The storing of

information was considered problematic. One opinion was that the information should

be stored in chronological order. A shared view was that the application group could

be developed with Lotus Domino environment.

The customer segment for the developer would be analysts and the profit

model could be a license fee per device. The participants had a very clear idea of the

market size for this application. They estimated that it would be twice the number of

people traveling in business class, plus the current Nokia Communicator users. It was

estimated that it would take only two weeks to develop such a Symbian OS based

application.

Perhaps the biggest risk for this set of applications would come from the firm’s

lack of knowledge of the financial sector and the financial institutions’ attitude against

‘outsiders.’ They could develop own applications to keep the value chain better in

their own control. End-customer needs were also considered problematic.

21
 My Financial Advisor

Inf omediary Stock Company

Exchange
In Fe
fo
rm e s
at
io
n
br
o ke
r in
g

Analyst Portfolio
Mass financial manager
information

M obile Network
operator
Customer

= information flow
= revenue flow
= logistic flow
(CSC M ap 7-8)

Figure 9. The Business model of My Financial Advisor.

MOBILE WALLET

Looking at CSC maps 4 and 5 (figures 5 & 6), the participants decided that they were

looking at similar models, perhaps from different perspectives, and decided to treat

them as if they were one model. The discussion resulted in the idea for the Mobile

Wallet, described in table 6. The mobile wallet allows customers to make payments

for small transactions, i.e., 0.1 to 50 Euro, anonymously and for medium sized

transactions, i.e., up to about 500 Euro, from a wireless device using credit or debit

card accounts. For the customer, value accrues as convenience and flexibility while

mobile. For other parties, value accrues in the form of reduced costs or increased fee

revenue.

22
Table 6. ‘Back-of-the-envelope’ description of the Mobile Wallet, an application idea developed
from CSC maps 4 and 5 (figures 5 and 6) at the Digia ideation workshop.

MOBILE WALLET
Description The means to conduct small amount transactions (FIM 1-10) with your
mobile device. Anonymity of money up to transactions of FIM 300
Architecture Security, Authentication, Synchronisation/replication, Database
(SIM/Smart Card), Profile, User interface
Supply chain players Bank, Shop, Customer, Loyalty card providers, Trusted 3rd party
(SmartTrust), Device manufacturers, Credit card issuers
Our customers for this Banks, Shop, Credit card issuers
system
Our profit model Free application, License fees from various plug-ins for different players

Benefits for players and End users: no coins, security aspect.

customer's) Shop: more difficult to rob and lower fixed costs.
Bank: transaction fees, avoidance of costs from plastic cards
Risks Lack of demand, reaching the critical mass, lack of standards, Microsoft,
customer resource management (CRM)

The business model for this group of applications resembles the that presented

earlier for the Personal Transaction Assistant. The authentication process is similar

and as are the parties involved in this process. Additional players are issuers of loyalty

cards and credit cards. The network business model is presented figure 10. The

network operator’s role in this model raised some questions. The operator is presented

to be in the middle between customer, card issuers and shop. The operator could also

be left out of some of the transactions, as in the case of Bluetooth technology, for

example, which does not need an operator. However, information from card issuers

probably would go through the operator.

The architecture of the application group resembles the one described for the

first Personal Transaction Assistant, with the addition of SIM or smart cards. These

could be used as storage space for anonymous money and identification information.

The identification information could contain points similar to those in passports and

key cards include currently. It should be noted that also with this application idea the

profile issue was dealt and the participants emphasized that it would be important for

users to have various profiles that suite their current needs according to place and

23
time. For the researcher it was fascinating that the workshop raised the same issues for

loyalty cards as the interviewees did previously and they were considered essential to

the concept.

The benefits to customers follow their performance goals. The biggest one was

the desire to avoid the use of coins that was expressed in both maps. For the retailer it

could be possible to reduce transaction costs, as the cost of processing currency and

the risks associated with handling currency are reduced. Likewise banks could

decrease their currency related costs and expenditures coming from issuing various

plastic cards to their customers.

The profit model is clearly different from the two previous ones. The

workshop participants expected that the client software for the consumer appliances

might be distributed free-of-charge and that license fees would be charged for a

variety of plug-in software that the other value chain participants need to contact the

Mobile Wallet application. For example, POS software developers would only have to

add the plug-in to their software and they would be able transfer data between the

wallet and their system.

Risks described for this application group included a lack of demand and

standards that could lead to difficulties in reaching the critical mass for profitability.

Customer resource management (CRM) issues were also raised. One of the

participants noted that supply chain participants would worry about whether big

software vendors like Microsoft would launch their own version of this product soon.

24
 Mobile Wallet
Service
Cre dit card fee
issue r

Shop
Customer Anonym mo bile payment

Debit/credit payment
Identification services

M obile Network operator

Re ceipt
Loyalty

credit standing check up

Authentication and
Card

Sales revenues
issuers

Authentication
Trusted 3rd party*2 process
= information flow
= revenue flow ? Bank
= logistic flow
= Loyalty card benefits
Government
*1 Population Register Centre *1
*2 Owned by Mobile Network Operator (CSC M aps 1-5 and 3-9-10)
Figure 10. The Business model of Mobile Wallet.

Digia representatives were very enthusiastic about the results of the workshop.

After the analyst presented the results of the workshop Digia’s Chairman remarked

that the workshop “positively…exceeded [his] expectations [about] the results…”

Digia plans to continue using the CSC method in IS planning. According to Digia

product analyst, Markus Ahonen, the firm regards CSC favorably because (1) the

method seems to work well and it is easy to see why it works, (2) the interviewer

collecting data for the method needs no special abilities or skills, and (3) the resulting

CSC maps are very helpful in understanding how people think about an issue.

CONCLUSIONS
At Digia, the CSC method, extended with network business models, helped

managers and technical experts to develop ideas for feasible application products for

25
the firm. These were developed at a ‘back-of-the-envelop’ level, then polished by the

analysts for presentation to the Digia CEO. Digia is currently developing products

related to these ideas for roll out for next generation mobile devices.

The CSC maps provided workshop participants with rich information about

features that the experts and sophisticated users wanted and the reasons why they

wanted them. This rich information helped workshop participants to develop

application ideas that, in turn, contained very rich information for developers,

including not just the desired application functionality, but information about the

interaction among business partners and customers, the flow of information and

revenue, architecture, and other issues. This rich information is expected to make it

more likely that developers can successfully develop the applications successfully.

Implications for managers

This extended CSC method should be helpful to planning for IT in a variety of

circumstances. We used experts and sophisticated end-users as our participant group.

Other groups could be used in different circumstances. For example, where a firm

wanted to plan infrastructure IS investments, a participant group that included

representative employees, plus supplier representatives, might be employed. For

systems involving integration with vendors participation of incoming logistics

employees, representatives from operations, as well as representatives from supplier

firms might be a good idea.

The extended CSC method is sufficiently economical so that participation

doesn’t have to be limited to just a few people. The marginal cost of adding an

additional participant was approximately the cost of the individual interview, about an

hour. At Digia, the interviewing process was spread out over about five weeks, but

that resulted from its scheduling during the holiday season. Ordinarily it might be

26
feasible to interview 30 participants in a little more than two weeks. The analysis

should normally take another week or two, then another two weeks should be allowed

for the workshop and post-workshop analysis. During most of this period, however,

only the two analysts are fully engaged.

The analysis for CSC is sophisticated enough so that it generally requires the

services of a skilled social scientist, at least for the first time. The interviewing is quite

simple, however, and can be done by an employee with interviewing skills after a

short period of coached practice.

27
 REFERENCES

Bickers, C., “Mixed signals on 3G phones,” Far Eastern Economic Review, Hong Kong, 164:38,
September 27, 2001, 36-37.
Boland, R.J. Jr.; Tenkasi, R.V.; and Te'eni, D. Designing information technology to support distributed
cognition. Organization Science, 5, 3 (1994), 456-475.
Broersma, M., “UK loses interest in wireless net,” ZDNET UK, 2002, accessed at
[Link] on April 19, 2002.
Earl, M.J. Experiences in strategic information systems planning. MIS Quarterly, 7, 1 (1993), 1-24.
Cornell, M. Going around the bend: rethinking the strategy-development, process in technology
companies. Upside, 10, 10 (November 1997), 153-156.
Fitchard, K., “Metricom’s stumble raises 3G questions,” Telephony, August 13, 2001, p. 34.
Hadden, A., “Great expectations for 3G,” Telecommunications, 35:7, July 2001, 47-51.
Hagel John III and Armstrong Arthur G., Net Gain – expanding markets through virtual communities,
Harvard Business School Press, 1997
Latour, A., “A telcom shakeout is looming in Europe—New wireless technology is set to winnow the
industry,” Wall Street Journal, March 20, 2002.
Lockamy, A. III and Smith, W.I. A strategic alignment approach for effective business process
reengineering: linking strategy, processes and customers for competitive advantage. Int. J.
Production Economics, 50 (1997), 141-153.
Nelson, M.G., “Third generation wireless: still on hold,” Manhasset, May 7, 2001, 130-132.
Nokia, The Demand for Mobile Value Added Services, Study of Smart Messaging, Market Study, 1999,
[Link]
Peffers Ken and Gengler Chuck, "Identifying High-Impact IS Projects," forthcoming in the
Communications of the ACM, 2001
Peffers Ken and Gengler Chuck, "Understanding Internal IS Customer models of Firm Performance To
Identify Potential High-Impact Projects," 33rd Hawai'i International Conference on System
Sciences, Maui, Hawai'i, January, 2000
Powell, K., “To meet revenue and market penetration projections, operators need to drive customer
acceptance, says Cap Gemini Ernst & Young Country,” News & Events Telecom Media
Networks, 18 October 2001.
Premkumar, G. and King, W. R. Organizational characteristics and information systems planning: An
empirical study. Information Systems Research, 5, 2 (1994), 75-109.
Quotient Communications, Making 3G networks pay, January 2001.
Rockman, S., “A mobile failure waiting to happen,” The Guardian, February 8, 2001.
Sangani, K., “Mobile banking’s tough sell,” Financial World, March 2002, 30-32.
Segars, A.H. and Grover, V. Profiles of strategic information systems planning. Information Systems
Research, 10, 3 (1999), 199-232.
Shank, M.E.; Boynton, A.C.; and Zmud, R.W. Critical success factor analysis as a methodology for
MIS planning. MIS Quarterly, 9, 2 (1985), 121-129.
Tillquist, J. Institutional bridging: how conceptions of IT-enabled change shape the planning process.
Journal of Management Information Systems, 17, 2 (2000), 115-152.
Timmers Paul (foreword by Ravikota), Electric Commerce – Strategies and models for business-to-
business trading, Wiley, 1999

28
 BIOGRAPHIES

Ken Peffers, Ph.D. (Purdue, 1991) is an Associate Professor of MIS at the

Hong Kong University of Science and Technology. His current research focuses on
making the right IS investments for the firm. He is editor-in-chief of the IS journal,
JITTA, accessible at [Link]
Tuure Tuunanen, [Link]. (econ) (Helsinki, 2001), is pursuing a PhD degree in
information systems science at the Helsinki School of Economics. His doctoral
dissertation focuses on evaluating and comparing various qualitative research methods
to support idea generation and software development in multi-channel
environment such as 3G mobile terminals and Digital TVs.

29