Peter Weill.

Information technology infrastructure for E-business online

. (page 1 of 3)
Font size
QR-code for this ebook

MIT LIBRARIES

'III II' ""l'f!'l"l|l' "'III



3 9080 02246 1



823



28 ^ \ (DEWEY

L4
\i

i



14 ^ [k\ i



Information Technology Infrastructure
for E-Business

Peter Weill and
Michael Vitale

February 2001

CISRWPNo. 313

Sloan WP No. 4157 ^



©2001 Massachusetts InstiUite of Technology. All rights reserved.



Center for Information Systems Research

Sloan School of Management

Massachusetts Institute of Technology

Three Cambridge Center, NE20-336

Cambridge, MA 02142



WIASbAUHUSt ITS INSTITUTE
OF TECHNOLOGY



APR 1 8 2001



LIBRARIES



^ns '] I



CISR Working Paper No. 313



Title: Information Technology Infrastructure for e-Business

Author: Peter Weill and Michael Vitale

Date: January 2001

Abstract: As firms migrate from their traditional physical business models to a combination
of physical and e-business models, the importance of information, particularly customer
information, raises the stakes for the management of the firm's information technology
infi"astructure. The number of opportunities for investing in IT infi^astructure, however, can easily
overwhelm senior management teams. To guide IT investment decisions this paper examines e-
business IT infi-astructure requirements. Based on a study of 50 e-business initiatives, the
researchers classified e-business initiatives into 8 different atomic e-business models and then
examined the need for each of 70 IT infrastructure services by each atomic model. The results
indicate that each type of e-business model demands a different set of five or six key
infi-astructure services. This finding should help executives identify IT investment priorities for
their e-business strategies. Recommendations are made for senior management to review their
firm's IT governance to ensure IT infi-astructure needs for e-business are idenUfied early in the
strategizing process.



22 Pages



This paper is also available as Sloan Working Paper No. 4157.



Information Technology Infrastructure for E-Business
Dot com, dot go: capitalizing on the E-Business Revolution

The rise and fall of the dot com has been a wake-up call to traditional businesses that are thinking
about the power of their business models in an on-line world. Much of the focus of the media,
academia, and Wall Street has been on the radical new business models being simultaneously
developed and market tested by highly publicized start-up firms in this first wave of e-business.
However, existing businesses will do much of the really hard work and make most of the profits in
the second wave of e-business. E-business will require existing businesses to change — but not
abandon everything that has made them successfiil. Traditional businesses must migrate fi-om the
physical to the e-business world. This migration includes making challenging leadership decisions
about which e-business models will succeed and how they will integrate with current ways of
serving the customer. Many of the assets (for example brand, cash, relationships, market share) of
successfiil place-based businesses will serve equally well in e-business, but some liabilities are
painfully apparent, including the lack of e-sawy leaders, corporate cultures resistant to change,
reward systems incompatible with e-business, potential channel conflict, and non e-compatible
information technology infi-astructures.

We expect the second wave of e-business will have four important characteristics:

1. The difficulty of building and sustaining a profitable dot com business will be reflected in
more realistic (i.e., lower) stock market valuations. We expect to see one or two successful
dot coms in each major business sector, and the rest struggle to survive.

2. Existing firms will evolve to e-business models combining the best of the physical and
electronic worlds. Existing firms will seriously challenge dot coms by providing higher-
quality services more cheaply that are integrated with their place-based channels.

3. The artificial distinctions between B2B, B2C, and C2C will disappear. Many viable e-
business models will serve a combination of business and household customers,
representing different customer segments requiring different value propositions.

4. To compete in this e-business world, traditional firms will need to continue to invest
heavily in IT infrastructure, which is the foundation of e-business.

Information Technology Infrastructure Capability

hiformation technology infrastructure is used in all e-business initiatives to connect different parts
of the firm and link to suppliers, customers, and allies, kiformation technology infrastructure
investments made by firms will be as critical for creating long-term shareholder value as the
previous waves of physical infrastructure investments in property, plant, and equipment. In this
section we first define information technology infrastructure and then present the findings of our
recent study of 50 e-business initiatives and their infrastructure requirements.



This paper draws heavily on Place to Space: Migrating to e-Business Models by Peter Weill and
Michael Vitale, to be published in April/May by Harvard Business School Press .



Our list of the 70 information technology infrastructure services needed for e-business is presented
in the Appendix.

We define a firm's information technology portfolio as its total investment in computing and
communications technology' (see Figure 1). The IT portfolio thus includes hardware, software,
telecommunications, electronically stored data, devices to collect and represent that data, and the
people who provide IT services. The portfolio includes both information technology capability
provided by internal groups ("insourced") and those outsourced to suppliers such as IBM Global
Services and EDS.

The foundation of the information technology portfolio is the firm's longer term information
technology infrastructure, which in turn is linked to external industry-based infrastructures such as
bank payments systems, airline reservations systems, and automotive industry supply chain
networks, and to public infrastructures such as the Internet and telecommunications networks. The
combination of the internal and external information technology infrastructures make up the firm's
information technology infrastructure. The various elements of intemal information technology
infrastructure are presented in Figure 1 , with the remaining parts of the portfolio referred to as
"Local Applications." At the base of this framework are the technology components such as
computers, printers, database software packages, operating systems, and scanners. These devices
are commodities and readily available in the market place. The second layer is comprised of a set
of shared information technology services. The technology components are converted into usefiil
shared services by a human information technology infrastructure composed of knowledge, skills,
standards, and experience. This human infrastructure binds the technology components into
reliable services that form the firm's information technology infrastructure.

The services notion of information technology infrastructure is very powerful. The concept
emerged from our discussions with business managers grappling with what they were actually
getting for their information technology investments. Business managers told us they have great
difficulty valuing technology components such as a server or a database package. New information
technology staff appointments are also difficult to value. However, business managers can more
readily value a service, such as the provision of a fully maintained personal computer with access
to all of the firm's systems and the Internet. Such services can be specified, measured, and
confrolled in a service level agreement. Perhaps most importantly, managers can price services in
the market place for comparison. Thinking of infrastructure as services places the intemal
consumer - the business manager - in charge, rather than the provider, whether the provider is the
information systems group or an outsourcer. The service notion also provides more certainty to the
provider as to their responsibilities, and allows for more precise planning.



Figure 1: The Structure of IT Infrastructure



The base foundation of budgeted-for IT capability (botti technical and human), shared throughout
the firm as reliable services, and centrally coordinated.



Local
Applications




Fast changing local business applications such as:
insurance claim processing, bank loan applications,
customer complaints support system, phone order
support systems.



Shared and standard applications which change less
regularly such as accounting, budgeting, human resource
management.



Services which are stable over time such as management of
shared customer databases, PC/LAN access, intranet

Human infrastnx^ture of knowledge, skills, polk^es,
standards and experience binds components.

Commodities such as: computers, printers, routers,
database software, operating systems, credit card swipeis.



Source: Figure 4-1 P. Weill & M. Broadbenl 'Leveraging the New Infrastructure: How Marttet Leaders Capitalize on IT , Hanard
Business Sctxwl Press, June 1998,



The infrastructure services within a firm often include telecommunication network services,
management and provision of large-scale computing (such as servers or mainframes), management
of shared customer databases, research and development expertise aimed at identifying the
usefiilness of emerging technologies to the business, and a firm-wide infranet. An increasing
number of firms have an additional layer of shared and standard infrastructure applications used by
all business units. These often include firm-wide applications that support shared services in areas
such as accounting, human resource management, and budgeting.



Historically, the set of infrastructure services required by a firm is relatively stable over time.
Similar services are generally required from year to year, with gradual improvements in the
componentry occurring over time to take advantage of new technologies and efficiencies. From
time to time new services are required to support a new initiative. However, occasionally a major
disruption occurs requiring a quantum leap in services required. Our research shows that e-business
introduced the need for significantly more and different infrastructure services for most firms. It is
likely that after a number of new services are created for e-business, the set of infrastructure
services will again plateau, hi contrast, the information technology required for business
processes — ^particularly e-business applications — changes continually, often on a quarterly or even
monthly basis as business processes are altered to better suit customer needs or in response to



competitor activity. In many firms, website functionality is increased every month. To understand
how IT infrastructure needs may have changed as a result of e-business, we conducted a study of
traditional firms launching e-business initiatives.



Figure 2: Information Technology Infrastructure for e-Business



Shared /
Centrally Coordinated




Corporate Infra



Business
Unit1



BU 1 Infra



Business
Unit 2



LOCAL
IT

BU 2 Infra



Order Processing
Knowledge Management
Financial Management



G

R

A

V

I

T

Y



PUBLIC INFRASTRUCTURE (eg. Internet, Vendors, Telco's, Industry Networl(s)



Intranet
Home page
PC/LAN service
- B Electronic mail

Large scale processing
Customer database
Telecommunications
network

service providers
Automotive industry supply
exctianges



Adapted from P. Weill & M. Broadbent "Leveraging the New Infrastructure:

How Market Leaders Capitalize on IT" , Harvard Business School Press, June 1998, pp 37.



The time required to implement a new e-business initiative will depend in part on the firm's
insourced and outsourced infrastructure capability. For example, in building a new web-based
housing loan system, a large bank needed to use the following informafion technology
infi-astructure services: mainframe and server processing, customer databases, security procedures
and systems, and both local area and national communications networks. Having those
infrastructure services already in place significantly reduced the time and cost to build the loan
system. However, the newly built system could not be released immediately, because firewall
security services were not in place to support a web-based application that integrated customer data
and credit scoring systems with the Internet customer interface. Direct customer access was not
considered in the initial design of the infi-astructure services. The firm had to postpone launching
the new inifiafive until the security services were ready. If well designed, the new security
infrastructure services will be re-used for many other web applications.



Where to place the IT infrastructure capabihty (e.g., firm-wide or in a business unit) is a strategic
choice made by senior management (see Figure 2). For example. E-business often involves a single
electronic point of contact to the firm by a customer. The firm's information technology
infrastructure must then integrate information from separate business units so that the customer can
obtain the desired business service from the chosen point of contact (see point B on Figure 2).
Firms taking flill advantage of a customer's transaction with one part of the business will attempt
to cross-sell products and services from other parts of the business. Alternatively, the single point
of contact could be made for only one business unit at point A in Figure 2, limiting the cross
selling and information gathering possibilities to that business unit only. This positioning within
the firm is a sfrategic issue for senior management.

frifi-astructure capability is difficult to create because it is a fiision of technology and human
assets." These capabilities have long lead times to emulate and can provide a source of competitive
advantage. We know from our research that firms with greater infrastructure capability have faster
times to market, higher growth rates, and more sales from new products, but lower short term
profitabihty.'" Building an infrastructure tailored to a firm's sfrategic context takes considerable
time and expertise. While the components are commodities, the management processes used to
implement the best mix of infrastructure capabilities to suit a specific firm are a much scarcer
resource.



Demands of E-Bu^iness on IT Infrastructure

Before gathering data from firms, we anticipated that e-business would bring about four types of
change with in a firms' IT infrastructure: more capability, gravity, external, and cooperative'".

More capability: To compete successfiiUy in an e-busmess world, IT infrastructure will continue to
become more important, resulting in both a greater number of services and increased spending on
these services over time.

Gravity: There will be a general frend towards providing infrastructure services on a firm-wide
basis rather than at the business unit or departmental level. For reasons both of cost savings (from
increased scale and reduced duplication) and of implementing sfrategic initiatives (e.g. single point
of customer contact in a multi-business unit firms), infrastructure will experience gravity and drop
from point A to point B on Figure 2. For example, Johnson and Johnson has recently announced all
n inJB-astructure previously provided independently by more than 175 business units will be
cenfralized and managed on a firm-wide basis.'

Outsourcing: The rapid implementation demands of e-business will drive an increase in IT
outsourcing, particularly for commodity and highly specialized services. As IT infiastructure
services become better understood, many commodity services will be outsourced. Furthermore, as
application service providers (ASPs) prosper and determine their pricing models, more
infrastructure applications (e.g. shared and standard business processes and applications such as



' Conversation with Gregory J Poorten, Director, Product Lifecycle Management, Johnson & Johnson Networking
& Computing Services, January 200 1 .



infrastructure applications (e.g. shared and standard business processes and applications such as
HR, purchasing, and accounts payable) will be outsourced. We expect firms to insource the IT
infrastructure that is important for their competitive advantage or core competencies (e.g.
workflow, infranets, knowledge management) and outsource or at least market test commodity
services.

Co-ooperative: Given the attractive economies and potential strategic confrol the owners of a
single, dominant infrastructure have, we expect to see competitors and allies within industries share
IT infrastructures. For example, automotive manufacturers are cooperating to compete with the
announcement of joint ventures in managing supply chains. Ford, GM and DaimlerChrysler have
combined forces to form a business-to-business integrated supplier exchange that will be the
world's largest Internet-based virtual marketplace.^' The new enterprise will offer open
participation to auto manufact\irers around the world and to their suppliers and dealers as well. The
three partners will have equal ownership in the business, which will operate independently from its
parents. The President and Chief Operating Officer of General Motors, G. Richard Wagoner Jr.,
commented on the industry's efforts to build independent exchanges and the decision to work
cooperatively in order to compete more effectively. "As we continued to build our separate
exchange sites, we quickly realized fraditional, individual, stand-alone models weren't the winning
sfrategy for us, our industry, our suppliers and, ultimately our customers. By joining together we
can fiirther increase the pace of implementation. . . We are excited about the opportunity to build on
what each of us started separately and create the best frading exchange in the world."

Study of Information Technology Infrastructure Needed for E-business

The purpose of our study was to identify and define the IT Infrastructure services relevant to
particular atomic e-business models and their combinations, and to identify frends in infrastiucture
provision. We sought first to identify and categorize all the information technology services
relevant for e-business, and then to identify statistically which services were more important for
each e-business model.

We spoke to CIO's, senior IT infrastructure managers, information technology managers, IS
planning managers, and E-commerce managers. We studied 50 e-business initiatives in a diverse
range of Asia-Pacific subsidiaries of global firms, Ausfralian-based global firms, and government
agencies, including manufacturing, natural resources, financial services, postal, retail, and
agricultural resource companies.""' The Appendix describes the study methodology in more detail.

Beginning with a list of 25 information technology infrastructure services from a 1997 study of IT
infrastructure, we expanded the services needed to include those we believed, on the basis of
discussions with several firms, were required for e-business."'"' We then piloted and iteratively
amended the list by gathering data from companies in questionnaire and interview format until a
stable set of services emerged. The result was a list of 70 infrastructure services, categorized into
the nine areas listed below. The figures in brackets are the number of services in each area. The
Appendix provides detailed definitions of each infrastructure service area and a complete list of the
70 information technology infrastructure services with the percentage of firms providing each
service and their relative investment level.



6



Areas of information technology infrastructure services

We grouped the 70 services into 9 areas:

Applications Management ( 1 3 services)

Communications Management (7)

Data Management (6)

IT Management (9)

Security (4)

Architecture and Standards (20)

Channel Management (7)

IT Ri&D (2)

Education (2)



Figure 3: IT Infrastructure Services for e-Buslness


^^^^^^^^^^H IS Sen/ice Offered Firm Wide Relative 11
^^^^^^^^I^^^^^^^BB Provided? II

1 Areas (Number Of Services) %have %fw %bu If^'JX!: '^ '"'"^ '"^o™ °''°^^ '''^^^ °''^°'^" ''™^pUcatioiwlnfrestnictura(13)


90


71


18


3.7


57


:s


20


23


1


66


10


2. Communications (7)


81


71


11


3.7


26


45


30


26


2


67


5


3. Data Management (6)


78


63


15


4.8


54


21


25


17


1


TT


5




91


80


11


3.8


50


23


26


14


1


84


1


5. SMUrity(4)


100


93


7


6.1


51


8


41


15





85





6. Architecture and Standards (20)


95


91


4


2.7


79





21


7





93





7. Channel Management (7)


51


41


10


4.6


40


31


29


20


4


76





8. IT & Research and Development (2)


83


77


7


2.5


56


8


36


16





84





9. rr Education (2)


82


71


11


3.9


24


40


36


12





76


12





Figure 3 summarizes our findings on the extent of provision of these infi-astructure services. Each
row represents one of the areas of infrastructure service, with the number of services in
parentheses. The numbers in the body of the table are the average firm responses for all the
services within each area. The columns in Figure 3 are interpreted as follows:



The first column is the percentage of firms that provide the service. For example, there are six
infi-astructure services in the area of data management. On average, these services are provided by
78% of firms. The percentage of firms providing each service within each infi'astructure area varies
(see the Appendix). For example, 100% of firms managed key data independent of ^plications.

The second and third columns are the percentage of the firms studied that offer the services firm-
wide or only to certain business units. For example, in the data management area, 63% of firms
provide the services firm-wide, while 15% of firms provide them only to specific business units.
The remaining 22% of firms do not provide these services centrally.

The column headed "Relative Investment" is an average of the firms' relative investment in each
service area, hi the interviews we asked managers whether there would be an increase or decrease
in spending on each service in the next year (2001) to provide support to e-business initiatives. We
asked managers to indicate this on a scale between -10 and + 10. A +10 indicates that, relative to
other services, the percentage increase in investment in this service was the highest in the coming
year. Conversely a -10 indicates that the service would receive the greatest relative disinvestment
over the next year. A zero means no change in relative investment in that service.

The first of our anticipated effects (i.e., more capability) of e-business on IT infrastructure was
strongly supported in two ways. Firstly, the increase (fi-om 25 in 1997 to 70 in 2001) of the number
of independent infi-astructure services found in large firms indicates a significant increase in the
complexity and variety of services offered. Secondly, all services except EDI had positive relative
investment in 2001 compared with the previous year, indicating a blanket increase in spending on
infi"astructure.

The areas with the highest relative investment were security, followed by data management and
channel management. Channel management provides the electronic channel (e.g. Web site, call
centers, interactive voice response) to the customer or ally to support multiple applications.
Channel management is a new category of infi-astructure services developed as a result of e-
business. Security was universally identified as critical to give customers and senior management
alike, confidence in e-business activities. All four security services identified were provided in
100% of firms, and 93% of the firms provided them on a firm-wide basis.

The columns headed "Sourcing" present the percentage of firms providing the service areas using
insourcing (%IN) outsourcing (%OUT) and a combination (%BOTH). The columns headed
"Service used for" describes the percentage of firms using the service areas for: B2B e-business
(%B2B), B2C (%B2C), both B2B and B2C (%BOTH), and not for e-business (%NON-EC).

About half of the IT infi-astioictiire services in this study were sourced purely internally (48%), with
22% outsourced and 29% provisioned both internally and externally. This is a significant increase


1 3