Smart Computing Innovation Must Be Guided by Eas: Executive Summary
Smart Computing Innovation Must Be Guided by Eas: Executive Summary
Executiv e S ummary
Forrester predicts that the latest cycle of technology innovation, “Smart Computing,” will reshape the
technology market through 2017. Smart Computing — which involves a blend of smart devices, smart
networks, smart analytics and applications, and smart data centers — will introduce new complexities
for enterprise architects to address, such as optimizing assets and minimizing risks and liabilities on
the corporate balance sheet. Utilities, for example, have implemented Smart Computing in the form of
smart grids and smart meters. For companies to benefit from Smart Computing, enterprise architects
will need to take the lead in designing the Digital Business Architecture that will support this new
technology model.
· Digital Business Architecture. Forrester’s Digital Business Architecture will be the backbone of
successful Smart Computing deployment and implementation. The IT-to-BT transformation is
built on a foundation of Digital Business Architecture, where technology embodies the design of
the business. By building on an established Digital Business Architecture, enterprise architects can
create a road map that encompasses new Smart Computing technologies and the capabilities that
come with them.
· The 5 A’s: awareness, analysis, alternatives, actions, and auditability. Smart Computing
combines new and existing technologies for awareness, analysis, alternatives, actions, and
auditability. Of these, both analysis and auditability will require new metadata-based tools for
Headquarters
Forrester Research, Inc., 400 Technology Square, Cambridge, MA 02139 USA
Tel: +1 617.613.6000 • Fax: +1 617.613.5000 • www.forrester.com
Smart Computing Innovation Must Be Guided By EAs 2
For Enterprise Architecture Professionals
compliance tracking, which merge with current EA activities in metadata management. Future
advances in Smart Computing can make sensitive customer data more accessible to other
platforms, and EAs must grapple with the security and privacy concerns that will arise from
that innovation. Forrester already recommends that EAs take the lead in creating metadata
management strategies, and because of increasing Smart Computing adoption, they should
reinforce their investment.
Firms’ Balance Sheets Will Expand Beyond Traditional Assets And Liabilities
Smart Computing will open the doors to firms creating broader balance sheets that reflect new
enterprise values — including social, environmental, country-based values as well as the existing
financial metrics in a new balance sheet. Why should EA care? Enterprise architects are in the
unique position of describing the relationships between these values and metrics. They should
address the balance sheet aspects of Smart Computing with a well-thought-out plan that accounts
for the new metrics that will become important to the business.
R ec o mme n d ati o n s
Forrester Research, Inc. (Nasdaq: FORR) is an independent research company that provides pragmatic and forward-thinking advice to global leaders in business
and technology. Forrester works with professionals in 20 key roles at major companies providing proprietary research, customer insight, consulting, events, and
peer-to-peer executive programs. For more than 26 years, Forrester has been making IT, marketing, and technology industry leaders successful every day. For
more information, visit www.forrester.com.
© 2010, Forrester Research, Inc. All rights reserved. Unauthorized reproduction is strictly prohibited. Information is based on best available resources. Opinions
reflect judgment at the time and are subject to change. Forrester®, Technographics®, Forrester Wave, RoleView, TechRadar, and Total Economic Impact are
trademarks of Forrester Research, Inc. All other trademarks are the property of their respective companies. To purchase reprints of this document, please email
[email protected]. For additional information, go to www.forrester.com. 56726
December 4, 2009
December 4, 2009
Smart Computing Drives The New Era Of IT Growth
A New Tech Investment Cycle Holds Seismic Promise — And Challenges
by Andrew H. Bartels
with Ellen Daley, Andrew Parker, Boris Evelson, and Chétina Muteba
Executiv e S ummary
The technology industry has entered a new cycle of tech innovation and growth, which we are calling
“Smart Computing.” Like prior cycles of mainframe computing, personal computing, and network
computing, Smart Computing will power a seven- to eight-year period when business and government
investment in technology grows at twice the rate of the overall economy. Smart Computing will be more
complex than what came before — blending elements of hardware, software, and network technologies.
Similar to earlier cycles, Smart Computing will grow rapidly because it will help business solve problems
that it couldn’t address before; in this cycle, Smart Computing will help companies optimize process results
and the returns from their balance sheets. Unlike the horizontal technologies of personal computing and
network computing, Smart Computing will have a highly vertical industry focus. Vertical solutions will
differ significantly from vertical offerings in the past — thus the advent of verticals 3.0 as a result. Tech
vendors will have great growth opportunities in this new cycle, but also big challenges in navigating the
shift to Smart Computing.
Three Reasons Why The Tech Market Will Boom between 2010 and 2016
We believe that the tech market in the US — and by extension, other countries — is poised for a
multi-year run of strong growth and innovation. While the 2008 to 2009 financial crisis muddies
this prediction, there are three reasons why we expect that the coming tech boom has been delayed,
not cancelled altogether:
1. History says we are due for a fourth wave of tech innovation and growth. Three times since
the 1950s the US tech market has gone through a 16- to 20-year cycle of rapid and then slowing
growth. In each case, the introduction of a new technology spurs an eight- to 10-year period
when tech investment grows twice as fast as the economy, followed by a similar-length period
when tech investment grows at the same rate as the economy. The last wave of tech innovation
growth started in 1992 and was due to end in 2008. Based on historical patterns, the US should
experience another wave of tech innovation and growth, starting around 2008.
2. Tech purchases started to grow much faster in 2008 until the financial crisis hit. The US
Department of Commerce data on business investment in technology shows that US tech
investment (starting in Q4 2007 and through the first two quarters of 2008) averaged 8.4%,
almost twice the average growth rate for nominal GDP of 4.6%. While the financial crisis and
resulting recession killed that growth spurt, the relatively strong growth in IT investment in late
2007 and early 2008 suggests that there is a lot of pent-up demand for technology goods that
will resurface as soon as the recession passes.
We call this new generation of technology “Smart Computing.” Why? It adds to existing
technologies new capabilities of real-time situational awareness and automated analysis. As a result,
technology moves beyond just proposing task solutions — such as executing a sales order — to
sensing what is happening in the world around it, analyzing that new information for risks and
possibilities, presenting alternatives, and taking actions. Smart Computing is:
a new generation of integrated hardware, software, and network technologies that provide IT
systems with real-time awareness of the real world and advanced analytics to help people make
more intelligent decisions about alternatives and actions that will optimize business processes and
business balance sheet results.
While what we are calling “Smart Computing” is still a long way from the intelligence in Star Trek
computers, we still see it as a seismic shift in capabilities from the technology of 2000 to 2001.
However, Smart Computing is not a brand new creation; it is an evolution as well as an extension of
three generations of computing. Similar to how Isaac Newton spoke of his own accomplishments,
Smart Computing’s achievements will rest on the shoulders of the technologies that came before.
The Tech Market Moves In Cycles of Tech Growth And Tech Digestion
Our study of the history of information technology since the 1950s shows that there have been three
cycles of new technology introduction and adoption (see Figure 1). This historical treatment is
focused on the US where historic data on tech investment is deep and long, but the other industrial
countries show similar patterns.
Figure 1 The US Has Seen Three Cycles Of Tech Introduction And Digestion Since 1960
Commercial software
2%
0.88%
Computer equipment
1%
0.83%
0.34%
0% Communications equipment
56
70
76
85
92
00
08
*
16
19
19
19
19
19
20
20
20
Source: US Commerce Department for 1956 to 2008; Forrester Research forecasts for 2009 to 2016
*Forrester forecasts
55157 Source: Forrester Research, Inc.
1. A new form of information technology comes to market and gains adoption. The new
technology, which had been under development with experimental adoption in earlier years,
suddenly starts to gain rapid and wide adoption because of its ability to solve critical business
challenges. New competitive offerings come to market, helping reduce solution costs and
accelerate purchases by more firms. This period of growth and innovation generally lasts eight
to 10 years, with a shift toward shorter periods in more recent decades.
2. The ratio of IT investment to GDP (IT-to-GDP ratio) rises during this period. With
investment in this new technology growing more rapidly than the economy, the ratio of
business and government investment in this technology as a percentage of revenue goes up.
Businesses and governments buy the new technology, seeing it as a way to quickly and easily
transform operations. Tech investment on average grows twice as fast as the overall economy
grows, causing the IT-to-GDP ratio to rise for eight to 10 years.
3. After eight years or so, growth in new technology slows as companies focus on absorbing
it. Technology by itself does not change how business gets done; business processes and
organizations need to change to take advantage in practice of what technology can do in
principle. Businesses and governments thus slow the pace of new technology acquisition and
focus instead on the business process changes that are needed to achieve better business results.
Vendors, responding to this skepticism, shift focus from new products creation to making
products that are easier to use and implement. This tech digestion and refinement period lasts
six to eight years, with a shift toward longer periods in more recent decades.
4. During the tech digestion period, the IT ratio stays flat — or declines. As a result of buyer
cutbacks and vendor focus on cutting solution costs, the growth in technology investment slows
to about the same rate as growth in the economy, or even slower. The IT-to-GDP ratio then
becomes stable and, in recent periods, has declined.1
The US Has Seen Three Cycles Of Tech Introduction And Digestion From 1960 To 2008
This cycle of tech innovation and growth, followed by tech refinement and digestion, has occurred
three times since the 1950s, with three generations of computer technology:
occurred frequently — like government benefit payments and airline reservations. Computer
investment rose from nothing in 1959 to 0.26% of GDP in 1969, with growth also in software
purchases pushing the total IT-to-GDP ratio to 0 .93% in those 10 years (see Figure 2-1). Then,
from 1969 to 1976, the computer-investment-to-GDP ratio edged up and down before hitting
0.27% in 1976, and the total IT-to-GDP ratio inched up to 1.1%, mostly due to wider adoption of
mainframe software products.
· Personal computing from 1976 to 1992 moved computing to white collar workers. The
personal computer was the center of the next wave of tech expansion — starting in 1976 with
the Apple-1 personal computer and the Wang 1200 word processor. From 1976 to 1985, a
series of innovations in PCs occurred, including the first IBM PC with a Microsoft MS-DOS
operating system; the Osborne 1 portable PC; WordStar, VisiCalc, and Lotus 1-2-3 software for
PCs; and the Apple Macintosh. These hardware and software products gained rapid adoption,
with both contributing to the total IT-to-GDP ratio reaching 2.33% in 1985, double the 1.1%
ratio in 1976 (see Figure 2-2). Why? PCs came to market when the US economy was shifting
from a manufacturing to a services economy — with white collar office workers representing
a growing portion of the workforce. Companies needed personal computing to help workers
prepare reports; do analysis; and make sales, marketing, or strategy presentations. It was not
enough to put PCs in front of employees. Changes in business processes or organizations had
to occur as well, such as getting employees to type their own memos rather than dictating
them to secretaries. This transition took time, so after 1984, the pace of innovation in personal
computing slowed, with vendors concentrating on making word processing, spreadsheet, and
desktop publishing tools easier to use and computers cheaper to buy — and investment in IT
was basically flat from 1985 to 1992.
· Networked computing from 1992 to 2008 drove process automation. The third wave of
computing was launched with the arrival of SAP’s R/3 enterprise resource planning (ERP)
software in 1992.3 Coinciding with the business process reengineering trend, SAP’s ERP
suite and those from rivals Oracle, Baan, and PeopleSoft took off from 1992 to 1996 as
firms aggressively redesigned their business processes to lower costs. Customer relationship
management (CRM) software from Siebel; supply chain management (SCM) software from
i2 and Manugistics; and product life-cycle management (PLM) from Parametric Technology
Corporation (PTC), Unigraphics (now Siemens PLM Software), and Dassault Systemes followed
to automate these business processes. Then, the Internet took off, powered by the browser
wars between Microsoft’s Internet Explorer and Netscape Navigator. New eCommerce sales
software from ATG and BroadVision and eProcurement and eSourcing software from Ariba
and Commerce One pushed enterprise applications into sales and purchase processes over the
Internet. As a result, IT-to-GDP ratio shot up from 2.36% in 1992 to 3.88% in 2000 — with the
Y2K frenzy adding about a half a percentage point of IT investment (see Figure 2-3). But the
tech bubble burst in 2001, and the resulting recession in the overall economy led to big drops in
tech investment in 2001 to 2003, followed by four years in which tech investment grew at about
the same rate as the overall economy.
Tech Tech
1.2% introduction digestion
and and
US IT investment-to-GDP ratio
0.0%
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
IBM 1400 series UNIX OS Ethernet
mainframe; IBM HP-2115 created
1301 disk mini- Data General
storage unit computer minicomputer Intel 8008
COBOL microprocessor
DEC PDP-8
programming minicomputer First email;
language first floppy Altair 880
IBM system/360; CDC 6600 diskette personal
IBM 7000 mainframes supercomputer; computer;
with transistors BASIC programming language Telenet as first
value-added
network
Source for photo: IBM
55157 Source: Forrester Research, Inc.
1.5% 0.81%
Computer hardware
0.19%
1.0%
0.27%
0.85%
0.5% 0.64% Communications equipment
0.0%
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
DEC’s VAX 11 Microsoft
minicomputer; Apple Lisa
introduces graphical Windows 3.0;
WordStar word
processor user interface (GUI) PeopleSoft World Wide
V1HRMS Web born
Commodore Pet and Commodore 64 PC: Lotus suite with HTML,
Apple II personal 1-2-3; Time magazine released HTTP, and
computers names the computer URLs
Apple-1 personal computer; “Machine of the Year”
IBM PS/2; Microsoft
Cray-I supercomputer; Windows 2.0; Oracle’s
CP/M PC operating system IBM PC with MS-DOS; UNIX-based apps
Osborne 1 portable PC
Source for photo: IBM
55157 Source: Forrester Research, Inc.
and and
3.0% growth refinement
2.5% 1.16%
0.70%
2.0%
Computer hardware
1.5% 0.81%
0.69%
1.0%
0.85% Communications equipment
0.5% 0.79%
0.0%
92
93
94
95
96
97
98
99
00
01
02
03
04
05
06
07
08
19
19
19
19
19
19
19
19
20
20
20
20
20
20
20
20
20
Unified communications
Netscape goes products start to be sold
public Y2K; Concept of service-oriented
Intel’s Pentium with industry architectures (SOA) starts to be
x86 architecture; eMarkets used; “utility computing” term gets
Mosaic Web browser picked up
Software products Cloud
SAP R/3 ERP eBusiness applications; computing
suite released Jeff Bezos of Amazon.com based on Web services
and software-as-a-service term
named Time “Person of introduced
the Year” (SaaS) enter the market
Smart Computing, Latest Cycle Of Tech Innovation And Growth, Began In 2008
Why does this history of technology since 1960 matter today? Because cycles of tech adoption and
digestion that occur with this regularity have a high likelihood of repeating themselves. Of course,
as financial investment literature always points out, past history is no guarantee of future results.
Still, cycles of booms, recessions, and recoveries have been a persistent feature of the world of
economics — and every time that optimists have predicted the end of economic cycles, a recession
has brutally thrown cold water on these hopes. We expect similar cycles in the world of technology.
Using this logic, the innovation and growth phase of the next tech cycle should have started in 2008,
eight years after the peak of the last cycle and after eight years of tech digestion and refinement.
Did it? Yes, for two reasons. First, the data shows that tech investment in 2008 started to grow
significantly faster than the economy — at least until the financial crisis in the late third quarter
and fourth quarter threw that trend into reverse. Second, a review of the technology landscape
shows a convergence of innovation in four domains of technology (software applications, data
center operations, client devices, and network systems) that together represents a new generation of
technology — Smart Computing.
Tech Investment Started To Outpace GDP Growth Until The September 2008 Financial Crisis
Apart from a five-quarter period from Q4 2004 to Q4 2005 when growth in business investment in
computer equipment, communications equipment, and software lagged nominal GDP growth, the
growth rates in these two metrics were more or less the same from 2004 through Q3 2007. Then,
starting in Q4 2007, while nominal GDP growth slowed as the US moved into a recession, IT
investment growth increased to 10% in Q4 2007, 8% in Q1 2008, and 9% in Q2 2008, before slipping to
a still positive 5% in Q3 2008 when the financial crisis hit the US and other countries (see Figure 3). In
each of these four quarters, tech investment growth was almost twice the growth in nominal GDP.
Despite the weakening of the US economy in late 2007 and the first three quarters of 2008, US
businesses were starting to invest heavily in technology.
Figure 3 Tech Investment Grew Much Faster Than GDP In 2008 — Until The Financial Crisis Hit
. . . and will do so
Tech investment started again in 2010.
to boom in late 2007 Business
18% and the first half of 2008 . . . investment
15% in IT
12%
9% Nominal GDP
6%
3%
0%
-3%
-6%
-9%
-12%
-15%
-18%
00 03 04 05 06 07 08 09 * * *
0 001 02 0 0 0 0 0 0 0 010 011 012
42 42 20 42 42 42 42 42 42 42 42 42 42
Q Q Q4 Q Q Q Q Q Q Q Q Q Q
Source: US Commerce Department, Bureau of Economic Analysis through Q3 2009; Forrester Research for
forecasts for 2009 to 2012
*Forrester forecast
55157 Source: Forrester Research, Inc.
What were businesses buying in this period? Certainly not servers or PCs from the largest vendors,
where the average growth rates for the four quarters through Q3 2008 were negative (-12% for
servers; -1% for PCs). Nor did vendors of storage equipment, computer peripherals, and software
operating systems see much growth, with average growth rates of 1% to 3%. Instead, the growth was
occurring in:
· Special-purpose computer equipment. Traditional servers, PCs, and even storage experienced
declining sales from Q4 2007 to Q4 2008. Bucking that trend were industry-specific computer
devices such as smart meters for utilities, network-connected sensors in highways and pipelines,
computer chips for airplanes and automobiles, or computer systems designed for physicians
offices to handle electronic patient records and digital images. For example, Texas Instruments’
embedded devices unit, which provides chips for cars, medical imaging equipment, and other
industry devices, reported revenue growth of 6% in Q4 2007, 11% in Q1 2008, and 12% in Q2
2008, before plunging in both Q3 and Q4 of 2008 as target industries like automotive crashed
(see Figure 4-1). Smart meter vendors like Elster Group and Landis+Gyr, although they don’t
report quarterly revenue growth, did report good revenue growth for 2008 as a whole compared
with 2007.
· SOA infrastructure and business intelligence (BI) and analytical software. Software in
general had positive growth in late 2007 and also saw positive growth the first three quarters of
2008. But sellers of application servers and service-oriented architecture (SOA) infrastructure
grew much faster, as did sellers of BI and analysis software. With the exception of pure-play BI
public companies like MicroStrategy and Actuate, all the big sellers of BI software are either
parts of larger software vendor or private companies; so it is difficult to track quarterly demand
in that area.4 However, anecdotally, the vendors did report strong revenue growth in that
period.5 In application servers and SOA, IBM WebSphere and TIBCO Software have quarterly
reported revenues that showed much stronger growth in early 2008 — although both got caught
in the downturn in software investment after Q3 2008 (see Figure 4-3).
Figure 4 Tech Investment Started To Take Off In Late 2007 But Not In Old Tech Categories
4-1 Specialized industry devices were still growing when PCs and servers fell
Percentage change from prior year
10%
6%
2%
-2%
-6%
-10%
Large computer
-14% equipment vendors
-18%
-22%
Example: Texas Instruments’
-26% embedded devices unit
7 8 8 8 08
2 00 200 2 00 200 20
Q4 Q1 Q2 Q3 Q4
4-2 Videoconferencing and unified communications boomed while routers and switches started to fall
Figure 4 Tech Investment Started To Take Off In Late 2007 But Not In Old Tech Categories (Cont.)
4-3 SOA and businesses intelligence outpaced ERP and infrastructure software
36%
32%
28%
24%
20%
16%
12%
8%
Example: IBM WebSphere
4%
0% Large software vendors
-4%
-8% Example: TIBCO Software
07 08 08 08 08
20 20 20 20 20
Q4 Q1 Q2 Q3 Q4
This pattern of IT purchases in late 2007 and early 2008 shows the emergence of a new generation of
technology — Digital Business Architecture — introduced by Forrester Vice President and Principal
Analyst Randy Heffner.6 This new generation of technology is more than just new computer
hardware and software as was the case with mainframe computing, personal computing, and
network computing. It is more than a movement to a new tier of computing, as occurred with the
transition from one-tier architectures of mainframe and personal computing to the two-tier client-
server and three-tier architectures of network computing. Digital Business Architecture represents
innovations in four domains of technology: 1) business services software; 2) Organic IT server-
based computing; 3) unified communications in network technologies; and 4) pervasive interaction
from client devices (see Figure 5).
Configuration
• Unified management • Business service
communications databases management
platform platform
Security
Source: July 3, 2007, “A Taxonomy Of Platforms For Your Digital Business” Forrester report
55157 Source: Forrester Research, Inc.
If we think about any concept of smart behavior or smart actions, these consist of five stages of
activity. Smart Computing uses digital business architecture technologies, either brand new ones or
new deployments of existing technologies, to support each of these five stages of intelligent activity
(see Figure 6):
· Awareness. New technologies for pervasive interactions such as radio frequency identification
(RFID), sensors, video cameras, global positioning system (GPS) chips, smart cards, and other
tools will capture data on the identity, status, condition, and/or location of people and physical
assets — data that indicates anomalies that present a business opportunity, activity, threat, or
risk.7 Unified communications technologies such as third-generation (3G) wireless networks
will transport this data from these client devices back to central servers for analysis.
· Analysis. Business intelligence and specialized analytical software such as data mining and
predictive analytics, video image analysis, pattern recognition, and artificial intelligence
algorithms will determine whether businesses or governments should act on or ignore a pattern
or an anomaly. Businesses and governments have already been using these analytical tools for
making sense of historical data, as well as for starting to make predictions about what may happen
next. But now, they will be deployed against the real-time data being transmitted from the new
awareness devices. Analyzing and storing the massive amounts of data that will be received
is only possible with the more flexible and adaptable servers and storage devices enabled by
server virtualization, data center automation, and storage life-cycle management — as well as the
potential for more flexible processing expansion and storage capacity through cloud computing.
Expect more of the basic processing that sifts out meaningful information from background noise
to happen at the fringes of the unified wireline and wireless broadband networks that connect
to the awareness devices. For this analysis to have business value, though, it will need to present
alternatives.
· Alternatives. Rules engines and workflow are the existing technologies for deciding which
alternative courses to pursue, either automatically through the application of a rule that says
“if this happens, do this,” or through human review based on workflow engines that route the
anomaly and alternative courses to the right person to make a decision.8 The basic function of
rules engines and workflow will stay constant — seismic leaps will be necessary in the data flow
and analytical inputs in a world of vastly expanded real-time awareness. For example, rules
engines will need to adapt and change their rules on the fly (based on new analysis of what the
best alternatives should be), and workflow engines will need to change rapidly what alternatives
should be presented to which people based on the seriousness of the issue. In either case, once a
human being or a rules engine makes a decision on what to do, that decision should trigger the
requisite actions.
· Actions. The action may be as simple as quoting a different price, placing a new order, making
a new offer to a customer, or initiating a customer service contact. Or the action may be as
complex as adjusting thermostats in tens of thousands of households and businesses to avoid an
electricity brownout. These actions will be executed through integrated links to the appropriate
process applications. The spreading conversion of process apps to service-oriented architectures
will allow these process apps to be adapted to business scenarios, with specific app components
pushed down to the awareness devices where they can execute that action, whether that is
alerting a citizen on her smartphone to the updated arrival time of a bus that was stuck in traffic,
notifying a doctor on a tablet device about the drug allergies of a patient he is about to
see, or directing the thermostat in an individual home to raise the temperature by turning up
the air conditioner by three degrees. But whether the right action was actually taken can only be
determined if there is auditability.
· Auditability. Tracking all steps in the process to aid in regulatory compliance, compliance
with company policies and goals, and improvement opportunities is critical. Any definition
of “smartness” includes elements of monitoring activity and learning how to do it better.
Technology needs to capture, track, and analyze information on each stage of this cycle to make
sure that the right actions were taken and to learn how to improve the analysis and identify
better alternatives.
Auditability
Using data on activity at each stage, record what happened and analyze for purposes
of compliance and improvement.
New capabilities
Improved analytical tools can then use this awareness data to identify locations where traffic is
slowing or a traffic jam is starting to clear, presenting the driver with alternatives as to whether it
will be better to take the next exit or stay on the road because traffic will soon be moving. And the
driver can now take the appropriate action — or perhaps in the future the car’s autopilot will do so.
Businesses and government regulators can audit each of these activities to make sure the provider
(for example) is not discriminating in its presentation of alternatives (e.g., all drivers of BMWs get
the best information, while those driving Dodges get the bad advice) and to improve the quality of
the analysis. But without the awareness of traffic conditions (created by the growing ubiquity of cells
phones and satellite navigation systems that provide real-time data on location), the rest of the A’s in
this scenario would not be possible.
What are these critical problems that prior technologies have not been able to solve? Look back at
the three prior waves of technology, and the answer starts to become clear through a process of
elimination (see Figure 8). Remember:
· Network computing helped automate key business processes. Business processes include
paying a supplier, billing a customer, taking an order, buying from a supplier, hiring and paying
an employee, and manufacturing a product or designing one. While opportunities still exist to
automate other processes (for example, generating a contract or creating a sales proposal) as
well as to link related processes, most of the big efficiency gains from process efficiency have
already been realized.
Figure 8 Smart Computing Will Help Solve Balance Sheet Business Problems
Optimized
Smart balance sheet
Computing results
These three waves of technology have mostly had a business impact on the income statement.
They reduced expenses through efficiencies that reduced the role of people and paper, eliminated
errors and re-work, and shortened cycle times. Or they increased revenues by improving customer
acquisition and retention, removing barriers to purchase, opening up new sales channels, and
shortening cycle times.
There are of course big income statement challenges that still need the right technology solution.
New business processes are arising to take advantage of what technology can enable, such as
customer-driven product innovation; crowd-based problem solving; and contract-driven pricing,
quoting, configuration, or services. And while many business processes have become more efficient
(thanks to network computing technologies), there is still room for making them more effective by
applying rules engines, workflow, analytics, and business process management to help processes
deliver better, more optimal results. These are all areas where Smart Computing solutions can deliver
real value to the business, through the introduction of Smart Computing process applications based
on the design principles that Forrester has called Dynamic Business Applications.10
Still, the next big business challenge lies more in the balance sheet — in optimizing the value of
and the returns on assets and minimizing the costs and risks from liabilities. New tools will provide
businesses and governments with far better real-time awareness of the status of their assets and
liabilities, as well as vastly improved analysis of how to maximize the returns from the assets and the
costs and risks from their liability. Today, companies and governments have awareness and analysis
of their financial assets and liabilities (although these areas still have ample room for improvement).
In the next phase, awareness and analysis of the balance sheet will focus on
physical assets and liabilities, such as cars, trucks, airplanes, office buildings, hospitals, transmitters,
pipelines, equipment, and machinery. And awareness and analysis technologies will quickly spread
to intangible assets like intellectual property, brand, customer or supplier contracts, or knowledge
workers in a workforce.
The abilities of Smart Computing to optimize the management of the balance sheet will meet a ready
audience because the current recession has heightened CEO awareness of the importance of the
balance sheet. The 2008 to 2009 recession was in many ways a balance-sheet-driven downturn. The
housing crisis that was the trigger for the downturn was the result of consumer mortgage liabilities
getting way ahead of the sustainable value of the home assets that consumers borrowed against.
Similarly, the financial crisis that pushed the global economy to the brink of a depression occurred
because of major imbalances between bank liabilities and assets — imbalances that had to be closed
through massive infusions of government support. CEOs understand that they need to pay much
more attention to the balance sheet, and the risks and opportunities that lurk in it, and not just the
income statement of revenues and costs.
· Real estate performance management systems. Until 2004, Accruent was a software vendor
selling a real estate contract management product, a focused application that about 300 clients
use to track and manage their real estate contracts. However, when Edward Lampert bought
Kmart in 2003 and then Sears Roebuck in late 2004 on the basis of their undervalued real estate
assets, other retailers suddenly realized that they were sitting on portfolios of real estate assets
that they knew nothing about. Seizing on this market opportunity, Accruent repositioned its
product in July 2006 as a real estate performance management solution specifically for retailers.
It could provide retailers with awareness of their real estate contracts, the location of their
properties, and the demographics of their property markets. It helps retailers analyze where
their properties were located in relation to competitors’ properties and to customers, and it
linked this analysis to application modules for lease administration, facilities management,
project management, and site selection. While Accruent may not have provided retailers with
real-time awareness of property status and condition (although it could be added), simply
consolidating existing data about properties has been a distinct improvement for retailers in
optimizing the value of their real estate assets and has allowed Accruent to double its price point
for its solutions.11
· Smart meter and smart grid systems for utilities. Smart meter systems use small computer
chips to capture real-time electricity usage in homes and businesses, with that information
relayed via wireless networks to central servers at the utilities. Smart grids put sensors
throughout the electric distribution networks that provide utilities with much better awareness
of end-customer demand and traffic through their power grids. Utilities then analyze that data
to detect heavy use patterns, make predictions about the future, and take corrective actions
in the form of sending instructions back to the home or businesses to shut off non-critical
equipment (such as a dishwasher) for a period of time or adjust thermostats up or down a few
degrees. Within the grid, power can be diverted into storage systems (such as pumping water
up to the top of a reservoir) or drawn from storage. Many vendors are active in this arena —
including smart meter vendors like Elster, Itron, and Landis+Gyr; power generation equipment
vendors like GE and Siemens; network equipment vendors like Cisco; existing software vendors
like Oracle and SAP or startups like GridPoint and Silver Spring Networks; and IT services
vendors like IBM and HP/EDS. Utilities like Pacific Gas & Electric in California, Hydro One in
Canada, and Enel in Italy are already seeing benefits.12
· Healthcare and patient records management systems. The healthcare industry, especially in the
US, has been slow to adopt technology. True, most major hospitals have hospital management
systems; invest in expensive computed tomography (CT) and magnetic resonance imaging
(MRI) technologies; and equip doctors, nurses, and technicians with PCs and increasingly
tablet computers. However, with some conspicuous exceptions, electronic patient records are
still few and far between, and many physicians and smaller hospitals have only the most basic
of computer equipment. Denmark has an electronic health records system that covers virtually
the entire population, creating awareness of all current and past treatments, medications, and
patient histories that physicians and nurses can access at the time of treatment. The US Veterans
Health Administration and Kaiser Permanente in the Western US provide such a system as well.
Once awareness of patient records and treatment has been established, then analysis can be done
across many patients to determine which treatments have been most effective with the lowest
cost, leading to better healthcare delivery and improved economics. However, the challenges are
significant, especially getting physicians, nurses, and other healthcare providers to use the systems
and reassuring patients that their records will be kept private. For example, the UK’s project to
convert the National Health Service to electronic records is behind schedule and well over budget,
with mixed results so far.13 Still, dozens of tech vendors are pursuing this market, including
Accenture, Cerner, GE, IBM, Siemens, and many others.
These are just a few examples of Smart Computing solutions that are emerging. Other examples
include solutions for transportation, such as improved traffic management like the London
congestion charging system that imposes variable fees for autos entering Central London, and
smart bridges like the I-35 bridge in Minneapolis, Minnesota, that contains more than 300 sensors
to track the effects of corrosion, temperature, and icing on bridges.14 In the future, we expectSmart
Computing solutions will emerge for education (combining display devices like smart boards,
student laptops, and video systems with online learning tools to help students learn at their own
pace, work collaboratively, and access the teachings of peers or remote teachers) and for professional
services industries (new tools for knowledge capture and knowledge sharing; collaboration on
projects; and improved project definition and project management to help legal, consulting,
accounting, engineering, or research firms optimize the returns from their human assets).
As these examples suggest, the benefits of Smart Computing for business will be real and substantial.
But the risks and obstacles are just as real and substantial. Apart from issues of security, standards,
and interoperability, as well as the process and organizational changes that accompany any new
technology, Smart Computing will raise some very troubling issues of customer data privacy. As
utilities start to gain awareness of how end customers are using electricity and natural gas, as
healthcare providers start to build awareness of the disease and treatment history of their patients,
and as GPS vendors gain insight into the location of the cars using their systems, the temptation of
these companies will be to use this information for their own benefit. That way conjures up images
of an Orwellian Big Brother (whether Big Brother is government or corporate giants). Addressing
up front the question of who controls and gains from Smart Computing will be key to its public
acceptance and thus to the success of its adoption by business (see Figure 9).
With Smart Computing looming on the horizon, it is critical to recognize that end users
and consumers will only accept it if they get control of the results. Rather than utilities
using the data on end consumer electricity use to try to control that use from their
end, they will need to provide this information to consumers to allow them to make
their own decisions about usage. Experiments that IBM has done with providing
consumers graphical data on their energy users has shown that informed users will
take steps to cut their own usage. Similarly, The Economist has argued persuasively in
its survey of healthcare and information technology about the need to give patients
control over their own electronic records if they are going to accept this technology.
versus
And, in the GPS example cited previously, it actually makes more sense to pass the
information about looming traffic jams to individual drivers to let them make their own decisions, rather
than having highway authorities posting electronic alerts to all drivers. After all, the best way to deal
with an impending traffic jam is to get 10% to 15% of drivers on their own shift to an alternative route,
rather than having 100% of drivers do so, thereby creating a new traffic jam.
Smart Computing Will Have Biggest Impact AS Verticals 3.0 Industry Solutions
The business examples that we’ve cited demonstrate another characteristic of Smart Computing — a
very high degree of vertical industry focus. As we noted, businesses will adopt Smart Computing
technologies because they help them address the key challenge of optimizing the value of their
balance sheets, allowing them to move beyond financial assets and liabilities to their physical assets
and liabilities (like electric grids or hospitals) and then to their intangible assets and liabilities (like
a skilled workforce or brand). Assets and liabilities tend to be very industry-specific, even more
so than processes that may be common across industries. And the task of optimizing the value of
these assets and liabilities is definitely industry-specific because what optimization means will vary
dramatically from industry to industry. For example:
· Financial services firms focus on financial assets and liabilities. A financial services company
will place most emphasis on optimizing the value of its financial assets and liabilities, a medium
emphasis on optimizing the value of its human assets, and a low value on optimizing the value
of its physical assets.
· Professional services firms value human assets the most. A professional services firm that
does primarily consulting will place most emphasis on optimizing the value of human assets,
medium on financial assets, and little on physical assets; while a professional services firm that
does outsourcing may place most of its emphasis on its physical assets, then its financial assets,
and finally on its human assets.
· Verticals 1.0 focused on core transaction applications. The first “release” of verticals 1.0
coincided with the mainframe computing era of the 1960s and 1970s. The very first mainframe
computers — ENIAC and UNIVAC — were custom-built for the US Census Bureau and
other government clients. Even after mainframe computers and minicomputers became more
horizontal technologies that were the same across industries — the companies built the software
running on this hardware specifically for vertical industries like airlines, utilities, banks,
insurance carriers, and government agencies. The second release of verticals 1.0 (V1.1, as it
were) came in the early days of network computing when ERP vendors like Baan, Oracle, and
SAP built core manufacturing management systems for manufacturers. Similarly, vendors like
Retek (acquired by Oracle in 2005) and JDA Software built similar systems for retailers, while
Cerner, Shared Medical Systems (acquired by Siemens in 2000), HBO & Company (acquired by
McKesson in 2000), and IDX (acquired by GE in 2006) created healthcare management systems
for hospitals.
· Verticals 2.0 shuffled process app modules appropriate for different industries. This version
of verticals started to surface in the 2000s as the large app vendors like Oracle and SAP started
to show how clients in different vertical industries could arrange modules of their ERP, CRM,
supplier relationship management (SRM), SCM, PLM, and other products to fit processes
common to those industries. SAP took this approach the furthest, creating process maps for
dozens of different industries. This approach allowed the vendors to eat the slices of vertical
solutions while still retaining the cake of horizontal process applications.15
· Verticals 3.0 will address core balance sheet problems in an industry. As discussed above,
verticals 3.0 will create solutions that combine elements of industry-specialized hardware
devices, vertical industry software, and industry-focused wireless/wired networks with
industry-oriented analytics to address core balance sheet challenges in a vertical industry. Some
current examples moving in this direction include Aptify for association management systems,
Blackbaud for nonprofit associations, Cúram Software for government social service agencies,
CygNet for oil and gas pipeline optimization software, and SunGard Higher Education for
universities.
However, vendors need to understand that tech purchases are not evenly distributed across
industries. In fact, our analysis of tech spending by industries shows that the seven industries that
can benefit the most from Smart Computing (healthcare, education, government, professional
services, transportation and logistics, telecommunications, and utilities) represent 60% of total US IT
spending.16 These are all industries where balance sheet optimization is more important than process
efficiency and where the need for Smart Computing technologies is critical (see Figure 11).
Figure 11 Asset-Intensive Industries That Benefit From Smart Computing Are Big IT Spenders
Industry groups
23%
Professional services
Government
Financial services These seven industries
represent 59% of the US
Healthcare tech market.
20%
Education
Media & entertainment
Industrial products
Telecommunications
14%
Consumer products
High-tech products
6% Retail
4% Insurance
5%
Transportation & logistics
4%
3% Construction & engineering
3% Chemicals & petroleum
3%
3% Wholesale trade
3% Primary production
2% Utilities
1%
Source: May 7, 2009, “US Enterprise Versus SMB IT Budgets In 2009” Forrester report
55157 Source: Forrester Research, Inc.
Remember, also, that these industries are major beneficiaries of the economic stimulus program in
the US.17 So, they have funds to invest in Smart Computing technologies (see Figure 12).
In short, vendors that choose to build Smart Computing solutions that focus on vertical industries —
and that pick the right verticals — are likely to find the market opportunity to be larger than those of
classic horizontal offerings.
Figure 12 These Industries Are Also Big Beneficiaries Of The US Fiscal Stimulus Program
Funding available
to IT vendors
(US$ millions)
Broadband Benefits
communications Telecommunications
$5,400
Tech investment
tax incentives
$1,500
Source: July 7, 2009, “There’s Gold For Vendors In Stimulus Packages” Forrester report
Note: Based on analysis of the American Recovery and Reinvestment Act of 2009
55157 Source: Forrester Research, Inc.
Playing The Angles: Where The Opportunities In Smart Computing Will Lie
Where will tech dollars flow in the next seven to eight years? It may seem that the answer is all into
Smart Computing technologies. But technology markets don’t work that way. Existing technologies —
servers, PCs, routers, switches, database management systems, ERP software, etc. — don’t go away.
Indeed, well-established, mature tech products will continue to represent the bulk of technology
purchases even eight years from now. The problem is that revenues from these products won’t grow by
much. Instead, almost all of the growth will occur in three areas:
Growth rates aside, we think that the total revenue from industry-focused solutions will outstrip
the revenues from Smart Computing foundation technologies by two to one and that they will
exceed the revenues from Smart Computing process applications by almost three to one. By 2016,
we predict that Smart Computing industry vertical solutions will equal 0.86% of GDP, while Smart
Computing foundation technologies will equal 0.42%, and Smart Computing process applications
will be at 0.30% (see Figure 14). Over this eight-year period, we expect that Smart Computing
industry vertical solutions will increase their share of the IT market from the 6% that they represent
in 2008 to 23% by 2016.
13-1 Current-generation technology will remain much larger than Smart Computing technologies or
industry-specific technologies . . .
Government and business spending
Smart Computing industry solutions on IT equipment and software
Smart Computing process applications (US$ billions) $787
Smart Computing platform technology $781
Current-generation technology $180 $213
$682
$615 $148
$569 $116 $63 $73
$508 $93 $47 $89 $96
$443 $71 $12 $21 $30
$419 $395 $68 $77
$2 $27 $369 $51 $47 $61 $448
$22 $27 $35 $5 $37 $7 $411 $404
$3 $23 $31 $395 $401
$348 $378
$368 $316 $324
2008 2009* 2010* 2011* 2012* 2013* 2014* 2015* 2016* 2017*
13-2 . . . but Smart Computing and industry-specific technologies will grow much faster
13-3 Unified communications, Smart Computing process applications, and vertical business apps will
be the largest of the new technologies
A spreadsheet with additional data is available online.
Government and business spending
on IT equipment and software
(US$ billions)
$644
$14
$46 Smart Computing
vertical solutions
27%
$120 CAGR
Smart Computing
$45 foundation
technologies
Industry-specific communications $25 19%
Industry-specific devices $19 CAGR
Industry vertical applications $63 Smart Computing
process applications
Unified communications $398 $1 50%
$5 CAGR
Microcomputing devices $21 $5 $108
$15 $3
SOA and BPM
$2
Smart Computing process applications
$108 $78
Current-generation communications
equipment
Current-generation
Current-generation computer $78 technology
equipment 3%
Current-generation software CAGR
$181 $262
2008 2016*
*Forrester forecast
55157 Source: Forrester Research, Inc.
14-1 Smart Computing process applications and vertical industry solutions will drive the expansion
of IT’s role in the economy
Government and business spending on IT equipment and software
(percentage of GDP)
4%
US IT investment-to-GDP ratio
0.98%
3% 0.18% Smart Computing industry solutions
0.02% Smart Computing process applications
0.15% 0.34%
2.55% Smart Computing platform technology
2% 0.44%
Current-generation technology 1.86%
1%
0%
2008 2009* 2010* 2011* 2012* 2013* 2014* 2015* 2016* 2017*
*Forrester forecasts
14-2 Smart technologies will expand their shares of the US economy, while current-generation
technologies will lose share
Government and business spending on IT equipment and software
(percentage of GDP)
2008 2009* 2010* 2011* 2012* 2013* 2014* 2015* 2016* 2017*
Industry-specific communications 0.00% 0.01% 0.01% 0.02% 0.03% 0.04% 0.05% 0.06% 0.07% 0.07%
Industry-specific devices 0.03% 0.04% 0.07% 0.10% 0.12% 0.15% 0.17% 0.20% 0.22% 0.25%
Industry vertical applications 0.15% 0.14% 0.15% 0.21% 0.28% 0.34% 0.40% 0.49% 0.57% 0.66%
Unified communications 0.03% 0.03% 0.04% 0.05% 0.07% 0.12% 0.14% 0.18% 0.21% 0.24%
Microcomputing devices 0.10% 0.10% 0.14% 0.14% 0.15% 0.15% 0.14% 0.13% 0.12% 0.11%
SOA and BPM 0.02% 0.02% 0.03% 0.04% 0.06% 0.07% 0.08% 0.08% 0.09% 0.10%
Smart Computing process
applications 0.02% 0.02% 0.03% 0.05% 0.07% 0.12% 0.16% 0.23% 0.30% 0.34%
Current-generation
communications equipment 0.75% 0.65% 0.64% 0.62% 0.63% 0.62% 0.57% 0.53% 0.51% 0.45%
Current-generation computer
equipment 0.54% 0.43% 0.39% 0.37% 0.39% 0.34% 0.36% 0.36% 0.37% 0.34%
Current-generation software 1.25% 1.15% 1.19% 1.23% 1.25% 1.25% 1.21% 1.18% 1.24% 1.06%
Total 2.90% 2.60% 2.70% 2.83% 3.04% 3.20% 3.28% 3.43% 3.71% 3.62%
Figure 14 Current-Gen Technologies Shrink, New Technologies Rise As Percentage Of GDP (Cont.)
14-3 Vertical industry solutions will expand from 6% of the US tech market in 2009 to 27% by 2017
20%
0%
2008 2009* 2010* 2011* 2012* 2013* 2014* 2015* 2016* 2017*
14-4 The tech market will become more complex as new technologies take share from, but co-exist with,
older ones
Government and business spending on IT equipment and software
(percentage of total technology investments)
2008 2009* 2010* 2011* 2012* 2013* 2014* 2015* 2016* 2017*
Industry-specific communications 0% 0% 0% 1% 1% 1% 1% 2% 2% 2%
Industry-specific devices 1% 2% 3% 3% 4% 5% 5% 6% 6% 7%
Industry vertical applications 5% 5% 6% 7% 9% 11% 12% 14% 15% 18%
Unified communications 1% 1% 2% 2% 2% 4% 4% 5% 6% 7%
Microcomputing devices 4% 4% 5% 5% 5% 5% 4% 4% 3% 3%
SOA and BPM 1% 1% 1% 2% 2% 2% 2% 2% 2% 3%
Smart Computing process
applications 1% 1% 1% 2% 2% 4% 5% 7% 8% 9%
Current-generation 26% 25% 24% 22% 21% 20% 17% 15% 14% 13%
communications equipment
Current-generation computer 19% 17% 14% 13% 13% 11% 11% 10% 10% 9%
equipment
Current-generation software 43% 44% 44% 43% 41% 39% 37% 34% 34% 29%
Total 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%
*Forrester forecast
55157 Source: Forrester Research, Inc.
For today’s technology vendors, the challenge will be how to play this market shift.
· “Stick to your knitting” will only work for a few scale players. For most vendors, selling existing
servers, PCs, storage devices, and enterprise applications will be a recipe for shrinking revenues,
given the impacts of cloud computing on these product markets. But demand for these purchased
technologies will not disappear. A few vendors that focus on being the consolidators of these
shrinking but still large product markets will be able to survive and even thrive.
· A feeding frenzy will go after the Smart Computing horizontal technologies. Vendors
that are committed to a horizontal sales model will view the projected growth rates for Smart
Computing technologies — whether for platform technologies or for new process applications
based on Dynamic Business Application principles — as their opportunity to escape the
prospect of commoditized current products with lousy growth rates. The problem is that many
other vendors will also be pursuing the same market opportunity. Lots of vendors fighting for
the same market will mean a few winners and many losers.
· Vertically focused vendors will thrive or struggle as their verticals do. The biggest
opportunity will be in Smart Computing vertical solutions to help companies in specific
industries optimize the returns from their balance sheets. But, this is by no means a risk-
free strategy. Focusing on some verticals means giving up potential opportunities in others.
Focusing on very narrow verticals with no leverage to other verticals puts a cap on growth.
Focusing on a vertical that goes into a steep economic decline (think of the auto industry or
banking in 2008 and 2009) means that you suffer as that vertical suffers. Yet, there are many
verticals — and even more microverticals — and each will be willing to pay a lot of money for
technologies that can get it significantly better returns from assets and liabilities.
· IT consulting and systems integration vendors have a mixed outlook. In the short run,
companies will need consulting help to plan strategies for SOA, unified communications, and
the emerging Smart Computing solutions. Systems integration (SI) projects will also do well
in 2010 as companies revive their postponed capital investment plans for CRM, ERP, SCM,
SRM, and other stalled applications. However, the emergence of Smart Computing solutions for
different verticals will present challenges for IT services vendors. The goal of many vendors will
be to create semi-turnkey solutions that pull together all the software and hardware needed: for
example, a smart meter solution for utilities or a healthcare records management offering for
hospitals, not a collection of different products from many different vendors that an SI would
assemble as a general contractor on a custom project basis. To the degree that this becomes a
reality, and admittedly it will take a few years before these offerings get productized this way,
it will make them much more affordable to clients and speed adoption. But it will also steal
revenues from the consultants. So, the consulting firms will need to build out their own software
capabilities to compete. As a result, we expect IT consulting and services revenues to track at
lower levels then the growth rates in software revenues (see Figure 15).
Figure 15 IT Services And IT Outsourcing Will Lag Software In This Generation Of Technology
*Forrester forecast
55157 Source: Forrester Research, Inc.
· IT outsourcing vendors have less promise in the next seven years. Certainly, outsourcing
of the last generation of technology will continue because it is common for companies to turn
the running and management of mature technologies that provide no competitive advantage
over to specialists who can run it more effectively. However, cloud computing will provide an
alternative to traditional IT outsourcing; so, some of the tendencies to turn existing technologies
over to IT outsourcing will get diverted into acquiring these capabilities as a cloud service. More
significantly, the new technologies will be resistant to outsourcing for a while. Whenever a new
technology comes along, the last thing that companies want to do is outsource it. Instead, they
want to assess it themselves and learn. This is hard to do with an outsourced technology, where the
IT outsource vendor controls it, sets terms and conditions for its use, moves at its own pace, and
limits adaptation to preserve its own efficiencies and scale economies. As a result, we expect that
IT outsourcing will grow at a steady but slow rate of 4% to 6% over the next six or seven years.
1. Ability to pull together the hardware, software, and network elements of Smart Computing.
These solutions will not work if they are a collection of separate hardware, software, and network
products from different vendors unless, and until, there are clear standards and protocols that
define how each part works with the others. Those standards and protocols for interoperability,
security, and performance will come in time. But, for now, vendors that can provide all the
elements of a full solution, and can do so as a product that can be sold to many clients, will have
the edge over consultants who stitch these pieces together as custom consulting projects or over
single-product vendors.
2. Expertise in the skills and technologies that will be differentiators in Smart Computing.
While Smart Computing involves combining different technology elements, not all elements are
equal. Of the five A’s of Smart Computing, it is awareness and analysis technologies that will be
the differentiators for vendors initially, with technologies for determining the right alternatives
coming to the fore in the future. For vendors that go after vertical industry solutions, asset
management software (especially IT asset management tools that are connected to networks and
have automatic asset discovery and management) and contract life-cycle management (to track
and manage the contracts related to these assets and liabilities) will be key technologies. And
successful vendors will need to have employees with statistical analysis skills, who can design
and set up the systems to create awareness of asset status, structure the analysis of this data,
define rules and workflow, and identify the right applications to initiate the appropriate actions.
3. Ability to play the angles between horizontal and vertical solutions. In a period when
horizontal technologies make up the bulk of IT purchases but grow slowly if at all (and vertical
technologies provide much of the growth, but still are only one-quarter of the market), solving
this dilemma will be the biggest challenge that many vendors face. Large vendors with dominant
shares and scale economies in horizontal technologies can add elements of cloud computing
and consolidation to outpace their stagnating market segments. Small vendors and new entrants
can easily prosper in new vertical niches. But everyone else will need to carefully pick the
horizontal technologies that they want to master and/or the verticals that they want to dominate
and give up the others.
4. In-depth understanding of the balance sheet issues facing specific industries. For those
vendors that pursue a vertical industry strategy, choosing which verticals to go after will be a
key success factor. Because balance sheet challenges tend to be unique to an industry, crafting
the right combination of Smart Computing elements to address these challenges requires deep
understanding of that industry and those challenges. Shallow understanding of all industries
will not cut it. So, vendors will have to focus on some verticals, or even microverticals, and forgo
others.
To find a good example of a vendor today that has combined hardware, software, and network
elements to create a compelling solution, you have to look outside the world of business IT and look
instead at consumer technology. That example is Apple Computer and its nexus of iTunes, iPod, and
iPhone products. Like Smart Computing solutions, the Apple product family combines servers and
server hardware (at Apple) with mid-client hardware (iTunes software for Macs and PCs) with end
client hardware (the iPod or the iPhone) running downloadable, reprogrammable software, linked
through generally available network systems. Apple has been adding analytics of user purchases of
music and apps to make recommendations for other music or apps that the user might want. Apple’s
success where other providers of digital music downloads have failed is largely due to its providing
a total solution with a well-designed and constantly improving user experience. While Apple
initially owned and controlled all aspects of the iPod/iTunes (with iTunes only deployed on Apple
Mac computers), it has relaxed that control in non-critical areas, such as allowing iTunes to run on
Microsoft Windows computers, allowing independent software developers to offer apps at the App
Store, and allowing different wireless carriers to offer iPhones.
Of course, Apple is primarily a consumer-focused tech vendor, with no need for vertical industry
capabilities. It hasn’t had to face the challenges that a vendor creating Smart Computing solutions
for the business world will face. Still, Apple provides a model for creating a Smart Computing
solution that pulls together technologies from multiple domains and packages that solution in a way
that wins buyer acceptance.
· IBM has a lead position but needs to add apps and overcome its consulting services bias.
IBM already owns many of the hardware, software, and network elements of Smart Computing.
Its Smarter Planet marketing initiative shows that it understands that balance sheet issues in key
verticals are the biggest opportunities, and it has built enough success cases in these verticals to
speak credibly to business executives in these industries. It is creating software frameworks in
these industries that combine IBM and partner products to provide solutions. But it will need
to acquire client device technologies and key software apps if it is to offer Apple-like product
consistency that speaks directly to the client desire for an easy-to-purchase-and-use solution.
· Oracle has many Smart Computing elements but only a nascent vertical strategy. Oracle
has the apps that IBM lacks, comparable analytics and SOA platform software products, and
a presence in hardware through its acquisition of Sun Microsystems. And, it has the potential
to offer separately branded vertical solutions as the result of acquiring many vendors and
sustaining their brands through its Applications Unlimited policy. Sun puts Oracle at the wrong
end of the hardware market, selling servers when virtualization and cloud computing is eroding
demand, and so it lacks the client devices needed for true Smart Computing solutions. Most
importantly, its vertical strategy is still stuck in verticals 2.0; it has made no clear commitment
to making its different brands truly vertical brands, and its understanding of verticals is shallow.
· Microsoft has to choose between competing with Apple in consumer tech or IBM in
business. Microsoft has the size and resources to compete with IBM and Oracle in Smart
Computing, as well as the analytical, SOA platform, and application software products to
provide credible solutions in services industries like education and professional services. Its
Xbox and Zune offerings, as well as years of working with PC and smartphone vendors, provide
experience in client devices. But Microsoft derives too much of its revenues from horizontal
technologies like Windows and Office to be able to pivot toward primarily vertical solutions.
Above all, Microsoft is as much a consumer tech vendor as it is a business tech vendor. Given
the competition that it faces in the consumer market from Apple and Google (among others),
Microsoft may not have the appetite to compete with IBM and Oracle in vertical Smart
Computing solutions outside of industries like professional services, where the dominant small
and medium-size businesses (SMBs) are natural targets for Microsoft Dynamics and Office-
based products.
· GE and Siemens use their strength in key verticals to move into Smart Computing. Most
people don’t think of GE or Siemens as IT vendors that compete in the same space as IBM,
Oracle, or Microsoft. That is changing. In key verticals like healthcare, GE and Siemens are
already providing core software products, as well as selling core medical technology equipment
that represents some of the core assets of hospitals and healthcare providers. Both are major
providers of power generation and transportation equipment (where they are leaders in adding
the sensors and analytical software for tracking the performance and condition of these assets).
They still need to add the software applications and analytical tools to provide full Smart
Computing solutions and may never be a factor in most industries. But in healthcare, utilities,
and transportation — three verticals that IBM has targeted in its Smarter Planet initiative —
they can give IBM a run for its money.
· SAP is stuck in verticals 2.0 and process automation. Like Oracle, SAP has both the apps
and the analytical engines to put into Smart Computing solutions. However, its SOA platform
lags behind IBM’s and Oracle’s, and it lacks any hardware products to evolve into client device
offerings. Perhaps most critically, it seems to be stuck in a verticals 2.0 mindset. The launch
of its SAP Business Suite 7 was all about the meta-processes common to all industries, with
analytics applied to improve the efficiency and effectiveness of processes. SAP, of course,
understands that balance sheet issues do matter, and its product suite can be, and has been, used
to address these issues in some industries. But perhaps because SAP’s product mindset is shaped
by the manufacturing and other goods-producing industries like wholesale that are all about
processes, it does not see how critical balance sheet issues are in the services industries where
demand for verticals 3.0 will be greatest.
· Cisco has aspirations to play in Smart Computing but has to overcome its network gear
bias. Network technologies are key elements of Smart Computing solutions, and Cisco has
been making acquisitions to strengthen its offerings in the enterprise and other markets. It also
has been developing software and client hardware capabilities that in principle could allow it
to provide awareness and analytical capabilities at the edge of the network in areas like video
analysis. But Cisco is just starting to develop a vertical industry focus, still has major gaps in its
software offering, and still tends to think and act like a network gear vendor rather than a Smart
Computing solution provider.
· HP and EMC are still in the horizontal tech space and slow to embrace Smart Computing.
While both vendors have made acquisitions of software vendors, and HP has added EDS,
neither has the portfolio of analytical tools and applications needed to offer smart solutions
on their own. EDS does bring HP some experience in smart meter systems and other industry
solutions, and HP has IT asset management tools and other middleware capabilities that put it
slightly ahead of EMC of Smart Computing capabilities. But at this stage, both look more likely
to try to dominate their horizontal technology domains then to play in vertical solutions.
· Dell makes a small move into verticals. Dell is still primarily a horizontal tech vendor,
concentrated in the commoditizing hardware categories of PCs and servers. Combining low
costs with high quality has historically allowed Dell to gain share in these markets, and that will
probably continue to be its play. However, its acquisition of Perot Systems and its partnership
with Wal-Mart to sell healthcare management systems to doctors are signs that it could become
a factor in creating vertical solutions in industries like healthcare (or in retail) that have many
small businesses.
Figure 16 Which Vendors Are Best Positioned For Success In Smart Computing?
Vertical industry
products
Horizontal
products
R ec o mme n d ati o n s
actions, and auditing all these steps, which can be embedded in Smart Computing solutions
provided by other vendors. Alternatively, vendors can push upward into the apps and
analytics space and create microvertical solutions.
· Computer and communications hardware vendor strategists will mostly stay horizontal.
While software will become more vertical, hardware will tend to stay horizontal. There are
exceptions, such as retail point-of-sale products or physician hardware systems, and there
will be even more examples in the future. But the most value will come from providing the
full industry solution, not pieces, so hardware vendors that want to provide microvertical
solutions will need to beef up their software portfolio. Otherwise, vendors that want to stay
in these stagnating horizontal product categories will need to be low-cost providers that
drive consolidation to grow.
· Telecommunication service vendor strategists need to partner. Wireless, broadband
telecom services will be key parts of many smart industry solutions. In some cases, there will
be opportunities for telcos to build special networks for machine-to-machine interactions,
especially in industries with unique security or deployment challenges. But, in general,
telecom services will be the most commoditized parts of the solutions, with limited room
for vendors to differentiate themselves. And telcos have not had a stellar track record as
systems integrators. Their best course of action is to partner with the top providers of Smart
Computing solutions, contributing the network component, delivering leads, and sharing in
the profits.
· Chip vendors should stay horizontal, with selective forays into vertical products. Of all
the tech vendor categories, chip vendors are the most likely to remain horizontal vendors.
Similarly, the strategies of chip vendors are the least likely to change. But even here, there
will be selective opportunities for chip vendors like Intel, Texas Instruments, and Qualcomm
to create vertical variations on their core products and sell these as end client devices. Texas
Instruments and Qualcomm have already had successes using this approach. These vertical
chips results may represent no more than 10% of the volume that these vendors sell, but
they could well be the most profitable segment.
W H AT I T M E A N S
· Other countries will follow the US with shorter, longer, or no lags. The US is still
the largest IT market in the world, the most advanced, and has the widest adoption of
technology across all industries and sizes of companies. However, there are other countries
that have similar levels of technology adoption, and, in specific areas, they could well
take the lead over the US. We have called these countries the Tech Twelve (they include,
in addition to the US, Australia, Canada, Denmark, Finland, Israel, The Netherlands, New
Zealand, Singapore, Sweden, Switzerland, and the UK), because their ratios of IT purchases
to GDP are 3% or more.19 China, Korea, Japan, France, and Germany are also likely to be early
adopters of certain elements of Smart Computing, with Brazil, Chile, India, and South Africa
being fast followers.
· The journey to Smart Computing will be long, with both hype and disappointments.
Today, we are at the start of a cycle, where the market is filled more with custom-built
solutions for single clients, partial solutions, and vendor promises of what their offerings will
be. Inevitably, there is a fair amount of marketing hype, with more to come. Even as Smart
Computing offerings become more productized and more substantive, the fine print about
the need for process and organizational change to accompany a new technology will get lost
in client desires for a simple technology fix and vendor willingness to accommodate those
desires. Think back to the hype versus the reality of ERP solutions in 1992 to 1995, or Internet
technologies from 1996 to 2000. But, the realities of benefits from Smart Computing will in
time come close to matching the promises.
· The growth cycle will end at some point, probably around 2017. Nothing lasts forever,
and that will certainly be true of Smart Computing. The period of growth in IT market
revenues that we have predicted in this report will run for several years, but will come to an
end by 2017 at the latest, and earlier if there is another global recession in the next three to
five years.
Endnotes
1
Since I first proposed this thesis five years ago, I have often been asked why the periods of tech innovation
and growth and the periods of tech digestion and refinement each seem to be about eight years long. I think
the answer lies in the tenure and mindset of CEOs, who are the ones who set the strategic and investment
priorities for their companies. As it turns out, the average tenure of a CEO (depending on the survey and
when it was conducted) is eight to 10 years. And most CEOs tend to be herd creatures, responding to and
following the key strategic moves of other CEOs. I think what happens is that at the start of a tech cycle,
CEOs at a few companies learn about this new technology from innovative colleagues in their company,
from a consultant, or from a key technology vendor. The success stories of these firms hit the media and
spread interest to other CEOs. However, the followers don’t see the same benefits as the early adopters, or
they miss the hidden process changes and organizational changes that the successful implementers had
done. So, these CEOs shift from being technophiles, in love with the promise of the new technology, to
techno-skeptics, unwilling to invest in new technology unless it has a hard, provable return on investment.
2
The mainframe computer got its start with pioneering products like ENIAC in 1946, Remington-Rand’s
UNIVAC I (delivered to the US Census Bureau in 1951), and the IBM 701 in 1953 and IBM 650 in 1954.
1960 ushered in the first DEC minicomputer, the first modem, and the COBOL programming language.
Over the next nine years, IBM released the IBM 1400 series mainframe (1961), the first disk storage
unit (1961), and especially the IBM System/360 (1964), which gave a major push to mainframe adoption.
Control Data released the CDC 6600 supercomputer in 1964, with DEC, Hewlett-Packard, and Data
General offering their own minicomputers in 1965, 1966, and 1969, respectively. The era of innovation
closed in 1969 with the release of the UNIX operating system.
3
While SAP had been selling versions of ERP software for the mainframe since the 1970s, and Oracle had
released UNIX-based applications starting in 1987, the SAP R/3 client-server-architecture-based ERP
product caught the attention of business and grew rapidly.
4
IBM owns Cognos and recently acquired SPSS, SAP owns Business Objects, Oracle has BI Enterprise
Edition (formerly Siebel Analytics) and Hyperion Essbase, TIBCO Software acquired Spotfire, and
Microsoft has several small BI vendors. Other leading vendors are private, such as SAS Institute,
Information Builders, and QlikTech.
5
For background information on business intelligence vendors as well as data warehouse vendors, see the
July 31, 2008, “The Forrester Wave™: Enterprise Business Intelligence Platforms, Q3 2008” report and see
the February 6, 2009, “The Forrester Wave™: Enterprise Data Warehousing Platforms, Q1 2009” report.
6
Digital Business Architecture was the conceptual framework that Forrester used to pull together new
developments in four domains of technology that were helping make it systems more adaptable, more
flexible, more extended, and more interconnected, and thus better able to match business needs and
requirements See the November 7, 2005, “Digital Business Architecture: IT Foundation For Business
Flexibility” report.
7
Not all awareness technologies are new of course. Special areas of application have used technologies like
telemetry for years. Even more important are the awareness technologies at Internet sites of what customers
are doing in real-time, using clicksteam analysis of page views, etc. In fact, many of the analytical tools
that exist and will be deployed against data coming from sensors, RFID, video analysis, GPS locators, etc.,
originated with analysis of clickstream activity.
8
The following report provides more insight into the role of rules engines in combination with business
intelligence and business process management in creating more optimal business results. See the May 14,
2008, “How The Convergence Of Business Rules, BPM, And BI Will Drive Business Optimization” report.
9
The September 2009 issue of The Economist magazine in the “Technology Quarterly” section provided this
profile of the future of satellite navigation devices. Source: “Rational Consumer: The Road Ahead” The
Economist, September 3, 2009 (http://www.economist.com/search/displaystory.cfm?story_id=14299710).
10
Forrester Vice President and Research Director Connie Moore and Vice President and Principal Analyst
John R. Rymer introduced the concept of Dynamic Business Applications at Forrester IT Forum
conferences in 2007 and in their report in September 2007. See the September 24, 2007, “The Dynamic
Business Applications Imperative” report.
11
We first wrote about Accruent in a report on four vendors that had adapted their software solutions to
address critical balance sheet issues in vertical industries. See the March 12, 2008, “Small App Vendors:
Optimize Business Results” report.
12
Forrester has initiated a series of reports on smart grids and their IT implications. See the October 19,
2009, “Smart Grid Technologies: Coming To A Utility Near You” report. The Economist has written two
good summaries of trends in the smart grid/smart meter market. Source: “Smart grids: Wiser wires,” The
Economist, October 8, 2009 (http://www.economist.com/displaystory.cfm?story_id=14586006&CFID=865
49758&CFTOKEN=72102578#); and “An internet for Electricity,” The Economist, June 4, 2009 (http://www.
economist.com/search/displaystory.cfm?story_id=E1_TPSNVRQP&CFID=87559706&CFTOKEN=869729
83&source=login_payBarrier).
13
The Economist magazine has also written a comprehensive review of healthcare information technology,
with a focus on patient records management systems. Source: “Medicine goes digital,” The Economist, April
16, 2009 (http://www.economist.com/specialreports/displaystory.cfm?story_id=E1_TPQPSJJD).
14
One early profile of these types of solutions comes from The Wall Street Journal. Source: Michael Totty,
“Smart Roads. Smart Bridges. Smart Grids,” The Wall Street Journal, February 17, 2009 (http://online.wsj.
com/article/SB123447510631779255.html)
15
ERP vendors like SAP and Oracle are starting to focus on a couple of dozen verticals each while retaining
presences in others. Smaller ERP vendors have been concentrating on a narrower set of verticals. See
the November 2, 2009, “The State Of ERP 2009: Market Forces Drive Specialization, Consolidation, And
Innovation” report.
16
The data for this statement comes from Forrester’s report on US enterprise versus SMB IT budgets in 2009,
which had our projections for IT spending by industry. See the May 7, 2009, “US Enterprise Versus SMB IT
Budgets In 2009” report.
17
The data for this statement comes from our report on the IT vendor opportunities in the economic stimulus
programs in the US and other economies. See the July 7, 2009, “There’s Gold For Vendors In Stimulus
Packages” report.
18
Forrester has published a long series of documents on the Information Workplace market and vendor
landscape. See the March 28, 2008, “Information Workplace Platform Vendors Light Up The World Of
Work” report.
19
We introduced the Tech Twelve concept in our January 2009 report on the global IT market. See the
January 12, 2009, “Global IT Market Outlook: 2009” report.
55157