The Death of CRM
It may seem odd to the casual reader that the same man who
founded the company that dominates the Customer Relationship Management (CRM)
space would now be calling that market dead. Yet that is exactly what Tom
Siebel of Siebel Systems did at DCI's CRM event in New
York. Instead
of promoting CRM, Siebel articulated a new vision for the software industry in
what may have been the most important speech of his career.
"There's no market for CRM," said Siebel.
"It's not there. There are a lot of companies trying to get into the space
and they're building generic sales force automation, customer service, or
marketing software. But they are going after a market that doesn't exist."
Having the man most closely identified with CRM declaring
its demise may seem a little like Henry Ford predicting the death of the
automotive business, or Frank Purdue announcing the end of the chicken
business. Yet Tom Siebel's prediction is something all together different.
Customer relationship management software is used to
automate and manage marketing, sales, and customer service information. It is
considered front-office software because it typically involves interactions
that the customer can see. As long as organizations continue to market, sell,
and service customers, they will require some kind of software to do it. What
Siebel is predicting is the direction the software industry will take to meet
this continuing need.
That CRM is dead will come as a surprise to destinationCRM, searchCRM, CRM
Forum, CRM Plus Weekly and all the other publications
devoted to promoting the market, not to mention the hundreds of software
companies that are betting their resources on creating products for it. To these groups of people, the CRM market is
like some kind of tide - when it returns it will lift all boats equally. This flies in the face of many signals that
the CRM market is not ever coming back.
From 1998 to 2000 the Customer Relationship Management (CRM)
market grew over 50% a year. Then in 2001, new license sales
fell 6% and is projected to fall another 2% this year, according to
Gartner Dataquest. The larger Enterprise Resource Planning Market (ERP) has
faired worse. In 1998 the ERP market enjoyed its greatest success as
new-license revenue grew 19%. ERP new-license revenue then fell 27% in 2001 and
is expected to fall 12% in 2002.
In Siebel's vision IT organizations will no longer buy
generic CRM software to adapt to their internal business processes. Instead,
they will buy software that incorporates industry best practices and processes
and then adapt their way of doing business to the software. Like most
revolutionary ideas, business process software may seem simple on the surface,
but it has far reaching consequences.
According to Siebel, the CRM market is dead, not because customers
no longer need it, but because it will be subsumed within a new layer of
software that offers much greater value, just as CRM once subsumed the Sales
Force Automation (SFA) and contact center markets.
If Siebel's vision is correct, the end of the CRM market
also means the end of the Enterprise Relationship Planning market, the Supply
Chain and Human Resources markets, and all other current categories of
application software.
Tom Siebel As Visionary
Those who are taken by surprise by his pronouncements forget
that Tom has a dual role as CEO and
industry visionary. As early as 1994, before Siebel Systems even had product,
Siebel used DCI as a platform to change the direction the business was headed.
Back then he also predicted that most of the current suppliers where headed out
of business because they didn't meet customer needs. At that time the industry
leaders were Brock Control Systems, Saratoga Systems, Aurum, and Salesoft: companies all but forgotten today.
In his keynote speech, Siebel neither
looked, acted or sounded like someone under a dark cloud. He appeared
confident, compelling, and knowledgeable. He struck me as someone who had
finally found the ground under his feet and knew what direction he wanted to
take his organization. In a one-on-one interview afterwards, Siebel was a man
of convictions. Among those convictions:
·
The CRM market is dead.
·
A new generation of application software is
about to be built, which, for lack of a better name, can be called business
processes applications.
·
The technology that enables this new generation
of applications is Web Services,
·
To be competitive, future application packages
will be not generic but industry specific.
·
The change in technology and functionality will
drive most existing software companies out of business and change dramatically
the shape of the industry.
What Are Business Process Applications?
The first question these ideas raise is, "What exactly is a
business process?" Simply put, business
processes are a set of activities organized in some pattern to accomplish a
business objective. This could be as simple as scheduling a meeting, or as
complex as launching a new product. The pattern activities take do not
necessarily have to be the most cost efficient or be accomplished in the least
amount of time.
A pattern that does measurably better than other patterns in
achieving an objective can be considered a best practice. Best practices are
also defined as proven methods of consistently achieving business objectives.
For a concrete example of a business process, imagine the
activities that must take place in order for a new employee to become
productive. In most business settings today this process is set in motion on
the first day a new employee reports to work and sometimes is not achieved for
many weeks. To complete the process and secure the objective, the following,
among other tasks, must occur:
·
The telephone switch must be touched to give the
new employee a phone extension.
·
The employee's contact information, including the
new telephone number, must be added to the payroll system.
·
If the employee is expected to work with Siebel
applications, territories and accounts must be assigned.
·
An email must be sent to the printer for business cards to be ordered.
The process can be considered a process because it is the
same for every new employee. It can also achieves a
discrete business objective because the employee is not fully productive until
the process is finished. Most of these tasks are already automated. Laid
end-to-end electronically this business objective could theoretically be
realized in a matter of minutes. In practice it takes a human agent to ensure
all the steps are completed.
Extending the business process represents a new level of
computing abstraction. Rather then buying applications in the future, IT
organizations will buy products with names such as service provisioning in
communications, formulary management in pharmaceuticals, and trade promotions
and funds management in consumer goods.
Why business processes are compelling to customers is easy
to see. They can be cost justified in terms of the resources they save an
organization compared with current processes and in terms of the value they add
- such as making new employees productive as soon as possible.
Web services
The technology that enables software to enter this new arena
is called web services. Web services promise to dramatically reduce the cost
and complexity of application integration and make cross application business
processes possible. No one organization controls the standards on which this
technology is based, although many are making a heartfelt attempt.
Since no single group can define what web services are, the
term itself is in danger of becoming amorphous. Web services are currently
defined by a bewildering array of different, and sometimes conflicting,
standards promoted by an equality bewildering alliance of organizations and
industry groups. Important among the standards at this writing are .Net,
BPEL4WS, J2EE, UAN, UML, WSDL, WSFL, and XSD (see the Emerging Standards
Section below), Some of thesemust inevitably drop by
the way.
With such an unsettled definition, many companies are waving
the web services flag in hopes of benefiting their short-term positions without
making any real change in their technology. This threatens to create more
confusion in the market and to further delay the eventual adoption of any
standard.
To give the definition of web services some teeth, any
number of public bodies and industry coalitions are publishing their
specifications with the hope of seeing them be adopted as the standards.
Specifications become standards in one of two ways, each with its advantages
and disadvantages.
Public standards bodies tend to favor an open specifications
process, where any organization, academic or individual, is invited to
participate. Specifications are discussed in public via drafts and open
committee meetings. As a result public bodies tend to take an extremely long
time to agree on anything, and by the time they do reach agreement, the market
has often trumped them by adopting a different standard.
In response to the shortcomings of the public standards
approach, many vendors release their own specifications in hopes of making them
the standard. In the closed-door world of vendor-defined specifications, the
scope and direction is kept confidential until published. Although this can
make for a much more timely result, lacking the wider consensus these
specifications are rarely accepted by the entire industry and are especially
scorned by the business competitors of the enterprise proposing the standard.
While there is no question that BPEL4WS is a standard; the
question of whether it will emerge as the standard remains open. Some of other
standards vying with it include BPML, the Business Process Modeling Language,
BPSS, the Business Process Specification Schema, which
is part of ebXML, WSCI, the Web Service Choreography
Interface, and XPDL, the XML Processing Description Language. Because of the
high stakes, some companies seek to hedge their bets by supporting multiple
standards. For example, BEA is a charter member of both BPEL4WS and WSCI.
How It All Works Together
Taken together, web services provide a standardized way to
build business process applications requiring information from multiple
applications, even if they run on different operating systems and hardware
platforms.
Returning to the example of the new employee business
process application, the new employee enters his or her identification
information - name, address, social security
number in a single screen which
sends a simple object access protocol (SOAP - part of .NET) query to the
telephone switch requesting its WSDL file.
The switch the sends the WSDL file back to the new employee
application describing the web services it can perform. The new employee
application parses the WSDL file, and selects the web service to create a new
phone number and sends a SOAP command (with the new employee's name) to the
switch. The switch verifies the security of the application and fulfills the
request by creating a new phone extension.
The switch sends the new telephone number back in XML format
to the new employee application, which then sends a SOAP query to the Human
Relations System requesting its WSDL file and so on until the process is
complete.
The Promise
It is no secret that the enterprise applications software
business is in something of a funk. The poor economy has contributed to the
worst performance the industry has seen in recent years. The software business
is also responding to its own product cycle; despite many attempts to introduce
new technologies, the products that are the market leaders today were first
introduced in the early-to-mid nineties. The growth of these products has slowed
as markets have become saturated and as the technology at their hearts ages.
For the industry to move forward it must go though a period
of regeneration similar to that which a forest experiences after a fire. In the
metaphor of the forest, the fire that swept thought the software industry was
the dot com which promised to turn everything it touched to gold, but instead
turned it to ash. Just as a healthier forest grows out of the ashes, a
healthier software industry can grow out of what was learned through the dot
com experience. What the industry needs to return it to sustainable growth is
leadership and a new direction in which to head. Given his enormous experience
and impressive accomplishments, Tom Siebel may be providing the industry with
the direction it needs to enter into a new era of prosperity.
How this vision of software will effect players in the
industry, including Siebel Systems, will be the topic of the next edition of
the Siebel Observer; Tom Siebel's Vision for the Software Industry - Part 2.
Emerging Standards
Although it is something most of us would prefer not to do,
it is helpful to examine some of these specifications and standards in more
depth.
.Net
.NET is a set of Microsoft software technologies for
connecting systems and devices together using XML. Microsoft sells the
developer tools, client applications XML Web services, and
servers necessary to participate in .NET. Microsoft and others are developing a
core set of XML Web services - from authentication to calendaring --
that can be combined with other XML services or used with applications.
BPEL4WS
BPEL4WS (the Business Process Execution Language for Web
Services) is designed to specify business process built with web services. In
the web services protocol stack, BPEL4WS is a layer
on top of Web Services Description Language (WSDL). It uses WSDL to specify
actions that should take place, and to describe the web services needed to
complete the process. BPEL4WS is the successor of two previously competing
standards: Web Services Flow Language(WSFL) from IBM, and Microsoft's XML
business process language in the BizTalk Server (XLANG). BPEL4WS is a
joint development of BEA, IBM, and Microsoft.
J2EE
J2EE, (the Java 2 Platform, Enterprise Edition) is designed
as a single standard that can sit on top of a wide range of existing enterprise
systems, including database management systems, transaction
monitors, and directory services.. It also defines a standard for developing
enterprise applications as sets of module components. J2EE takes advantage of
many features of the Java, such as portability, JDBCTM API for database access,
and CORBA for interaction with existing applications. The J2EE standard includes
specifications and compliance tests to ensure portability of applications across
the wide range of platforms.
Sun Microsystems invented the Java programming
language and pioneered its use in enterprise services.
UAN
Launched by Siebel Systems, the UAN
(the Universal Applications Network)
is a standards-based approach to the design and development
of cross-application business processes. It is Siebel's answer to the standard. The Universal
Application Network solution comprises three basic components: (1) a library of
business processes, common object models, and transformation maps; (2) a
business process design tool; and(3) an integration server to
coordinate communication.
UML
UML (the Unified Modeling Language) is designed as a
standard way for software developers to specify, visualize, and document programs
before coding. UML is similar in concept
to the blueprints, site maps, and physical models architects use to plan a
building, UML allows applications to be modeled as a means of managing their complexity,
promoting code reuse, and minimizing future maintenance. UML is promoted by the
Object Management Group, a public standards body.
WSDL
WSDL (the Web Services Description Language) is build on
SOAP, the standard designed to address the interoperability problems between
Web Services written in different languages,
running on different platforms, and connected only by the Internet.
WSDL is an XML format for describing Web Services as a set
of messages containing either programs or data. The message content is first
described abstractly, and then bound to a
specific network protocol and format to define an endpoint.Related endpoints are combined into a
complete Web Services of programs and data. The WSDL standard is promoted by the World Wide Web
consortium (W3C).
XSD
XSD (the XML Schema Definition) specifies how to formally
describe the elements in an Extensible Markup Language (XML) document. This
description can be used to verify each item of
content in a document. For example, in a document describing a Web site, you
would define a Web site element and each Web page element as a set of tags.
In contrast to the earlier languages, XSD is written in XML.
Some detractors claim the language is unnecessarily complex. XSD is promoted
by the World Wide Web Consortium (W3C).
XSLT
XSLT (the Extensible Stylesheet
Language Transformations) is a language for transforming XML documents into
other XML documents With XSLT you are not restricted to present your
information in the same order you created it, nor are you required to present
all of your information. XSLT allow XML sources to be transversed
multiple times if that information need to be reused more than once in a single
result. XSLT is promoted by the World Wide Web Consortium (W3C).
