At the March meeting, the Object Management Group (OMG) issued a Request for Proposals (RFP) for a Value Delivery Metamodel specification. The RFP is based on the concept of value chain models for business analysis and planning. This specification should link strategic planning to operational activities, including SOA, and provide an important tool for enterprise transformation. I've talked about the importance of value chain modeling in past blogs, see Value Chain Modeling Is Essential for SOA Management.
In the OMG RFP, the term "value delivery" was chosen to avoid conflicts with current techniques and to recognize that an enterprise will need to manage multiple value chains and their relationships when more detailed models are developed. Paul Harmon provided some useful background on value chains and related techniques in his recent BPTrends article Value Chains and Other Processes. In another BPTrends article, George Brown describes the evolution of these concepts in Value Chains, Value Streams, Value Nets, and Value Delivery Chains.
These articles highlight the diversity of techniques that have emerged from the basic concepts presented by Michael Porter in his1985 book, Competitive Advantage: Creating and Sustaining Superior Performance. The amount of activity over the last 24 years highlights the importance of the fundamental concepts. The Value Delivery Metamodel is important for conventional enterprise management as well as SOA.
I believe the diversity of approaches to some extent reflects the need for different abstractions of the underlying concepts in order to provide manageable models with limited tools. A metamodel is essentially a specification of a computer-based modeling language. The OMG process should bring together a group of submitters with different perspectives to define the basic concepts and resolve variations in terminology. The resulting specification will enable the development of interoperable modeling tools, supporting different business viewpoints, to more effectively capture and manage value delivery models.
The OMG has also organized another forum for discussion of such issues. The Business Architecture Work Group meets in conjunction with the quarterly OMG meetings, and involves business architects as well as information technology people interested in developing better modeling tools for business planning, analysis, design and governance. I hope that the RFP along with the BAWG will bring together a cross-section of perspectives for a new generation of business modeling capabilities.
About 9 months ago I discussed the potential development of an OMG (Object Management Group) standard for modeling value chains: A Value Chain Modeling Standard?. Progress has been slow, but over the past year we have gained additional insights. OMG issued an RFI and received four responses: Cordys, Hewlett Packard, Value Chain Group and VUA Amsterdam. An operating level view of value chains would provide important business insights for a conventional enterprise, but such a view is essential to effective business design and management in a Service Oriented Architecture (SOA). A new draft RFP for "Value Delivery Modeling" has been prepared for discussion at the next OMG meeting in March.
Paul Harmon provides perspective on the evolution of the value chain and related concepts in Value Chains and Other Processes. The value chain models needed for SOA are not the executive-level models of Michael Porter, but are more operationally focused models of the activities involved in the delivery of customer value.
A value chain will represent a dependency network of activities needed to produce customer value for a line of business or similar products. The dependencies correspond to the flow of work products from activities where they are produced to activities where they are needed as inputs. The activities reflect the application of business capabilities at an operating level. Essentially the value chain is a use-case of business capabilities.
An activity may be linked to a capability type that defines the resources, skills, facilities and intellectual capital needed to perform the activity as well as similar activities. In the actual operation of the value chain, a capability type may be filled by a particular organizational unit that possesses the needed capability, or there may be alternative organizational units that can perform the needed activity.
In a SOA, the capability type might be described as a service specification, referring to the activities or "services" that a particular organization, or "service unit" offers to perform. The difference between a capability and a service unit is that a service unit has a well-defined interface that enables it to be used in different business contexts, particularly, different lines of business.
Paul Harmon, in his book, Business Process Change: A Guide for Business Managers and BPM and Six Sigma Professionals, (page 71) describes the difficulty of deciding how many value chains an enterprise should have, in other words, what level of granularity and product differentiation should be reflected in the value chain model. SOA provides an answer: the activities in the value chain should be those performed by service units. Different products can share a value chain as long as the activities don't require participation by different service units. Of course the specific actions performed by a service unit may vary significantly based on the product specifications.
Value chain models are essential to SOA because they define the service units that are engaged to produce customer value for each line of business. Conversely, they define the contribution a service unit makes to each line of business in which it participates. This provides the context for evaluation of service unit performance and the context for considering the merit of investing in service unit improvement. It also provides a means to configure existing capabilities to address a new market opportunity. I explored these ideas in more depth in my book, Building the Agile Enterprise with SOA, BPM and MBM.
Value chain models should incorporate the full cost of activities along with activity durations to support analysis of value delivery cost and timeliness. Value chains also can provide insights on the consequences of capability failures. Linkage to internal value chains for supporting services can provide insights on overhead costs and identification of non-essential business capabilities.
An Object Management Group Request for Proposals (RFP) has been drafted by Henk de Man of Cordys and myself to develop a standard for enterprise value chain modeling. There is opposition to moving forward on this. There seems to be a concern that there are too many different views on value chain modeling and analysis to develop a standard. At the same time, there seems to be a belief that there would be no value in a standard.
Variations have been proposed and discussed for the past 20 years, ever since Michael Porter published his book Competitive Advantage: Creating and Sustaining Superior Performance in 1985. A standard would provide a foundation for moving forward and the ability to develop tools and more robust models that would ultimately reconcile the value chain with other business models. I believe current value chain analysis is limited by a lack of tools to manage the complexity of a detailed value chain.
Porter's value chain describes the creation of customer value through high-level phases of product development and delivery. The Value Chain Group describes creation of value through a dependency network where each "process element" receives inputs that are needed to produce value, and produces value outputs that are input to other process elements to create their value. The cumulative result is value flows leading to customer value. A robust value chain model based on this dependency network would support analysis of the cost, timeliness, quality and risks of value creation.
An enterprise may have different products with different value chain networks that use some of the same business capabilities. These business capabilities and the business activities that manage them can be characterized as service units in a Service Oriented Architecture (SOA). Service units produce value by execution of business processes-business processes define the flow of control while the value chain dependency network defines the flow of work products, in other words, customer value.
Some people think the term value chain implies a single thread of operations-some prefer value network. Others discuss value streams. Stabell and Fjeldstad propose value chains, value shops and value networks as three different enterprise configurations for producing value. I believe there are fundamental concepts that can unify these various approaches and could provide a valuable business modeling capability.
A unified model has the potential to link strategic planning using high-level abstractions for executives with operational details for analysts and product planners. A model can provide accountability and a basis for targeting performance improvements. It also has the potential to align business process models and SOA models with the creation of customer value.
A standard can enable the marketplace. It enables users to develop models with one tool and move their models to a new tool when a more advanced capability is offered. A standard enables small vendors to compete and advance the state of the art.
Should there be a modeling standard for value creation? What should be the essential elements? Issue of an OMG RFP starts the process of developing consensus.