Assessing and achieving alignment between an organization's strategies and its IT/business functions has long been recognized as a critically important question. This paper reports on a project that seeks to overturn established management orthodoxy by establising that strategies can be adequately modeled using conceptual modeling notations and that methodological and tool support can be provided for the task of assessing and achieving alignment between the strategies of an organization and its service o erings. A key element of this enterprise has been the design of SML - the Strategy Modeling Language. This paper presents an interim report from this project that describes how a notation inspired by i* has been used to obtain the diagrammatic modeling component of SML, and how i*-like notions have been used to represent strategy decomposition (required to be able to re ne strategies to a level where there is an onotlogical match between the languages used to describe strategies and services). We also comment on how i*-like notions would play a greater role in this project, as a complete model of the enterprise context is brought to bear on the alignment exercise. We provide a brief illustration, and a description of the toolkit implemented on the Eclipse platform.
A large body of literature, spanning several decades, has highlighted the critical
importance of strategic alignment to the management and information systems
communities. Yet there is also considerable pessimism about the prospects for
solving this problem using computer-mediated tools and methodologies. This
? Funding of this research was provided by the Smart Services CRC Initiative
http://www.smartservicescrc.com.au/
paper presents an interim report from a project on strategic service alignment,
within the Australian Cooperative Research Centre for Smart Services. We have
approached the problem by rst devising a language for modeling organizational
strategy in a manner amenable to machine processing (the Strategy Modeling
Language - SML), then de ning a high-level business service modeling language
(the Business Service Representation Language - BSRL) [
In this paper, we focus on the modeling of strategy. The Strategy Modeling
Language (SML) has been designed to include both a textual and diagrammatic
interface for modeling and visualizing the strategic landscape of an organization.
We describe how the i* framework [
Our current work has mainly focused on alignment within a single enterprise context, but is currently being extended to address cross-enterprise value chains. The capabilities o ered by i* SD diagrams to model enterprise structure via actor/dependency models is critical for modeling strategy in a cross-enterprise context, and represents an important direction for future development.
Current work on strategic alignment looks at how strategies can be speci ed in
relation with other artifacts such as actors, business processes, resources. The
key approaches in this space (outside of i*) include e3 Forces [
Modeling the decomposition of business strategy plays a vital role in the landscape of enterprise strategic alignment. To be able to align business services to organization's strategy, we need to have a rich model that represents the organization's strategy to work with. i* is the main the source of ideas that particularly in uences our vision on strategy modeling. Our initial idea in this
direction explores the notion of contextual consistency in goal decomposition
[
In this section, we present how we extend i* to cope with strategy decomposition (Subsection 3.2), which is illustrated by an example (Subsection 3.1).
Let us consider an example that describes a multi-national book-seller whose management decides to provide their services via the Internet. The management sets out the strategy for the book-seller is to become the market leader amongst book-sellers in Australia. As the business of selling books involves marketing, optimizing operating costs and dealing with book suppliers, the main strategy is then broken down into three more concrete component strategies (i) To rst gain market-share in New Zealand, then use United Kingdom market credibility to enter Australia (ii) To minimize operating costs (iii) To manage supplier relationships by providing purchase volume guarantees, and fast payment against invoices. This process can be carried out until the management reaches a set of strategies that are concrete enough to map to business services or business processes3 that operationalize them.
Figure 1 gives the decomposition hierarchy of strategy for the book-seller.
Note that each strategy is pre xed by a string followed by a colon that is in turn
followed by textual description of the strategy being represented. The pre x is
actually a concatenation of an abbreviation and a number. The former denotes
the strategy type and the later signi es the hierarchical branch at which the
strategy being represented is. To reason on strategy decomposition more e
ectively, we di erentiate three types of strategy: business plan, functional goal and
optimization objective [
In our project on strategic alignment, we have been developing a toolkit called ServAlign. This tool manages a repository for strategy and a catalog of business services. In addition, the tool permits diagrammatic representation of strategy and decomposition of strategy. The diagrammatic notation of strategy modeling 3 Business services and processes can be regarded as the main vehicle for the operationalization of an organization's business strategy in a manner akin to the way in which object-oriented components have provided the basis for implementing software requirements in traditional software engineering thinking. The topic of how to align business services and processes to organization's strategy is out of the scope of this paper. used in ServAlign is tailored from i* (see Figure 2). We reuse the i* pictogram of hard goal for our functional goal while introducing additional pictograms for other types of strategy. This addition includes a block arrow for business plan and triangles for optimization objectives (i.e. either maximization or minimization).
Figure 3 is a screenshot of our ServAlign prototype that is implemented as an Eclipse4 plug-in. The Eclipse perspective to the left o ers a tree-view that shows the entire strategy decomposition hierarchy. Each strategy in Figure 1 is now represented as a tree node. An out-zoomed notation is attached as icon to each tree node to visually illustrate the type of the strategy being represented by the tree node. The panel in the middle of the Eclipse window is dedicated to a diagrammatic interface of ServAlign. The diagram shown in Figure 3 shows the decomposition hierarchy of the book-seller's main strategy using the notation shown in Figure 2. Textual description of each strategy is printed below its pictogram. In this diagram, lines represent decomposition links.
The novel ways in which i*-like notions can be deployed in strategic service alignment can provide useful insights to the i* community. We also expect to further leverage i* as we extend our account to cross-enterprise value chains. We are working towards (semi-)automatic, ontology-based strategy decomposition and establishment of strategic service alignment.