Town planning the path to ICT governance

17/08/2007 14:55:00

Governance is increasingly prominent in our ICT consulting assignments. Some events, like mergers, create acute governance pressures. In larger enterprises, especially information-centric ones, ICT governance can become a chronic illness.

ICT governance is a relatively new phenomenon, not listed on board agendas as recently as the 1980s.

It has gained prominence for several reasons. ICT has become pervasive: businesses are riddled with ICT applications and infrastructure, and in some cases the business and its ICT are indistinguishable.

Domestic banking, wagering and share trading operations stop functioning when the systems are down. Retail traders rely on EFTPOS, travel bookings rely on reservation systems.

ICT has become complex and option-rich. In larger enterprises a portfolio of 1000 applications wouldn't be unusual. Accurate diagrams of applications and their interconnections are useless for decision-making purposes, because no business decision maker could understand them without weeks of study.

Presentation layers sit in front of access and security layers protecting integration layers that interact with application services partitioned from data services. There are product choices for every link in the chain.

The world is becoming interconnected, and inter-enterprise connections are common. The user community for a single application extends beyond the enterprise, to customers, suppliers, regulators or intermediaries.

ICT practitioners allege that business management should set business objectives for ICT. But how do business managers know what practical options are available to them? ICT practitioners have a duty to present options to the business, but ICT professionals are increasingly immersed (sometimes sinking) in technological intricacies.

There aren't many ICT professionals who can present technological front lines in business terms. The old-time system analyst, who knew everything about the business and its systems, is critically endangered.

Projects having significant business impacts range from large to enormous, so it's not surprising that company directors and top management are struggling with ICT governance.

There are dozens of ICT governance models on the Internet, from ITIL's comprehensive three volumes through government policy frameworks to those privately merchandised. Generally, these resources are for experts, and to my knowledge, none of these provides a protocol for company directors or top management.

It is possible that we could learn something from another field which, from a governance perspective, is remarkably similar to ICT: that field is town planning.

Town planning is contentious and sometimes disputatious, but it takes place within a well-established governance regime.

Everyone (in industrialized societies) accepts the need for a town plan because the alternative is urban chaos. The components of a modern city are complex, interconnected and interdependent. New residential areas demand services and transport. Industrial developments compete with recreational facilities for scarce land.

There are lots of stakeholders in town planning: developers, administrators, residents, commuters, industries, energy suppliers and retailers have competing interests and requirements. All of the stakeholders expect a degree of certainty, so the plan has to be stable, robust and, most of all, graphically visible. The best town plan in the world is useless if the stakeholders can't understand it.

Changes to urban infrastructure are costly, high-impact, long-lived and are hard to modify once built, so stakeholder engagement in major projects is essential. Decision makers are administrators and elected representatives having limited expertise in town planning disciplines like air quality or transport engineering and therefore have no choice but to rely on advice from experts.

The attributes of town planning can be applied to large-scale ICT planning. Every senior manager would accept the proposition that ICT should be conducted within an overall plan. The complexity of an ICT portfolio in a large bank, telco, insurance company or government department is indisputable.

Stakeholders in major ICT ventures are numerous and extend beyond the enterprise itself. Major (and, sometimes, minor) changes are costly, slow to implement and hard to modify once built. Business managers and directors are not ICT experts, and they must rely on ICT managers and professionals with limited business expertise.

Arranged side by side in a slide pack, the similarities between ICT planning and town planning are remarkable. Company directors, top managers and consulting partners recognize the similarities as soon as they are pointed out.

Town planning works because it has some important features: first and foremost: 80 per cent of the governance structure used in any particular town plan originates outside the town.

In the state of Victoria, for example, every city and shire has a planning framework, and most of the contents are supplied by the State Government, because only the very largest and richest local government bodies could marshal the resources and expertise to create their own governance regime.

It is also because lots of planning regulations are best managed in a regional or national context. (The same is true of ICT).

Second, town planning is governed by a body of laws and regulations, supported by robust enforcement mechanisms. Individuals can try to evade them but only at risk and with difficulty.

It is of interest that another governance area - financial reporting - shares these two characteristics with town planning. All large enterprises in Western economies use accounting standards that are supplied to them by the accounting profession. And financial reporting is buttressed by stringent laws and regulations, of ever-increasing rigour. Company officers and directors who attempt to misrepresent financial positions are likely to be jailed.

Town planning has some neat features that lend themselves to local adaptation. Zoning is a well-known, widely used mechanism for preparing and implementing town plans. It enables authorities to exercise controls over land use without acquiring the land. The controls can apply to all aspects of land use - purpose, density, height, appearance or services.

Zoning first appeared in the USA around 1920. It became widespread in less than 5 years - a remarkably short time. Effective zoning is enforceable and widely accepted, and its success is underpinned by a strong legal and political foundation.

How can we apply the lessons and the well-established disciplines of town planning and financial reporting to ICT governance?

We believe that it's possible to describe a future state of ICT governance that incorporates the successful features of town planning and financial management. This future state has four sets of attributes:

• An internationally accepted set of governance policies, standards and processes • Legislative and regulatory enforcement of compliance • In every large enterprise, a widely publicised, simplified, concise statement of ICT strategies and enterprise architectures • Local regulations for managing coarse-grained architecture domains (zones)

The internationally accepted set of governance policies, standards and processes will probably be sourced from ITIL. Sources in the US Federal Government are another option. The entire framework will be summarised in guides for company directors and senior officers, whose ICT obligations will be incorporated in "Duties and Responsibilities of Company Directors and Officers" published by the Australian Government.

These obligations will centre on the professional disciplines that are required to sign off major ICT ventures. An ICT equivalent of the quantity surveyor will be one of the signatories. Other obligations will refer to consultation and sign-off processes when the enterprise architecture changes, or when important procurement strategies are adopted.

Obligations of directors and officers will be tied to laws and regulations. Corporation law will include appropriate provisions for ICT governance and compliance with international frameworks and standards. There will be an ICT equivalent of enforceable building regulations and a statutory body (like the Victorian Building Commission) that oversees them.

If you think that the last three paragraphs are far-fetched, consider this: in 10 or 20 years' time, almost every ICT service will interact with services outside the originating enterprise. A major ICT failure can already paralyse a company, wreck a supply chain or ground a national transport system. Regulation is inevitable.

The means to produce and publicise concise architectures and strategies has been with us for years. However, ICT practitioners aren't good at getting full attention from their audiences. Town planners regularly engage in community consultation and ICT professionals, especially enterprise architects, will master these skills.

Business communities will have no choice but to listen and engage. If they don't, compliance processes will strangle them.

ICT plans will be expressed in the form of coarse-grained architectures. For compliance purposes, "zones" will comprise groups of closely related architectural domains. International standards will include a standard set of domains and domain decompositions.

These sets already exist in public sources such as TOGAF and the US Government's Federal Enterprise Architecture models. We have found that domain definitions which suit the requirements for abstraction in service design and procurement need to be aggregated for enterprise compliance purposes.

Effective aggregation is industry sector-dependent; zoning for a telco is different to zoning for a bank. The key feature of compliance-based domain definition is that each super-domain can be managed according to a clear set of rules and regulations. For example, the rules for managing data services are quite different to those for managing platform infrastructure services.

Standard templates for domain definition and management will be available within internationally accepted standards. Large enterprises will modify the templates to suit their local needs, in the same way as a shire develops its local zoning regulations. The modified templates will require Board or ministerial approval.

This idealised future state would represent ICT nirvana for enterprise architects and ICT strategists. If the end-state can be defined, it's possible to conduct a gap analysis and to devise a works program to implement the target state. Town planning offers an extensive field for research for students of ICT governance and compliance.

Allen Shatten, proprietor of Lattice, is a specialist in large-scale ICT planning using town planning principles to support the alignment of business and IT strategies. Email: shattenco@ozemail.com.au


[ Printer Friendly Version ]

[ Other stories about Surveyor, Marshal8e6, OzEmail ]