The proper establishment of the scope of an ICT project at an early stage can avoid the perception of cost or schedule overruns, says Neil Brown of the Ministry of Health.
Defining the scope — the range of desired outcomes — is one of the crucial early steps of designing an enterprise architecture, Brown told an audience at the Enterprise Management stream of last week’s one-day Govis “mini-conference” in Wellington.
These outcomes, he emphasises, should be expressed in business terms, not the language of technology.
Enterprise architecture, which he defines as the “effective and efficient implementation of business processes”, should align the design of the whole ICT system to business needs, identify the dependency of one business process on others, locate overlap and duplication and help consolidate processes which are trying to achieve similar or related objectives.
Correctly done this process helps improve decision-making in the design of ICT systems, and minimises total cost of ownership, he maintains. TCO, though sometimes regarded as a cliché, is a crucial measure in deciding on alternative approaches to development of a system.
The only way of comparing apples and oranges is to reduce both to dollars, he says.
Enterprise architecture, Brown says, also promotes innovation, by allowing the mundane to be done quickly and efficiently so the development team can devote more attention to innovation.
However, both Brown and another speaker, Todd Forsythe of Victoria University, admit the concept of enterprise architecture can be difficult to sell to management and staff. It is vital, they say, to identify some obvious early wins, by way of process improvement, to the people controlling the budget They may not be the architect’s immediate or even indirect bosses.
As with the architecture of a building, the architect may have a grand concept of the whole — an edifice as harmonious in its design as the Taj Mahal — but unless the architect can answer “the person building his own little minaret” when they ask “what’s in it for me?” the battle may be lost, Forsythe says.
The knack of doing that has something in common with the software as a service trend, where a complex suite of software appeals to particular users only insofar as specific parts of it directly address their needs. Providing “enterprise architecture as a service” in a similar way is the only way to introduce it into an organisation, he says.
Both agree the process of accommodating an organisation to an enterprise architecture is an incremental loop, a “virtuous circle”. The organisation’s maturity and the quality of its enterprise architecture are mutually dependent on each other and the architecture is “as much an effect as a cause”, says Forsythe.