1. DIY is the dominant service specification approach.
Respondents report there’s a wide range of approaches to service specification in use, and the most common approach by far (50%) is “do it yourself”. Sure, DIY will mean that practitioners are like magpies, they pick up ideas from a variety of sources, and create their own capability. But this, by any measure, must represent a failure of industry standards. While consistency of specification meta data may not be a primary goal for designers and developers, architects should consider a range of risks including use of development automation support tools, standardization of technical components of outsourcing contracts, opportunity to find/reuse pre-existing services, and crucially standardization of tooling. Which takes me onto the next point . . .
2. Documents and Spread-sheets still dominate tool usage for service planning, reporting and management.
In the early stages of service adoption almost everyone simply uses the tools that are available to hand. But as the service portfolio evolves and matures these tools become a strategic liability in areas of planning, reuse management, portfolio and asset management, governance and reporting. I note not one respondent mentioned the EA Repository in context with service support tools. And again this is closely connected with the third interesting conclusion . . .
3. 64% dissatisfied with tool support.
There was a very high level of dissatisfaction with tools, particularly in the context of communicating service specifications between the different roles and groups across the organization. We all appreciate high quality service specifications are the result of multi-disciplinary efforts, spanning Business Analysis, Architecture, Design, Developers, Product Management, IT Service Management, Outsourcing, Procurement, Governance etc . So this is much more than reporting, there’s an evident need for modern tools that facilitate the collaboration process.
See Service Specification Practices Survey 2013 - Report