Building and maintaining a mature architecting capability is not cheap, especially in large global organisations. Typically, with a program/project portfolio in the hundreds, it is impossible to assign good architects to cover the entire portfolio.
Even merely trying to achieve an effective architecture governance, well integrated with the broader IT governance processes, is in itself challenging and often impossible for a portfolio of that size!
Relatively to many other IT areas, architecting is an expensive activity, and investing in it must make business sense. Just like architecting in the construction’s industry, you wouldn’t hire a top architecture firm to extend your kitchen or build a standard block of flats; not only you would struggle to hire and retain a top architect and make the most of her expertise, but also the cost of a top firm would not be justifiable for such projects.
Allocating the best architects where it is most needed happens ad-hoc in small IT shops, but with portfolios in the region of several hundreds IT programs, and a typically resource-constrained architecting practice, in large organisations the work of highly skilled architects must be prioritised systematically.
This can be achieved by introducing the concept of “architectural significance”, and classifying the entire portfolio accordingly. Typical values of architectural significance are High, Medium, Low and Unknown. As a rule of thumb, less than 5% of your portfolio should be of Unknown Architectural Significance (UAS) at any time, and publishing monthly metrics on the architectural significance of your portfolio is a good incentive to minimise the “unknown” and build a good understanding on the architectural relevance of the program portfolio.
Once you understand where the architectural need is greatest, a more optimal resource allocation is possible. For programs of High Architectural Significance (HAS) it makes sense to assign good architects from early on to collaborate with the program on a continuous basis, often with the responsibility of leading the architecture and design work. For programs of Medium Architectural Significance (MAS), a one-off architecture review is normally sufficient, whilst no architecture engagement is typically required for programs of Low Architectural Significance (LAS).
To make this work, an effective collaboration with portfolio/program/project managers is required, and these are some good practices to encourage that:
- don’t capture architecture significance on an isolated spreadsheet; much better to extend the data model of the portfolio management system to capture architectural significance and other architecture engagement attributes (Lead Architect, Lead Reviewer, architecture milestones, etc.)
- involve the portfolio and the program manager when deciding whether a certain program is of high architectural significance
- publish criteria to help decide the degree of architectural significance; budgeted expenditure is one of many criteria, others should be alignment with target architecture, impact on strategic systems, etc.
- produce a short e-learning module and other artefacts (architecture engagement in 1-page) targeted for that specific audience
I have introduced architectural significance as a way to modulate architecture engagement twice now, and I know it works.