But what is Enterprise Architecture actually for?
We do a lot of work on designing organisations. Organisation structure has a massive impact on how well organisations perform: how well they can fulfil their purpose, how well they can set and execute strategy, and how efficient they are. We also tend to look at the information in the organisation. Part of this is about good information flow, and part is about providing the right information for decision support.
Call these business architecture and information management, and it’s not surprising that we tend to work with enterprise architects. And there is something that really puzzled us and it’s the diversity of this thing called Enterprise Architecture. Depending on role and organisation, enterprise architects are doing a vast range of different things, from managing technology platforms to supporting managers in the operations of the business, from designing suites of IT applications to providing information to inform transformation planning. So it’s maybe not that surprising that we got a wide range of answers when we’ve asked enterprise architects: “But what is EA actually for? The one that really resonated was from Sǿren Martensen from Copenhagen, who gave us a view which we found very helpful:
“… what we really get brought in for, is to make the organisation’s complexity understandable.”
And certainly that matches other insights we have, including an overwhelming majority of CEOs (78%) who say that reducing complexity is a personal priority, and that managing complexity demands special tools. Really interestingly, only 5% think they have those special tools. It also matches our experience that the really big organisational issue is around modelling complexity and finding ways to harness it rather than being buried by it.
So we set about having more conversations with enterprise architects, and a set of core themes emerged, with that handling of complexity at its heart. We focused on what it’s for, not what it does and here’s our current thinking, in brief.
- To provide technical input to the design of the enterprise, its organisational and its information flows, and to design underpinning IT where appropriate.
- To make the organisation’s complexity understandable to management so they can actually manage the organisation, both running it and changing it.
- To manage ongoing changes to the architecture as the enterprise changes though time.
- To design the information and decision structure of the enterprise to provide decision makers with the information they need to take the right decisions at the right level with the right information.
- To enable governance of the coherence and integrity of the enterprise’s architecture.
It’s still Work in Progress, but already this has triggered some really interesting discussions. We get a positive reception from most enterprise architects we’ve tried this on, with some comments that some of these require a higher degree of EA ‘maturity’ than others. Fair enough. But there’s also been a consistent sense that the current EA toolkit is insufficient to handle all of these well, particularly where is a lot of complexity and high rate of change. So just as the CEOs reported, it seems we need additional and specialised tools to be able to handle complexity and dynamics. We’ve started working with a small group of enterprise architects to explore how and where systems approaches could supplement the existing architect’s toolkit. And in discussion and in practice, we are finding that systems tools bring new types of value to the delivery of enterprise architecture. In particular, systems approaches focus on the relationships between things as well as on the things themselves, and this style of cause-effect thinking is critical in a world which is complex and dynamic.
So, early days, but encouraging. What we really want is for enterprise architects to be able to design viable organisations – those that can adapt themselves to their environment (or adapt their environment to them), and that they have the best available tools to do that.