Value from EA Doesn’t Require Mountains of Data
It’s hard to argue with the idea that decisions made with limited information usually lead to limited results. Better information, used well, increases your chances of achieving the outcomes you’re aiming for. That principle sits at the heart of Enterprise Architecture (EA). And yet, for all its logic, many organisations still struggle to justify EA in terms of value versus effort and cost.
One of the most common objections goes something like this: “The volume of data required to make EA meaningful is enormous. The effort needed to capture and maintain it just isn’t worth it.”
On the surface, that sounds reasonable. If EA means documenting everything about the organisation – every process, every system, every integration, every technology – then yes, that’s a huge undertaking. But here’s the catch: that argument is built on a false assumption. It assumes that “enterprise” must mean the entire organisation, end to end, top to bottom, all at once.
In practice, that approach is exactly what causes many EA initiatives to stall or fail.
Think of it like decorating a palace. You wouldn’t have decorators sprinting between dozens of rooms trying to paint them all at the same time. You’d focus on one room, complete it properly, and then move on to the next. The same principle applies to EA. Success comes from maintaining focus on a well-defined scope, tackling one meaningful area at a time.
Defining Scope
Scope can be defined in several ways.
First, by business domains or capabilities. If you know the outcome you want – perhaps reducing IT costs, simplifying complexity, eliminating redundancy, managing risk more effectively, or improving time to market – then start where those problems are already known to be prevalent. Pick a specific domain or set of capabilities where the pain is real and measurable. Concentrating effort there makes it far easier to demonstrate tangible value quickly.
Second, by architecture layers. Not every problem requires a full-stack architectural deep dive. If your goal is to reduce application redundancy, for example, your focus will primarily sit within the business and application layers. There may be no immediate need to dive deeply into infrastructure or technology layers. Being selective about which layers you examine dramatically reduces the amount of data you need to capture and maintain.
Third, by level of abstraction. This is often overlooked but hugely important. Do you need to understand the what (conceptual), the how (logical), or the who and where (physical)? The more concrete and detailed the view, the larger the data footprint becomes. Starting at the appropriate level – often conceptual or logical – allows you to gain insight without drowning in unnecessary detail.
Finally, scope can be reduced by organisational boundaries. In large or complex organisations, trying to engage everyone from day one is rarely effective. If possible, focus on a specific business unit or organisation where stakeholders already recognise the potential value of EA. Early advocates make progress easier, and demonstrable results build momentum for broader adoption.
The common criticism that EA demands overwhelming amounts of data is only true when the scope is overwhelming. When you are laser-focused on a clearly defined area, the volume of required data becomes manageable – and, more importantly, purposeful.
By narrowing the scope, you shorten the path to demonstrable value. And in doing so, you directly address the original concern: EA doesn’t need masses of data to prove its worth. It needs clarity, focus, and disciplined scoping.
Useful Links:
- Practical Jump-Start to Enterprise Architecture by Eetu Niemi
- Building Enterprise Architecture Capability: A Strategic Engine for Change by Xin-Kuan (Leo) Yeh