Architecture has a bit of a reputation problem. Too often, delivery teams see it as something that’s done to them: another approval meeting, a thick pack of diagrams or a perfect future that has very little to do with today.
I’m not suggesting we stop being rigorous. Quite the opposite. We just need to make that rigour useful. The real job is to help people understand the problem well enough to make a good decision early, then carry it through with confidence.
Trust comes first
You can have the best answer in the room and still achieve almost nothing. People need to know you understand what matters to them. That could be clinical safety, keeping a service running, staying within budget, making life easier for users or simply having enough people to do the work.
So spend time with people before you present to them. Listen for the worry behind the formal requirement. Say when you don’t know something. And explain the trade-offs in words the person making the decision would use themselves.
People rarely resist architecture. They resist decisions that feel detached from their reality.
Take healthcare. A technically elegant change can still be completely wrong if it ignores a round-the-clock clinical workflow, shared devices or what even a short interruption would mean for staff and patients. Context isn’t a footnote to the design. It’s part of the design.
Make it clear, not impressive
A good architecture view should make a difficult decision easier to understand. It shouldn’t exist to prove how much the architect knows.
You’ll usually need different views of the same truth. A board wants to understand the choices, risks, cost and result. A delivery team needs the boundaries, dependencies and enough detail to get moving. The people looking after governance need to see how the choice fits with everything else. One view isn’t more “correct” than another; each has a different job.
A useful test
- Can a senior decision-maker explain the choice after the meeting?
- Can a delivery lead identify the next action and the important constraint?
- Can another architect understand why the decision was made six months later?
- Can the people affected see how their needs shaped the answer?
If the answer is no, the document might be accurate, but it isn’t doing its job yet.
Stay close to the delivery
The destination matters. Standards and sensible boundaries do too. But architecture quickly loses credibility if the architect disappears after approval and only comes back for the next review.
Reality will change the design. A supplier will uncover a limit. The data will be messier than anyone expected. A dependency will move. That’s normal. Good architecture treats new information as useful feedback, not as failure. Stay involved, protect the result and let the route change when it needs to.
That’s also how trust builds. Delivery teams see that architecture can unblock a decision rather than just police it. Leaders see that the advice still works in the real world. The next difficult conversation gets a little easier.
Governance should help things move
Good governance makes the safe route the easy route. It tells people who can decide, what they need to show and how to handle a genuine exception. It also gives teams patterns they can reuse instead of solving the same problem again and again.
Bad governance creates queues, duplicated paperwork and decisions made elsewhere to avoid the process. Sometimes more control gives you less visibility.
Keep asking a simple question: does this step help us get a better result? If it doesn’t, simplify it. If it does, explain why so teams can see the value too.
The work behind the work
People can see the models, roadmaps, principles and designs. What they don’t always see is the real work: joining the plan to the investment, the technical possibility to day-to-day reality and a group of people with different priorities to one shared direction.
That’s why trust isn’t a soft extra. It’s part of the architecture. Without it, even the right design can go nowhere. With it, a complicated organisation can make a bold change without forgetting what matters.