Connection Over Control — de Visscher — Reading the Map

Part 7 · The Practitioner’s Compass

The final part of this book steps back from the mechanics and asks the larger question: what is enterprise architecture ultimately for, and where is it heading? Three practitioners offer three arguments. This chapter takes the first — Niek de Visscher’s contention that enterprise architecture is not, at its best, about control at all. It is about connection.

The distinction sounds gentle. It is in fact a sharp critique of how much of the discipline operates, and a different theory of where architectural value comes from. An architect who internalises it does the job differently — and, on the evidence, more effectively.

The Argument

The conventional, often unstated, theory of enterprise architecture is that its job is control. The architecture function exists to impose order: to set the standards, enforce the principles, gate the decisions, and prevent the parts from doing things that harm the whole. On this theory, the architect is a kind of regulator, and the function’s value is measured by what it stops.

de Visscher’s argument is that this theory is mistaken about where architectural value actually comes from. The value, he contends, is in connection — in linking the things that organisations habitually keep apart. Strategy to execution. Ambition to reality. The board’s intent to the delivery team’s work. Today’s decision to tomorrow’s consequence. The architect’s distinctive contribution is to be the connective tissue that runs between these separated things, carrying meaning across the gaps where organisations otherwise lose the thread.

On this view the architect is not the regulator who stands at the gate saying no. The architect is the translator and connector who runs between the strategists and the builders, the present and the future, making sure that what was intended at one end is understood and realised at the other. The gatekeeper’s value is in refusal; the connector’s value is in linkage. de Visscher’s claim is that the linkage is worth far more, and that the discipline’s drift toward control has been a drift away from its real source of value.

Why the Gatekeeper Produces Worse Outcomes

The argument is not merely that connection is nicer than control. It is that the architect who positions as a gatekeeper produces worse outcomes than the one who positions as a connector — and the mechanism is worth understanding.

The gatekeeper accumulates resistance. Every refusal creates an adversary, and an architecture function experienced as the body that says no is a function that teams learn to route around, withhold information from, and engage as late as possible to minimise its interference. The gatekeeper, paradoxically, ends up with less influence over decisions, because the decisions are increasingly made out of its sight, presented as faits accomplis precisely to avoid the gate. Control, pursued directly, produces evasion, and evasion produces exactly the uncoordinated estate the control was meant to prevent.

The connector accumulates the opposite. An architecture function experienced as the body that helps decisions be better, that connects a team’s local problem to the resources and context it lacks, that carries the team’s constraints back to the strategists — that function is sought out, engaged early, and given the information it needs, because teams want it in the room. The connector ends up with more influence over decisions, because the decisions come to it willingly. Connection, pursued directly, produces cooperation, and cooperation produces the coordination that control could not compel.

This is why the governance chapter (Chapter 28) insisted that a good ARB is one teams cooperate with rather than route around, and why the function chapter (Chapter 33) made connection-over-control its central mandate strategy. de Visscher’s argument is the principle underneath those practical recommendations: govern by connecting, not by gating, because gating defeats itself.

de Visscher Applied to Meridian

Meridian’s architecture function lives this argument out, and the contrast between its two modes is the clearest illustration of de Visscher’s point.

In its early form, before the EA had learned the lesson, the function tried to operate as a gate. The principles were asserted; the integration project that violated them (Chapter 28) was, in the gatekeeper instinct, something to be stopped. Had the EA simply enforced the principle and blocked the project, she would have created an adversary, taught the project team to engage architecture as late as possible, and very likely seen the integration built anyway, out of her sight, presented as already done. The gate would have produced the evasion it was meant to prevent.

What she did instead was the connector’s move. She connected the project’s real constraint — its deadline, the unbuilt integration platform — to the architecture’s real concern — the cost of another point-to-point connection. The conditional yes (build it now, own the debt, remediate in Wave 5) was not a softer form of control; it was connection in action, linking the team’s reality to the estate’s interest in a way both could accept. The project team left that conversation as an ally of the architecture function, not an adversary, and engaged it early on the next decision rather than late. The connector’s move bought more actual control over the estate than the gate ever would have.

The deeper illustration is the EA’s whole trajectory. Her effectiveness rose not when her analysis improved but when she stopped positioning herself between the organisation and its decisions as a checkpoint, and started positioning herself as the tissue connecting the parts that could not otherwise reach each other — the CFO to the technical reality, the clinicians to the migration plan, the engineers to the strategy. Every gain in her influence came from a gain in connection, not from any increase in her authority to control. Meridian is de Visscher’s argument rendered as a year in an architect’s working life: the connective architect succeeds where the controlling architect would have failed, not because she is gentler but because connection is where the value and the influence actually are.

The Limits of the Argument

In the spirit of the honest reading this book has tried to maintain, the connection-over-control argument has limits worth naming, and de Visscher’s defenders would not deny them.

Connection is not the absence of authority. The connector still needs a governance mechanism with real teeth (Chapter 28); the argument is about how that authority is positioned and used, not about discarding it. An architecture function that connects beautifully but can never, in the end, prevent a genuinely harmful decision is not the connective ideal — it is the advisory body dressed up, the Architecture Theatre of Chapter 28. The connector’s authority is exercised through relationship and framing rather than refusal, but it is still authority, and there are moments — a decision that would genuinely endanger patients, say — where the connector must be able to say a hard no. Connection is the default mode and the source of value; control is the reserve the connector rarely has to draw on precisely because the connection is working.

The argument also assumes an organisation capable of being connected — one where the parts, once linked, will cooperate. In genuinely dysfunctional organisations, where interests are actively hostile, the connector’s tools may not be enough, and more formal authority may be necessary. de Visscher’s argument is a theory of how value is normally created, not a guarantee that connection alone suffices in every pathology.

Translator Panel

The control theory says: “The architecture function enforces standards and governs compliance across the enterprise.” > What that means: Architecture’s value is in what it stops — the bad decisions prevented, the standards imposed. de Visscher’s critique is that this positioning produces evasion: teams route around the gate, decisions are made out of sight, and the function ends up with less real influence over the estate, not more.

de Visscher’s view says: “The architect is connective tissue — linking strategy to execution, intent to reality.” > What that means: Architecture’s value is in the connections it makes across the gaps organisations leave between strategy and delivery, present and future. The connector is sought out rather than routed around, and so ends up with more influence over decisions — governing by linkage rather than by refusal, and getting cooperation where control gets evasion.

The Key Idea

Niek de Visscher argues that enterprise architecture’s value lies not in control but in connection — in being the connective tissue that links strategy to execution, ambition to reality, intent to outcome, and today’s decisions to tomorrow’s consequences. The argument is sharp because it claims the gatekeeper produces worse outcomes than the connector: control pursued directly produces evasion, as teams route around the gate, while connection pursued directly produces cooperation, as teams seek the function out, so the connector ends up with more influence over the estate, not less. Meridian renders the argument as lived experience — the architect’s effectiveness rose with every gain in connection and not with any increase in her authority to control. The limits are real: connection is not the absence of authority, the connector still needs governance with teeth in reserve, and the theory assumes an organisation capable of being connected. But as a theory of where architectural value normally comes from, connection over control reframes the whole job.

Next: Chapter 52 takes the second practitioner’s argument — Vince Dooher’s claim that architecture is, underneath everything, a teaching profession, and that the architect who teaches outperforms the one who polices.