Why IT–Business Alignment Remains a Structural Illusion
The crash withdraws the capital; it does not touch the wiring.
The Ritual We Mistake for Progress
There is a conversation that recurs in every large organisation I have observed, and it recurs with the reliability of a season. The business complains that technology is slow, expensive, and disconnected from what the enterprise actually needs. Technology replies that the business cannot say clearly what it wants, changes its mind without warning, and fails to appreciate the complexity beneath the surface. A workshop is convened. A steering group is refreshed. A new liaison role is created and given a title with the word partnership in it. Everyone leaves the room believing that alignment has been improved. And within two quarters the same conversation begins again, almost word for word.
We have been holding this conversation for the better part of a decade. The collapse of the past year has done nothing to end it — it has only made it louder, and more expensive to ignore. The money that flowed so freely into technology during the boom has stopped, and as it recedes it has left the beach littered with the evidence: systems that were bought but never adopted, integration programmes that consumed fortunes and delivered interfaces nobody trusts, strategies that were written with great ceremony and then quietly abandoned when the people who commissioned them moved on.
The standard diagnosis of all this is that business and technology are not aligned, and that if only they communicated better the problem would dissolve. I want to argue that this diagnosis is not merely incomplete but actively misleading — and that treating alignment as a relationship problem is the reason we never fix it.
Alignment Is Not a Relationship Problem
The language we use gives the game away. We speak of alignment as though it were a matter of goodwill, mutual understanding, and better meetings. We send technologists on courses to learn the language of the business. We ask business sponsors to be more engaged. All of this assumes that the two sides are pulling in different directions because they misunderstand one another.
They are not. In most of the organisations I have seen, business and technology understand each other perfectly well. They are pulling in different directions because the structure they operate inside rewards them for doing so. Alignment fails not despite everyone’s best efforts but because the wiring of budgets, ownership, and measurement quietly guarantees the failure, and no amount of partnership language re-wires anything.
The pattern that recurs is this: we treat as a problem of relationships something that is, in fact, a problem of structure. And structural problems are indifferent to goodwill.
Consider three pieces of that structure — each of them so familiar that we have stopped seeing them as choices at all.
The Budget Is Wired for Misalignment
In the majority of enterprises, technology is funded as an overhead. It sits in the same mental category as facilities and telephones — a cost of doing business, to be contained. When money is plentiful, this shows up as indulgence; every function gets the systems it lobbies for. When money is scarce, as it is now, it shows up as indiscriminate cutting.
But notice what funding technology as overhead actually does to behaviour. It tells the technology function that its job is to minimise its own cost. It cannot be held accountable for creating value, because it was never funded to create value — it was funded to keep the lights on as cheaply as possible. Meanwhile the business units, who do carry value targets, treat technology as a free good to be requisitioned rather than an investment to be justified. Neither side is behaving irrationally. Each is responding exactly as the budget structure asks it to.
- The technology function optimises for cost, because cost is what it is measured on.
- The business optimises for its own local outcome, and externalises the cost onto a shared overhead.
- No one optimises for enterprise value, because no one is funded to.
You cannot fix this with a workshop. You can only fix it by changing how the money flows: by treating significant technology spend as investment attached to a named business outcome, sponsored by the person who will carry the benefit — or the blame.
The Ownership Vacuum
The second structural fault is subtler and, in my experience, more corrosive. A large system is commissioned by one group, paid for out of a shared pool, delivered by a programme that disbands on go-live, used daily by a fourth population, and — this is the crucial part — owned, in any meaningful ongoing sense, by no one.
We saw the consequences everywhere in the enterprise resource planning wave of recent years. Organisations spent years and extraordinary sums installing integrated systems, and a striking number of them ended up with the software running and the change unrealised — the modules live, the old behaviours intact, the promised integration existing on the diagram and nowhere else. The technology was delivered. The transformation was not. And when you ask who was accountable for the difference between those two things, you find the vacuum: the programme has gone, the sponsor has been promoted, and the users were never asked.
“A system with no owner is not a neutral asset. It is a liability that has not yet presented its invoice.”
Ownership cannot be conjured after the fact. It has to be designed in — a single accountable owner for each significant capability, present before the programme starts and still present long after it ends, whose performance review actually depends on whether the thing delivers.
We Measure Delivery and Call It Value
The third fault completes the trap. We measure technology programmes by whether they land on time and on budget, and we hold the celebration at go-live — the precise moment before any value has been created. The benefit case that justified the whole endeavour is filed away and almost never revisited. Nobody returns eighteen months later to ask whether the savings materialised, whether the promised productivity appeared, whether anyone is actually using what we built.
This is not an oversight. It is the natural result of the first two faults. If technology is funded as a cost and owned by no one, then of course we measure it by cost and delivery — those are the only things anyone is accountable for. Outcome measurement would require an outcome owner, and we have just seen that there isn’t one.
- We approve spend on a promised benefit.
- We manage the programme against time and cost, not benefit.
- We declare success at go-live, before the benefit could exist.
- We never audit whether the benefit arrived — and so we never learn.
The loop never closes. And because it never closes, we repeat it, boom after boom, crash after crash, each time believing the failure was one of execution rather than design.
Why the Crash Will Not Fix It
There is a comforting theory abroad at the moment that the downturn will impose the discipline the boom lacked — that scarcity will force organisations to spend on technology only where it genuinely pays. I do not believe it, and the reason is the argument I have just made. The crash withdraws the capital; it does not touch the wiring. Faced with pressure, an organisation whose technology is funded as overhead, owned by no one, and measured by delivery will do the only thing that structure permits: it will cut, indiscriminately and often stupidly, hitting the investments that were creating value alongside the ones that never would. Then, when conditions ease, it will spend again on exactly the same terms. The illusion of alignment survives every cycle precisely because we keep treating it as a mood rather than a mechanism.
What a Structural Answer Looks Like
If alignment is structural, then only structural moves will shift it — and they are neither mysterious nor new. They are simply harder than convening another workshop, which is why we avoid them.
- Fund technology as investment, not overhead. Significant spend should be attached to a specific business outcome and sponsored by the person accountable for that outcome. If no one will put their name to the benefit, the enterprise has learned something valuable before spending a penny.
- Give every capability a standing owner. Not a programme, which disbands; a person, who remains — accountable for adoption and value long after the integrators have gone home.
- Measure the benefit, not the delivery. Return to the business case after go-live, on a fixed schedule, and ask the uncomfortable question: did the value we promised actually appear? Then feed the answer back into the next decision.
None of this requires better relationships between business and technology. It requires the enterprise to stop pretending that the problem lives in the relationship at all. The two sides are not misaligned because they misunderstand each other. They are misaligned because we built them a structure that rewards misalignment, and then we act surprised, every quarter, when it delivers exactly what it was built to deliver.
The organisations that break the cycle will not be the ones that talk most warmly about partnership. They will be the ones willing to change where the money sits, where the ownership lives, and what we are honest enough to measure. Everything else is ritual — and we have performed it long enough.