The RACI Trap

The RACITrap

Why Clear Roles Still Cause Confusion

RACI charts clarify roles, but not authority. Learn the five tests boards and CEOs should use to turn responsibility into real accountability.

I have sat in steering meetings where every workstream had a neat RACI chart, every box had a name in it, and nobody could answer the most important question: who can actually make the decision?

The project was not suffering from unclear roles. It was suffering from something more dangerous: the illusion of accountability.

That distinction matters.

For years, conventional management wisdom has told us that when a programme becomes messy, clarify the roles. Define who is Responsible, Accountable, Consulted and Informed. Put it on a page. Socialise it. Get everyone to agree.

I have used RACI matrices. They are useful.

But I have also seen organisations spend weeks perfecting them while the underlying decisions remain just as slow, political and ambiguous as before.

My view is simple: RACI is a responsibility map. It is not an accountability system.

And treating it as one is where the trap begins.

Why RACI Looks Better Than It Works

RACI solves a genuine problem.

Large organisations naturally create overlapping responsibilities. Technology programmes make this worse because business units, finance, operations, technology, risk, procurement and external partners may all have legitimate interests in the same outcome.

A RACI matrix brings order to that complexity.

The problem is that most major initiatives do not fail because people cannot identify who is supposed to perform an activity.

They stall because nobody has made clear:

Who owns the business result?

Who gets the final say when two executives disagree?

Who can commit money or people?

How quickly must a deadlock be escalated?

What happens when a commitment is repeatedly missed?

None of those questions is reliably answered by placing an “A” in a spreadsheet.

That is the RACI trap.

The chart creates the appearance of organisational clarity while the operating reality remains unresolved.

An “A” Does Not Automatically Create Accountability

This is where I challenge the conventional wisdom most directly.

Organisations often assume that assigning one person as Accountable means accountability now exists.

It does not.

Imagine the executive accountable for a transformation milestone cannot approve additional expenditure, cannot reassign key staff, cannot resolve a dispute between two business units, and cannot overrule a functional leader whose cooperation is essential.

On paper, that executive is accountable.

In practice, the organisation has made that person responsible for an outcome without giving them the authority required to produce it.

That is not governance. It is organisational theatre.

The opposite problem is equally common.

Several senior stakeholders are labelled Consulted, but culturally each believes consultation gives them an informal veto. Every significant decision therefore needs another meeting, another round of alignment and another layer of reassurance.

The RACI looks disciplined.

The decision process is anything but.

Confusion Usually Appears at the Boundaries

Inside a well-run function, roles are often reasonably clear.

The problems emerge between functions.

Consider a global company replacing a fragmented set of business systems with a common enterprise platform.

Technology may own delivery.

Operations may own process adoption.

Finance may control the investment case.

Procurement may own the supplier relationship.

Risk may define mandatory controls.

Regional leaders may own local business performance.

A conventional RACI can assign each activity neatly.

Then the difficult question arrives.

The global design improves standardisation and lowers long-term cost, but one large market argues that it will disrupt revenue during the transition. The regional leader wants an exception. The transformation leader wants standardisation. Finance wants the savings case protected.

Who decides?

That is not a task-allocation question.

It is a business trade-off involving capital, execution risk, local revenue and long-term operating leverage.

A RACI matrix may tell us who must be consulted.

It rarely tells us whose judgement prevails.

That is precisely where senior management attention is required.

The Cost of False Clarity

Poor accountability is not an administrative inconvenience.

It is expensive.

Decisions get deferred.

Suppliers wait while internal stakeholders debate.

Teams build workarounds because approvals take too long.

Contingency budgets increase.

Senior executives spend time resolving issues that should never have reached them.

Milestones slip, but everyone can demonstrate that they fulfilled their assigned part of the RACI.

This is one reason troubled programmes can produce remarkably clean governance packs.

Every activity has an owner.

Every meeting has minutes.

Every risk has a colour.

Yet the business outcome continues to deteriorate.

Boards should be wary when governance reporting demonstrates activity more clearly than accountability.

The relevant question is not, “Have responsibilities been documented?”

It is, “Can this organisation make and execute the difficult decisions at the speed this initiative requires?”

Five Tests of Real Accountability

For material transformations, strategic investments and cross-functional programmes, I would put every critical outcome through five tests.

If any one of them fails, the RACI is not enough.

1. Is there one owner for the business outcome?

Responsibility for activities can be distributed.

Accountability for an outcome cannot.

If a programme is expected to reduce cost, improve customer experience, shorten cycle time or generate revenue, one executive must ultimately own that result.

Not the presentation.

Not the implementation activity.

The result.

This distinction sounds obvious, but it changes the conversation.

Instead of asking who owns the data migration, training programme or system deployment, senior leadership asks who owns the business benefit the investment was approved to deliver.

Boards fund outcomes, not workstreams.

Governance should reflect that.

2. Does that owner have genuine decision rights?

An accountable executive who cannot decide is not accountable.

For every critical outcome, specify the decisions the owner can make without seeking further consensus.

Can the programme leader reject a local exception?

Can the business sponsor reprioritise resources?

Can the executive responsible for the investment alter scope within agreed financial limits?

Can a functional leader stop deployment because a risk threshold has been breached?

The purpose is not to centralise every decision.

It is to remove ambiguity before a difficult decision arrives.

A surprising amount of organisational friction comes from executives discovering their real authority only when they try to exercise it.

By then, the delay has already begun.

3. Does the owner control, or have guaranteed access to, the required resources?

Accountability without resources is wishful thinking.

Many transformation leaders are given ambitious objectives while critical staff remains controlled by functional departments whose incentives point elsewhere.

The programme becomes a negotiation for people's spare capacity.

Then leadership is surprised when delivery slows.

For strategic initiatives, resource commitments should be explicit.

If an executive is accountable for the outcome, either the required resources should sit within that executive's control or there should be an enforceable organisational commitment to provide them.

Capital allocation matters here as well.

If every minor adjustment requires escalation through multiple financial approvals, the organisation has effectively separated accountability from the ability to act.

4. Is there an escalation clock?

Most governance frameworks describe where an issue should be escalated.

Far fewer specify when.

That omission creates enormous delay.

A disagreement can circulate between teams for days or weeks because everyone believes another meeting might solve it.

I prefer explicit escalation clocks for material issues.

If two functions cannot resolve a decision within an agreed period, it moves automatically to the named executive forum.

No embarrassment.

No politics.

No accusation that someone “escalated too quickly.”

The mechanism is already agreed.

This matters particularly in large transformations where dozens of small unresolved dependencies can quietly become one large schedule problem.

Good governance does not eliminate disagreements.

It prevents disagreements from becoming paralysis.

5. Is there a consequence when commitments are not met?

This is the uncomfortable test.

Many organisations are willing to assign accountability but reluctant to attach consequences to it.

If a critical business unit repeatedly fails to provide promised resources, what happens?

If an executive sponsor repeatedly postpones decisions, is the resulting schedule impact simply absorbed by the programme?

If benefits are not delivered after implementation, does ownership remain with the business, or does responsibility somehow migrate back to the technology team?

Without consequences, accountability becomes descriptive rather than operational.

Consequences do not need to mean punishment.

They may mean transparent benefit ownership, budget adjustments, formal risk acceptance, escalation to the executive committee, or changes to the investment case.

The principle is what matters.

A commitment without a consequence is usually only a preference.

The Board Should Ask Different Questions

Boards and CEOs do not need to review a 70-line RACI matrix.

They need to test whether the operating model beneath it is capable of delivering the promised outcome.

For a major initiative, I would ask five questions:

1. Who owns the business result?

2. Which important decisions can that person make directly?

3. Do they control the resources needed to deliver?

4. How long can a cross-functional disagreement remain unresolved?

5. What happens when a critical commitment is missed?

If the answers are vague, the governance is vague, regardless of how polished the responsibility matrix appears.

These questions are particularly important for technology investments because they often cross organisational boundaries more aggressively than traditional capital programmes.

A factory expansion has a physical location.

A digital transformation may touch sales, finance, supply chain, HR, risk, customer service and every geography simultaneously.

The interfaces become the programme.

That makes decision architecture as important as technical architecture.

Should Organisations Stop Using RACI?

No.

That would be solving the wrong problem.

RACI remains useful for clarifying participation, particularly where multiple functions interact.

The mistake is expecting it to do something it was never designed to do.

A RACI should tell teams how responsibilities are distributed.

It should sit underneath a stronger layer of governance that defines business outcomes, authority, resources, escalation, and consequences.

Think of RACI as a map.

A good map tells you where everyone is positioned.

It does not decide who has the steering wheel.

Why This Matters More as Organisations Become More Matrixed

Modern organisations increasingly rely on matrix structures, shared services, product teams, external partners and cross-functional transformations.

That means authority is becoming more distributed while business outcomes remain interconnected.

The natural response has been to produce more governance.

More committees.

More role definitions.

More matrices.

More approval stages.

But adding governance artefacts does not necessarily increase accountability.

Sometimes it does the opposite.

When too many people participate in every decision, individual accountability becomes weaker.

When every stakeholder must be comfortable before action is taken, speed becomes optional.

When the executive with the “A” lacks control over money, people or trade-offs, the letter becomes ceremonial.

In competitive markets, that is not merely inefficient.

It is a strategic disadvantage.

The organisations that move fastest are not necessarily those with fewer controls.

They are often the organisations that are clearer about where authority sits and when it moves.

The Real Test of Governance

The quality of governance is not revealed when everyone agrees.

It is revealed when two legitimate priorities conflict.

Revenue versus standardisation.

Speed versus control.

Local flexibility versus global scale.

Short-term cost versus long-term capability.

That is when the organisation discovers whether its accountability model is real.

If the answer is another alignment meeting, another steering committee discussion and another revision of the RACI, the problem is probably not unclear roles.

The problem is unclear authority.

Three decades around enterprise programmes have reinforced one lesson for me: organisations rarely suffer from a shortage of intelligent people who understand their responsibilities.

They suffer when intelligent people are placed inside systems that make decisive action difficult.

A RACI matrix can clarify who is in the room.

Leadership must still decide who is allowed to decide.

Where have you seen RACI create genuine clarity, and where has it simply documented confusion more neatly?

Subscribe to TechnologyTrends, or add your perspective in the comments.