The 88% Failure Rate: Why You Can’t “Roll Out” a Transformation
- Darren Emery
- Jan 12
- 6 min read

If you’re leading a transformation right now, the odds are not in your favour.
According to Bain’s 2024 analysis, 88% of business transformations fail to achieve their original ambitions. Not underperform. Not stall. Fail.
That should be an intolerable number.
In any other part of the business, an 88% failure rate would trigger immediate intervention.
If your cloud platform failed 88% of the time, the vendor would be gone.
If sales missed quota 88% of the time, the Sales Director would be replaced.
Yet when it comes to transformation - Agile, product, digital, customer-centricity - we explain failure away.
We blame "culture." We blame "middle management oversight." We blame "change fatigue."
We are blaming the wrong things.
And we’re diagnosing the wrong problem.
An 88% failure rate isn’t just a missed ambition.
It’s sunk transformation budgets, years of executive attention diverted from customers, and organisational debt that compounds long after the consultants have left.
Most failed transformations don’t leave you where you started.
They leave you slower, more cynical, and structurally harder to change next time.
In a slower, more predictable market, you could survive this failure rate.
In today’s market, it compounds faster than most leadership teams realise.
The root cause is not execution. It is a fundamental misunderstanding of how decisions, risk, and incentives actually work in your business.
You Can’t Command a Complex System
Most leaders attempt to transform an organisation as if it were a machine.
Machines respond to instruction. Organisations don’t.
A company is a complex adaptive system: shaped by incentives, constraints, power, and risk.
Behaviour emerges from the system whether you like it or not.
Which leads to the uncomfortable truth:
You cannot transform a company. A company can only transform itself.
And it will only do so when you stop trying to change people - and start changing the physics of their environment.

There Is No Resistance - Only Rational Behaviour
“People are resisting the change.”
This is the most common complaint I hear from executives. Usually followed by a plan to do more communication, more training, or another town hall.
But from a systems view, resistance doesn’t exist.
What you’re seeing is coherence.
Behaviour is simply the system doing exactly what it was designed to do.
People are not irrational. They are highly attuned to how survival actually works inside your organisation.
If your leadership narrative says “Be bold and experiment” but your promotion system rewards predictability and zero defects, people will choose predictability, every time.
That’s not resistance.
That’s competence.

If a Product Manager spends weeks building a deck to justify a £40k experiment, they’re not lacking empowerment. They’re navigating a governance structure optimised for control.
If an engineer pads estimates by 300%, they’re not disengaged. They’re responding to a system where being late is punished more than being slow.
When leaders label this behaviour as resistance, they reach for narrative fixes: clearer vision, better storytelling, more Agile coaching.
But when behaviour is coherent, the fix isn’t narrative.
It’s structural.
Culture Isn’t the Lever You Think It Is

“Culture eats strategy for breakfast” is quoted endlessly in transformation work.
What’s usually forgotten is the reality:
Culture follows structure.
Culture is not something you install.
It’s the shadow cast by your operating model.

It’s formed by how people experience:
funding cycles
approval gates
decision rights
risk distribution
who gets promoted - and who doesn’t
You don’t change culture directly.
You change the system that generates it.
Which is why most transformations break in exactly the same places.
Where Transformations Actually Fail
1. Strategy Is Forced Into a Delivery Shape (The OKR Trap)
Most organisations now use OKRs.
Most of them misuse them.
The problem isn’t OKRs themselves - it’s using them as commitment contracts instead of learning instruments.
Strategy is probabilistic.
It’s a set of bets: If we do this, we believe that will happen.
Delivery is deterministic.
We will build X by date Y.
When strategy is forced into a delivery-shaped container = “Objective: Launch feature by Q3” - Value is sacrificed for compliance.
Teams are required to commit to solutions before they understand the problem.
Adaptation becomes non-compliance.
Discovery becomes failure.
Which is why many leadership teams say they want learning - until it contradicts the plan they’ve already committed to.
The Fix:
Use OKRs to describe changes in outcomes or behaviour (not outputs)
Allow teams to choose - and change - solutions
Treat learning that avoids unnecessary work as success, not deviation
If your organisation can’t tolerate that ambiguity, you don’t have a product problem.
You have a control problem.
2. Responsibility Is Pushed Down - Authority Is Not
Everyone says they want empowered teams.
Very few are willing to govern or fund them properly.
Empowerment is not a mindset.
It’s decision physics.
Decision physics is simple: Who can decide what, with which money, without asking.
If a team needs permission to pivot, they aren’t empowered.
If a team needs approval to run a small experiment, they aren’t empowered.
If teams own outcomes but executives retain decision rights, you get theatre - not speed.
This is how organisations end up with Agile rituals layered on top of Waterfall power structures.
The Fix:
Define thresholds of autonomy:
a clear outcome
a bounded budget
explicit guardrails
For example:
“You have £250k and one quarter to move this metric. Stay within legal and brand guardrails - no approval required.”
Then genuinely step back.
This feels uncomfortable for most leaders, because it removes the illusion of control.
But in complex markets, centralised certainty was never real anyway.
The Real Enemy: Comfortable Ambiguity at the Top
Most failed transformations aren’t blocked by resistance below.
They’re softened to death by ambiguity above.
When leaders won’t make hard trade-offs explicit - speed vs certainty, learning vs predictability, autonomy vs control - the organisation fills the vacuum with the safest possible interpretation.
Ambiguity at the top doesn’t create flexibility. It creates conservatism at scale.
Selective Violation: How Leaders Kill Change Under Pressure
Employees are sceptical, for good reason. They’ve seen this all before.
They don’t judge your transformation by your roadmap.
They judge it by what happens when pressure hits.
When delivery slips, do we reduce scope - or mandate heroics?
When data contradicts senior opinion, which one wins?
When an experiment fails, is the learning rewarded - or punished?
Most leaders fail this test.
They support the new model right up until it constrains them personally.
“We trust teams - but I need to sign this off.”
“We’re Agile - but this project needs fixed scope.”
“We empower Product - but not on this decision.”

Every selective violation teaches the organisation the same lesson:
The old rules are still the real rules.
And the organisation learns that transformation is optional - but hierarchy is not.
If you are not willing to be constrained by the system you are building, you are not leading a transformation - you are running a rebrand.
Your Role Has Changed - Whether You Like It or Not

In the old model, executives were Captains.
You set direction.
You issued orders.
You corrected course.
In modern organisations, that role doesn’t scale.
Your job now is Architect.
You design the system that determines:
Funding: persistent teams, not temporary projects
Staffing: dedicated, cross-functional units, not resource pools
Governance: outcome-based review, not permission gates
This is harder than command-and-control leadership.
Because it requires dismantling the very mechanisms that once made you feel safe.
But the pattern is consistent.
The 88% who fail try to force behaviour.
The 12% who succeed change the system so the behaviour makes sense.
A Non-Negotiable Test
If your transformation is real, you should be able to answer these without hesitation:
What decisions can teams make today without asking permission?
What funding can be reallocated without executive escalation?
When was the last time a senior leader was overruled by data - and accepted it?
If those answers are vague, your transformation isn’t failing because people don’t “get it”.
It’s failing because the system still points backwards.
Stop trying to transform your company.
Build a system where the behaviour you want is the path of least resistance.
Then get out of the way.
If that feels risky, it’s because your current system is protecting you more than it’s protecting the business.




Comments