Transparency Is a System, Not a Speech

There is a small moment in Chapter Ten that readers skip past, and it is the one I would defend hardest.

Anika reaches into her bag and slides a worn leather notebook across the table. David opens it, expecting — I think — something quotable.

Instead he finds diagrams.

Decision visibility. Context-sharing. Rationale documentation. Stakeholder loops.

Not statements. Not principles. Plumbing.

I wrote it that way deliberately, because every transparency initiative I have watched fail, failed in the same manner: it was announced rather than built.

The all-hands. The open-door policy. The leader saying my door is always open to a room of people who have already learned otherwise.

Those are speeches. They cost nothing, which is exactly why organisations reach for them.

What People Actually Judge

For a long time I assumed the goal of transparency was for people to agree with the decision. That if I explained well enough, they would come around.

They often didn't. And the ones who didn't come around were frequently the people I most needed.

What changed my thinking was discovering that researchers had studied this for decades, and had reached a conclusion that runs against the instinct of most leaders.

The field is called procedural justice, and it goes back to work by Thibaut and Walker in the 1970s. The core finding is that people evaluate the fairness of the process separately from the fairness of the outcome — and that the process carries far more weight than we assume.

In 1980, Gerald Leventhal set out six rules that determine whether people judge a procedure to be fair. Consistency. Bias suppression. Accuracy. Correctability. Representativeness. Ethicality.

Read that list again and notice what is missing.

Not one of the six says anything about the outcome.

You can satisfy every rule on that list and still tell someone no. That is not a flaw in the framework. It is the entire point.

Later work by Tom Tyler found that when people have genuine voice in a process — a real opportunity to put their case — they become significantly more willing to accept a decision that goes against them.

And the effect is strongest precisely where you would want it: it matters most when the news is bad.

This has been replicated across methods, across settings, and across cultures including Asian ones, which matters to me because a great deal of management research does not travel and this body of work largely does.

So the goal was never agreement.

The goal is that someone can disagree with your decision and still trust the way you reached it.

That is a much lower bar than persuasion, and a much higher bar than communication.

The Part Nobody Says Out Loud

Here is the conclusion I have come to, and I hold it uncomfortably.

You cannot make an unfair process transparent without making things worse.

Which means most resistance to transparency is not really about information at all. When a leader pushes back on making decisions visible, they are rarely protecting confidential data. They are protecting a process that would not survive being looked at.

Inconsistent criteria. Decisions that would have gone differently for someone better connected. Rationale constructed after the fact to justify a conclusion already reached.

Expose that, and you have not built trust. You have documented the problem and put your name on it.

I think this is why so many transparency programmes quietly die. They are layered on top of a process nobody has fixed, and the first genuinely awkward decision kills them.

So the sequence matters. Fix the process. Then make it visible. Not the reverse.

The Four Things in the Notebook

Which brings me back to Anika's diagrams, because each one is an artefact of a process that can withstand inspection.

Decision visibility. Who decided, when, and on what basis — recorded somewhere findable, not living in the memory of whoever was in the room. If the only record of a decision is what people remember, you do not have a process. You have a folklore.

Context-sharing. The constraints you were actually operating under. Almost every decision that looks stupid from outside looks reasonable once you know the three things the decision-maker could not change. People do not need to agree with your constraints. They need to know they existed.

Rationale documentation. Why this, and not the obvious alternative. The alternative matters more than the choice. Anyone can justify what they picked; the credible thing is showing you seriously considered what you rejected, and why.

Stakeholder loops. Who was consulted, and what happened to what they said. This is the one most often faked. Asking for input and then never referring to it again is worse than not asking — it teaches people their input is decorative, and they stop giving you the real version.

None of that requires a platform, a tool, or a programme. It requires writing things down at the moment you decide, when it is inconvenient, rather than reconstructing them later when it is comfortable.

Two Honest Caveats

The research also carries a warning I would not skip.

There are diminishing returns. Where a process is absent entirely, small additions improve perceived fairness dramatically. Past a certain point, more process buys very little — and I have seen organisations well past that point, where documentation had become a ritual performed to protect people rather than to inform them.

So this is not an argument for more governance. It is an argument for enough of it, applied where decisions actually get made.

The second caveat is that explanation is a separate skill from process. Researchers separate procedural justice — was the process sound — from informational justice, which is whether people received an honest, adequate and timely explanation.

You can run a genuinely fair process and still damage trust by explaining it badly, or late, or not at all. The process earns the right to be trusted. The explanation is what delivers it.

The Test I Use

Two questions, borrowed almost directly from Leventhal's list, and both uncomfortable in the right way.

Consistency: would this decision have been the same if it had involved someone I like less — or someone more senior?

Correctability: if we have got this wrong, what is the route by which we would find out and reverse it? And does the person affected know that route exists?

If the answer to the first is no, the problem is not transparency. If there is no answer to the second, you do not have a process — you have a verdict.

Transparency is not the speech you give after deciding.

It is whether the deciding itself was built to be seen.