Six Thinking Hats in a Modern Workshop

Six Thinking Hats is a facilitation method in which the entire group deliberately thinks in one mode at a time - facts, feelings, risks, benefits, ideas, process - instead of doing all six at once and calling it a discussion. Edward de Bono called it parallel thinking, and the practical claim is simple: a room that argues about whether an idea is good while also debating what the numbers say and how everyone feels about it will produce heat rather than a decision.

The version most people have encountered is not that. It is a workshop where six people are each handed a coloured hat and told to play a role for an hour, which converts a thinking discipline into typecasting and gives the sceptical person a licence to be difficult all afternoon. That version deserves the eye-rolling it gets.

This guide covers what each hat is actually for, the sequences that work for different session types, a sixty-minute script you can run as written, how to run it remotely without the props, where the method genuinely fails, and how to get from six columns of output to a decision somebody owns.

The six hats, and what each one is really asking for

Each hat is a mode of attention, not a personality type. The instruction is always the same shape: for the next few minutes, everyone contributes only this kind of thinking, and anything that belongs to another hat gets parked until that hat comes round.

Why one mode at a time beats a normal discussion

In an ordinary meeting, the person who speaks first sets the mode. If they open with a risk, the next four contributions are risks, and an idea that was worth exploring dies in minute three. If they open with enthusiasm, the room's genuine concerns get held back because raising them now looks like being negative about something everyone else seems to like.

That is a mechanical problem, not a culture problem. Everyone holding the same hat removes the social cost of the mode: when the whole room is doing black hat, listing a risk is compliance rather than dissent, and when the whole room is doing yellow, finding the upside in a colleague's proposal is not flattery. The method is one of the few structural defences against a group converging early and calling it agreement.

It also fixes a sequencing problem. Groups habitually evaluate before they have generated, which kills the option that needed twenty more seconds of development. Putting green after black forces the room to respond to the objections it just raised rather than stopping at them.

Pick a sequence that matches what the session is for

There is no single correct order. Blue always opens and closes; what happens between them depends on whether you are evaluating something specific, generating options, or trying to unstick a group that has already dug in. Choose the sequence before the session and put it on the agenda so people can see the shape.

What to prepare before you run it

The hats examine something, so you need the something written down before anyone arrives. One page: the proposal, what it would cost, who it affects and what decision the session is being asked to make. A group given a vague topic will spend white hat arguing about what is actually on the table, and you will lose fifteen minutes you needed later.

Send that page and any supporting numbers 48 hours ahead. Pre-reading is what lets you compress white hat from eight minutes to three, and the compression matters because green and blue at the end are the blocks that produce the decision. Sessions run out of time at the end, never at the start.

Then pick your sequence and write the timings on the agenda. Deciding the order live looks flexible and reads as improvisation, and it makes the parking rule impossible to enforce - you cannot tell someone their objection is four minutes away if you have not yet decided when black hat is.

A sixty-minute run you can follow as written

Announce each switch out loud and put the current hat on the screen. The visible marker matters more than it should - people genuinely lose track of which mode they are in, and half of the method's value is knowing that a colleague's objection is black hat rather than opposition.

Collect each block as simultaneous written input rather than open discussion. Session Flo works well here because you can run six back-to-back open-text or word-cloud activities from a single room code, so switching hats is one click rather than a new link and thirty seconds of re-joining. It also means the black hat block is not dominated by whoever objects fastest.

Running the hats remotely and in hybrid rooms

Teaching the hats to a group that has never used them

Do not explain all six upfront. Two minutes of colour-coded theory before any work has started is the fastest way to make the method feel like a management fad, and people will have forgotten four of the six by the time you get there.

Introduce each hat at the moment you switch to it, in one sentence, with the instruction attached: 'Next four minutes are gut reaction only - one line, no justification, nobody has to explain it.' The group learns the method by doing it, and the sceptical people find they have already used three hats before anyone said the word framework.

Debrief it once at the end, briefly. Ask which block produced the most useful material and which felt artificial. Groups almost always name black hat as easy and yellow hat as unexpectedly hard, and that observation is more persuasive than any explanation you could have given at the start about why the structure exists.

Where Six Thinking Hats actually fails

The method has real limits and pretending otherwise is how facilitators end up with a room that has decided the whole thing is a gimmick. It is a structure for examining something that already exists, so it does badly as a blank-page tool: green hat with no proposal on the table produces scattered ideas with nothing to attach them to.

It also assumes a group that will accept a facilitator interrupting them. In a session with a strong hierarchy, the senior person who wants to make a point during the wrong hat will make it anyway, and if you let that pass once the structure is gone. Either get explicit agreement from that person before the session that you will park them like anyone else, or run something with less enforcement in it.

The two-hat version for ordinary meetings

You do not need the full ceremony for most decisions. The compressed version is two blocks of four minutes: everyone writes the risks, then everyone writes what would have to be true for it to work. Blue hat is you, saying at the start what the decision is and at the end what was decided.

This takes fifteen minutes including the close and it prevents the most common failure in a normal meeting - the first objection setting the tone and nobody bothering to make the positive case. Teams that use the fifteen-minute version weekly get more out of the method than teams who run the full sixty minutes twice a year and then forget it.

Turning six columns of output into a decision

Facilitator moves that keep the method honest

Your main job during the blocks is enforcement, and it needs to be light. When someone offers a risk during yellow hat, do not debate it - say 'hold that, it is four minutes away' and write it down somewhere visible so they know it will not be lost. Parking with a visible list is the difference between enforcement that feels like structure and enforcement that feels like being shut down.

Watch for the two characteristic failures. The first is black hat sprawl, where the risk block runs to twenty minutes because it is the mode the group finds easiest and most socially safe; stop it on time even mid-sentence. The second is yellow hat as politeness, where the upside block produces bland agreement rather than mechanism; push once with 'why specifically would that work?' and the quality changes immediately.

Finally, resist running the hats on something that does not need them. A group that already disagrees openly, tests its own assumptions and separates generating from judging is doing the thing the method exists to install. Using it there adds process to a room that was already working, and that is how a genuinely useful structure earns a reputation as workshop theatre.