The whole agenda is on this page. You could run it yourself, and some of you should.

One question, agreed in writing — Not a theme. A question the day can actually answer, short enough to fit on one line, signed off by whoever will have to live with the answer
Six one-to-one calls — Thirty minutes each, before the day, with the people who'll be in the room. This is where I find out what nobody is going to say in front of the others
The numbers, dull and complete — Whatever the question depends on, in whatever state it is really in. A day spent arguing about whose figure is right is a day wasted, and it is the most common way this fails
Who is not coming — Deliberately. A room of fourteen is a presentation; the day needs six to eight and needs them there the whole time, not dropping in for their bit
The decision rights, stated out loud — Who actually decides at the end. If the honest answer is somebody who won't be in the room, we fix that before booking anything

09:30 — The question on the wall — Written up, read out, and challenged for twenty minutes before anything else happens. About one day in four changes its question here, which is the cheapest change that will ever be made
10:00 — What we already know — The numbers, presented by whoever owns them, with the gaps named as gaps. No slides longer than five minutes and no recommendations yet
11:15 — The uncomfortable list — The things raised in the one-to-ones that nobody has said in a group. I read them out, unattributed. This is the part people remember and the part that needs an outsider
12:30 — Lunch, and not a working one — Forty-five minutes away from the room. Working lunches feel efficient and reliably ruin the afternoon
13:15 — Three options, written — Small groups, one page each, including what each option gives up. An option with no cost attached hasn't been thought about
14:30 — Argue properly — The options against each other, out loud, with me making sure the quiet people speak and the senior person speaks last
15:30 — Narrow it — Down to one, or down to one plus a thing we'll test first. "We'll do all three" is not an outcome and I will say so
16:00 — Write it down before anybody leaves — The decision, the name against it and the date. On the wall, photographed, sent that evening

A decision, in one sentence — Written in the room, in plain words, short enough to be repeated accurately by somebody who wasn't there
A name against it — One name, not a department and not a committee. A decision owned by everybody is a decision owned by nobody and it will still be undone in March
A date — The date something visible happens, not the date of the next meeting about it
The things we deliberately are not doing — Written down with the same weight. Half the value of a strategy day is the permission to stop something, and it evaporates unless it is on the page
What would change our mind — Named in advance, while nobody is defending anything yet. It is the only cheap moment to agree what would count as being wrong
No deck afterwards — You get the photograph of the wall and one page typed up that evening. A deck takes a fortnight, arrives after the energy has gone, and is where decisions go to die

What you're actually paying for — Somebody with no stake, no history and no next appraisal riding on the answer. That is quite hard to find internally, and impossible to fake
The one-to-ones are the real difference — People tell an outsider things they will not put in a group, and they tell somebody who is leaving afterwards more than they tell somebody who isn't
Timekeeping is a job — Somebody has to end the eleven o'clock session while it is still interesting. Nobody internal enjoys doing that to their own colleagues
The senior person gets to participate — Rather than chair. Most of the value of hiring this out is that the person who usually runs the room gets to be in it instead
When to just do it yourself — A question with an obvious answer, a small team that already trusts each other, or a decision that is genuinely already made. All three are common and none of them needs me

I won't run a day whose decision is already made — It happens, and it is usually well meant — a decision taken upstairs needing to look consulted. Everyone in the room can tell, and the cost is that they trust the next real consultation less
I won't write your strategy — I don't know your business and one day would not fix that. The room writes it; I keep the room honest and on time
I won't stay for the implementation — The day ends and so does my involvement, apart from the typed page that evening. Anything else would make me a stakeholder in the answer
I won't take work that follows from it — No consulting engagement, no retainer, no introduction fee to anybody you go on to hire. If the day points at more work, I am the wrong person to be pointing
I won't report to one person privately — Nothing gets said to whoever booked me that wasn't said in the room. The one-to-ones are confidential in exactly that direction as well
I won't do it for fourteen people — Above eight it stops being a working session. Where fourteen genuinely need to be there, what's wanted is two days and a different shape

The third option is not a trick — It is bought a few times a year and it works. If you have somebody who can hold the room, the pre-work is the half that is genuinely hard to do from the inside
Travel and the room are yours — Booked by you, paid by you. I'd rather not mark up a train ticket, and somewhere off site is worth more than somewhere impressive
One rescheduling, free — Things happen. The second one is charged at half, because the one-to-ones will have gone stale and have to be redone
Charities and very small teams — Ask. There is a lower number and it is not means-tested beyond a straight conversation about it