Designing the Agenda

01 Title

On screen "Designing the Agenda" · Module 4 · Facilitation Foundations

I once spent an entire Sunday building a chest of IKEA drawers. Flat pack, forty-something parts, one of those little metal keys that ruins your thumb.

I got to about step fourteen, feeling quite good about myself, and I picked up the back panel and realised it was supposed to have gone in at step six.

So I took the whole thing apart. All fourteen steps, on the floor, in reverse.

What stayed with me is this. I hadn't used a single wrong part. Every piece was right. Every screw went where a screw was supposed to go.

The parts were never the problem. The order was.

Remember that chest of drawers.

02 A skeleton is not an agenda

On screen the five-block skeleton next to a real timed agenda: Frame, Diverge, break on the seam, Converge, Commit, Buffer 10%, Close

Module three's skeleton is five blocks with an exercise in each. Frame, diverge, converge, commit, close. That's real work and it's the hard part.

But a skeleton isn't an agenda.

A skeleton says what happens. An agenda says what happens, in what order, for exactly how long, with what happening in the gaps, and what you do when it all runs late. Which it will.

This module is where the timings become honest, the activities get chosen for the outcome instead of for novelty, and the whole thing survives contact with a room that arrives eight minutes late and spends the first quarter of an hour on the parking.

03 Backward design

On screen an arrow running right to left, artifact back to the door

Start at the end. Always.

Take the objective, the artifact the group has to leave holding, and put it on the right-hand side of the page.

Then ask: what has to be true immediately before that, for that thing to exist? And then what has to be true before that? And keep going until you reach the door.

That's backward design. Wiggins and McTighe named it in a book about curriculum in 1998, and it works the same on a Tuesday afternoon as it does across a school year.

Design forwards instead and you'll write a lovely opening, a good warm-up, a rich discussion, and then look up with eleven minutes left to produce the actual outcome. Everybody designs forwards the first time. Everybody runs out of day.

04 Every activity must earn its slot

On screen "What does the group have at the end of this that they didn't have at the start?" · "Warmth is not an output."

One question, applied to every single row on your agenda.

What does the group have at the end of this that they didn't have at the start?

Write the answer next to each row. Not think it. Write it. And if you can't answer it, cut the row.

This question will delete about a third of your first draft. It's brutal on icebreakers that exist because you felt you ought to have one, and on 'context setting' slots that are really you talking for fifteen minutes.

And here's the honest version of it, which people find slightly cold. Warmth is not an output. Warmth is lovely, safety is genuinely necessary, and module six is largely about building it. But a session can be warm all the way through and produce nothing. Hopes and fears makes people comfortable and tells you what the room walked in carrying, so it earns its slot twice. A fun fact about your weekend earns it once, and not very well.

1:06 · then: the outcome at the top, three candidates, two crossed out

Which is also how you choose in the first place. People go wrong here in a very specific way: they start with the activity they want to run and reverse-engineer a reason for it. You saw a great Crazy 8s session once, you've been waiting to use it, and now there it is on row three.

Do it the other way round. For each block, ask what the group needs to produce, then pick the exercise that produces it.

One worked pick, so you can hear the reasoning. You're converging, and there's a genuine disagreement about trade-offs. That's impact and effort, because plotting things on two axes gives the disagreement coordinates. People stop arguing about whether something is 'important' and start arguing about where it sits.

Module three's kit is ten exercises and covers every block. What matters isn't knowing forty activities. It's picking the right one from ten.

2:05 · then: "Structure beats charisma" · "Make the thinking visible"

Two rules worth repeating here, because they're design decisions rather than theory.

Structure beats charisma. A well-structured session run by a nervous facilitator beats a charming one improvising, every time, and structure is the part that survives your bad days.

And make the thinking visible. Wall, whiteboard, shared doc, doesn't matter. It exists so the group can argue with its own thinking instead of arguing with each other's memory of what somebody said twenty minutes ago.

05 The 3x rule

On screen "A five-minute discussion is fifteen minutes."

Now the timings, and here's the number that will do the most for you.

Activities take about three times longer with a group than they do in your head.

Three times. A five-minute discussion is fifteen. A quick round of introductions with twelve people is not two minutes, it's twelve. 'Let's just quickly vote' is not one minute.

Full disclosure: 3x is a rule of thumb from practice, not a measured finding. But I've never once had a session where the head number was closer than the room number.

0:35 · then: six small overheads

It's worth knowing where it goes, because that's what makes it believable.

Explaining the instruction. Re-explaining it to whoever missed it. Getting people into pairs, which somehow takes ninety seconds every time. Getting the room back, which takes longer the more fun they were having. And the tail on every activity, where two people are still writing and you have to decide whether to wait.

None of that is in your head at eleven o'clock on a Tuesday when you're writing the agenda. All of it is in the room.

06 Buffer, and where it sits

On screen an agenda with a 10% block, unallocated, sitting between Commit and Close

Ten percent of the session, unallocated. In a half day that's about twenty minutes. Write it in as a block on the agenda. Don't leave it as optimism.

Because something will happen. Discussion runs long. Someone arrives forty minutes late carrying the one objection that actually matters.

And where you put it matters as much as having it. The buffer sits between commit and close. Not before commit, and not spread evenly across the day, where it's quietly gone by mid-afternoon.

Because without buffer, here's what gets cut, every single time: the close, and then the commit block sitting right next to it. The two blocks that carry all the value of the day are the two closest to the thing that runs out. Buffer between them is what stands between an overrunning convergence and 'I'll write this up and send it round.'

07 The Rule of Thirds

On screen a bar split three ways

Rough first-draft allocation. A third divergent activity. A third convergent activity. And a third for opening, transitions and closing.

People always push back on that last third. It feels like a third of the day spent on admin.

It isn't. That third is the frame, every handover between blocks, the commit, and the close. It's where the session gets set up and where it gets landed. It's the two ends of the chest of drawers, and without them you've got a stack of panels.

0:33 · then: a three-hour agenda, thirds marked

Numbers on it. Three hours, so 180 minutes. Sixty minutes of divergent work, sixty of convergent, and sixty for the frame, the transitions, the commit and the close.

Which breaks down as roughly fifteen for the frame, five in total across the handovers, twenty-five for commit, and fifteen for the close.

Look at that commit number. Twenty-five minutes to work out who's doing what by when, on maybe four or five actions. That feels enormous when you're planning and it's almost exactly right in the room, because commitments made properly involve people negotiating dates out loud with each other.

New facilitators underbudget that last third every time, and then run out of day at precisely the moment the commitments were supposed to be made.

08 Breaks are a design decision

On screen a break placed on purpose, on the seam between diverge and converge

One thing that gets treated as admin and is actually structural: where the break goes.

A break should land on a seam, not in the middle of a block. Break between diverge and converge and you get a genuine reset. People come back ready to choose, and the break does some of the transition work for you. Break in the middle of a convergence and you lose the thread. People spend six minutes rebuilding where they were, and somebody has changed their mind in the corridor for reasons nobody else heard.

Two numbers. Anything over ninety minutes needs a break, and people will take one whether you schedule it or not. And a ten-minute break is fifteen minutes. Always. Plan fifteen and start on time.

And say the return time out loud, then write it on the wall. 'Back at twenty past.' Not 'back in ten minutes', because ten minutes starts at a different moment for every person in that room.

09 Remote changes the numbers

On screen the same agenda, two sets of timings

If the session is remote, the design changes, and not by a little.

Everything takes longer, so the 3x rule is closer to 4x for anything involving discussion, because turn-taking online has a delay built into it and people wait for gaps that never quite arrive.

Blocks get shorter and there are more of them. Ninety minutes in a room is a normal block. Ninety minutes on a call is punishing. Break it into three thirty-minute pieces with something different happening in each.

And silent written work gets better, not worse. This is the one genuine advantage. Everyone typing into a shared doc at the same time is faster and more equal than the equivalent in a room, because nobody's waiting for the pen.

So a remote session has more written work, shorter blocks, and about a third fewer agenda rows. Same five blocks. Different arithmetic.

10 The two diamonds

On screen two diamonds, Discover and Define / Develop and Deliver · "They arrive with a solution. Do you accept it?"

One more shape, and don't confuse it with Kaner. The Double Diamond, from the Design Council in 2005. The first diamond diverges and converges on the problem. The second does it on the solution. Kaner describes what happens to a group emotionally inside a single decision. The Double Diamond describes a whole design process.

Here's why that matters practically. Most groups arrive having skipped the first diamond entirely. They walk in holding a solution and want help executing it. 'We need a workshop to plan the new onboarding flow.' Onboarding is already the problem and a new flow is already the answer.

And you have a real choice. Accept the framing and design a good solution session, or send them back to the problem, which is a harder conversation and has to happen before the day rather than at eleven o'clock in front of everyone.

The question I'd ask the sponsor, once, politely: 'What makes you sure that's the problem?' If the answer's solid, design the solution session. If they hesitate, you've found your real objective.

11 The run of show

On screen an agenda next to a run of show, columns compared

Last piece. Your agenda and your run of show are two different documents and you need both.

The agenda is what participants see. Times, activities, breaks. Keep it simple. They don't need your internal machinery, and it makes the day look more complicated than it is.

The run of show is yours. Same rows, extra columns.

0:22 · then: five columns

Five columns. Time. What actually happens. The exact words you'll use to open the activity. The transition line out of it. And what you cut if you're behind.

That last column is the one people skip and the one that saves you. Decide now, on a Tuesday, calmly, which rows can go. Because at three o'clock on the day, forty minutes behind, tired, with the room looking at you, you'll make a much worse decision than the one you'd make right now.

Write the cut list before you need it. Module seven is about using it.

12 The steel-man

On screen "Isn't this all a bit rigid?" · "Plan tightly. Hold loosely."

The pushback, and it's a fair one that experienced facilitators make.

It goes: this is over-engineered. The best sessions are responsive. If you plan to the minute you'll spend the day serving your agenda instead of the room, and you'll miss the moment where something real happens because it wasn't on the sheet.

That's half-right, and the half that's right is important. A facilitator who won't deviate from the plan is worse than one with no plan, and module eight is entirely about deciding when to abandon it.

But you can only deviate from a plan you've got. What looks like graceful improvisation is almost always somebody who planned carefully, holds it lightly, and knows exactly what it costs to drop a block, because they wrote down what each block was for.

Plan tightly. Hold loosely. You can't do the second one without the first.

13 Find the collision

On screen Handout A, the flawed agenda

Exercise. This is a two-hour agenda for a product prioritisation session. Nine people, Tuesday afternoon. It looks entirely reasonable, and I've built three faults into it.

Find where divergence and convergence have been allowed to overlap. Find where the Groan Zone has been designed out. And find a transition that's never named.

In pairs. Eight minutes.

Facilitator tip: circulate but don't confirm. If somebody asks "is this one of them?", say "keep going" and move on. Pairs that get confirmation stop looking. Strong pairs also spot that the session has no frame and no commit block at all, which is the biggest fault and isn't on the list.

14 The Block Checklist

On screen eight questions, on the back of Handout B

And here's what you run over every agenda before you send it. Eight questions. Every one is a failure that's already happened to somebody, usually more than once, usually to me.

Is the outcome an artifact, not a feeling.

Are all five blocks there, and does commit have real time.

Does anything converge before the group has diverged.

Is there a Groan Zone, with time protected for it.

Is every transition written in the words you'll actually say.

Does every activity produce something.

Is there ten percent buffer, and does it sit between commit and close.

Does the hardest thinking sit in the first ninety minutes.

And if you've only got two minutes before you send it, check numbers two and three. A missing commit block and an inverted diamond account for most of the workshops that felt good in the room and changed nothing afterwards.

0:58 · then: five short lines

Five to watch for. You design forwards, because the beginning is more fun to write. You pick the activity first and justify it afterwards. You use head numbers instead of room numbers. You leave buffer as optimism instead of writing it in as a block. And you build a beautiful agenda and never write the run of show, so on the day you're improvising the cut list at the worst possible moment.

15 The practice

On screen This week's practice

Here's your one thing. Take the flawed two-hour agenda from the exercise and run the full eight-question Block Checklist over it. All three planted faults are in there. Find all three.

Then repair it. Put clock times on it, not durations. Write ten percent buffer in as its own block, between commit and close. And re-time the three longest activities against the 3x rule, honestly.

If you're holding a real agenda of your own this week, run the same eight questions over that one too.

That's the rep. It's what you'll do to every agenda you ever write.

16 Callback close

On screen the chest of drawers, assembled

I hadn't used a single wrong part in that chest of drawers. Every piece was right. Every screw went where a screw was supposed to go.

The parts were never the problem. The order was.

That's an agenda. Not a list of the activities you like. The order they have to happen in for the thing to stand up at the end.

0:24 · then: Next: Module 5 - Opening Strong and Giving Instructions

You've now got a way of turning any skeleton into a runnable agenda. Everything from here is the craft of running one well, starting with the first ten minutes, which is module five, because a beautifully designed agenda dies on the spot if nobody in the room understands what you just asked them to do.

## The practice (recap)

Run the eight-question Block Checklist over the flawed two-hour agenda from the exercise, find all three planted faults, then repair it: clock times rather than durations, ten percent buffer written in as its own block between commit and close, and the three longest activities re-timed against the 3x rule. Optionally run the same eight questions over a real agenda of your own. This is a standalone rep of the audit skill and nothing later depends on it.

## Callback

Every part was right. Every screw went where a screw was supposed to go. The parts were never the problem, the order was.

## Sources

Wiggins, G., & McTighe, J. (1998). *Understanding by Design*. ASCD. Source of backward design, Beat 3.

Design Council (2005). The Double Diamond. Beat 14.

The 3x rule, the 10% buffer and the Rule of Thirds are mine, from practice. They're rules of thumb, not measured findings.