Two weeks before the February feature, Helen runs a readiness meeting: one plan, one number, and a name next to every risk — before the first week on the new capacity tests all three.
Right — two weeks to the feature. Twenty-five percent, gondola ends, four weeks, Harlow's only. Today we leave with one plan, one number, and a name next to every risk. What we're not doing is reopening the terms. They're signed, they're good, and nobody in this room renegotiates them by accident.
Helen opens with the outcome ("one plan, one number, a name next to every risk") and then says what the meeting is NOT for. Naming the out-of-scope topic early stops the meeting drifting back into it.
For the record, Richard's already tried twice. Politely.
Of course he has. This is the first feature on the new capacity, so I want it boring. If anything is going to surprise us, I'd like to hear it today, while it's still cheap.
A kickoff's real job: surface surprises when fixing them costs little. "I'd like to hear it today, while it's still cheap" invites bad news without making anyone the messenger.
Then I'll start with the part that isn't boring, so nobody relaxes.
Hold that thought. Customer first. Edmund — where are we with Harlow's?
The chair controls the order without shutting anyone down. "Hold that thought" promises the point will come back — and in Scene 3 it does.
Dates and volume are signed. Four weeks from the Monday after the reset, an end in each of the top two hundred stores. And their new commercial director has asked to see the week-one numbers personally, so this feature is his first impression of us.
What do we know about him?
Not much. Richard described him as "numbers first, lunch never". I think that was a warning.
Good. Now — what's confirmed in writing, and what was agreed on a call?
One of the most useful questions in any kickoff. "Confirmed in writing" is a commitment; "agreed on a call" is a memory — and two companies rarely remember a call the same way.
Dates, volume, mechanic, the availability clause — all in writing. On a call, their category manager asked if we could hold back extra stock for the flagship stores. I said I'd look at it.
"I'll look at it" is a sentence Harlow's will remember as "yes". Park it — we'll come back to it. And the penalty clause is live from day one?
Polite and vague — and the customer will hear "yes". Helen doesn't criticise Edmund; she names the risk and parks the topic with a promise to return. Safer customer versions: "Let me check capacity and come back to you by Thursday."
Day one. Top lines at a hundred percent, in writing, with teeth. So a gap isn't just an awkward phone call any more — it's an invoice. Personally, I'd rather have the awkward phone call.
An agreement "with teeth" has a real consequence if it's broken — here, a financial penalty. Edmund turns it into one plain image: the gap is no longer a phone call, it's an invoice.
Right, the part that isn't boring. I'm at nine hundred cases a week over base for the four weeks. But I'll be honest about the shape. A twenty-five percent mechanic on a product people already like gives a wide range. It could be seven hundred. It could be twelve hundred.
I'm going to say the thing I always say.
One number. I know. Nine hundred is the number — that's what we plan, what we order to, what Luca builds. I'm not giving you a range to plan to. I'm giving you the range so you know how much I trust the number.
Amanda separates two things people often mix up: the number the business plans to, and her confidence in it. The range doesn't replace the number — it tells the room how firmly to hold it.
And how much do you trust it?
Medium. At launch, the first week came in twelve percent over the ambitious plan — at a normal price. So here's my trigger: if sell-out runs more than fifteen percent over nine hundred for three days in a row, I don't wait for the weekly. I call it.
The full forecast in three parts: the number (nine hundred), how sure you are (medium, with evidence), and the pre-agreed signal that changes the plan (fifteen percent over, three days running). A trigger turns "let's keep an eye on it" into a decision made in advance.
Number, confidence, trigger. That's the version I want in every forecast from now on.
Can I borrow that for Richard? He asks for a number, and he always gets a number with a story wrapped round it.
Borrow it. Just don't give him the trigger.
The same structure works with the customer — but some parts stay inside the company. The trigger is an internal decision rule; sharing it with the buyer would hand him leverage.
Then here is my honest number. The co-packer carries the feature volume — up to about a thousand cases a week. Line Two keeps the core bars. At nine hundred, we are fine. Not comfortable. Fine.
Two short words with different meanings: "fine" = the plan works; "comfortable" = there's room if it goes wrong. Separating them tells the room exactly how much margin there is.
Has the co-packer actually run our bar at volume yet?
Two trial runs, both in spec. Trial runs are not four weeks. I would call it proven on paper.
Honest calibration: the evidence exists, but it hasn't been tested under real conditions. A useful phrase when you don't want to overclaim or sound alarmist.
Then the first week is the real test, not the trials.
Agreed. I'll be at the co-packer for the first run, in person. If something looks wrong, I'd rather see it on the line than read it in a report.
And at Amanda's trigger?
Fifteen over nine hundred is a thousand and thirty-five. That's above the co-packer. They can go higher, but they need ten days' notice, and a four-week feature doesn't wait ten days. So the extra comes back to Line Two — and Line Two is feeding the core range, the lines under the penalty clause.
Luca doesn't say "that could be a problem". He walks the room through the numbers step by step until the risk is obvious. Visible arithmetic persuades better than adjectives.
So a good week on the feature could empty the shelf on the bars we guaranteed.
Yes. The last time I said the next constraint out loud, it arrived a week early. So I'm saying it today, not in a crisis meeting.
"I'm saying it today, not in a crisis meeting" — raising a future risk while there is still time to prepare. It protects the speaker too: nobody can later say they weren't warned.
What do you need?
Two weekend blocks, approved now, sitting in a drawer. If the trigger fires, I open the drawer. The crew already knows — this time I asked, I didn't assume. And I've told the co-packer we may use the ten days in the second week, so they can pencil it in.
A contingency that is approved in advance but only used if needed. Asking for the approval now — not during the crisis — is the whole point of a readiness meeting.
Approved. Both blocks. Put the cost on Amanda's one page, next to the feature margin, so we can see what we're buying.
Now, Edmund — the flagship stores.
Their category manager has a point. The twenty biggest stores sell about a third of the volume. If one of them runs empty on a Saturday, that's the photo that lands on the new director's desk. I'd like three hundred extra cases held back for them.
Held back from where, though? Nine hundred is the whole number. If I ring-fence three hundred for twenty stores, the other hundred and eighty are planned short from day one. That's a gap we'd be choosing.
A calm, powerful challenge: every extra allocation comes from somewhere. Amanda doesn't say "no" — she asks the question that makes the cost visible.
Or a gap we'd be choosing somewhere quieter, instead of somewhere with a camera.
You're both right, and we're not going to settle this with opinions. What do we actually know about how the flagships sold at launch?
When two reasonable positions collide, the chair stops the debate and asks what is actually known. "We're not going to settle this with opinions" is firm without taking sides.
Very little. The shared sell-out feed only starts this month.
Then it's an assumption, not a fact, and we write it down as one. No ring-fence today. Amanda — on day five, flagship sell-out against the rest. If they're running ahead, we move stock then, with data. Edmund, tell the category manager exactly that: we're watching the flagships daily, and we'll act on day five. Not "I'll look at it".
Instead of forcing a decision without evidence, Helen turns the disagreement into a dated check: an assumption, an owner, and a day to decide. Both sides get a fair hearing, and nobody has to "win".
Watching daily, deciding on day five. That's a sentence I can say out loud.
Edmund tests the decision by imagining saying it to Harlow's. A good internal decision should translate into one clear external sentence.
Owners. Edmund, you're the only voice Harlow's hears from us. If Richard rings Amanda, Amanda sends him to you. Amanda — daily sell-out for the first two weeks, from the feed, in one line: plan, actual, gap. Luca, you own the trigger.
A kickoff assigns owners, and the most important one here is the single point of contact. If a buyer hears three slightly different stories from three people, he will use the difference.
And the one line goes to all four of us at eight in the morning — before anyone's had time to have a feeling about it.
And if Richard asks me whether we're ready for a big week?
Tell him we're reading sell-out daily and we have capacity ready to switch on. Both true. Neither one is the trigger.
Helen gives Edmund an honest customer answer that doesn't reveal the internal rule. Being truthful doesn't mean sharing every detail of how you decide.
And if the trigger fires on a Saturday?
Then you open the drawer. You don't need to ask me, and you don't need a meeting. Tell me after, in one message. The worst version of this is three people waiting for permission while the shelf empties.
Delegating decision rights explicitly. Owning a task and being allowed to act on it are different things; Helen makes sure Luca has both — and asks only to be told afterwards.
Recap. One plan: nine hundred a week. One trigger: fifteen over, three days. Two weekend blocks, approved. Flagships decided on day five, with data. Edmund talks to Harlow's; nobody else does. Last question, and I want a real answer. Three weeks from now, looking back at this meeting — what do we regret?
"Looking back, what do we regret?" asks people to imagine the failure before it happens. It surfaces the risks people noticed but didn't think were worth raising.
Trusting a medium-confidence number more than I said we should.
Not writing the trigger down.
It's written. And Amanda — yours goes on the page too. Anything else, Edmund?
Only that if it's dull, it worked.
Then let's have a very dull February.
The meeting ends where it began: Helen wanted it boring. A short closing line that repeats the goal gives everyone the same picture to walk out with.