The protein bar is at Gate 2, and for the first time capacity speaks first. Helen has one rule from last autumn to test: is the factory in the room before the product has a name?
Right — Gate 2 for the protein bar. One change from how we usually do this. Capacity speaks first. Not in the last five minutes, after we've all fallen in love with the idea. First.
The order of an agenda decides what the meeting pays attention to. By putting capacity first, Helen makes sure the hardest constraint shapes the discussion instead of arriving too late to change it.
I've been in this company eleven years. This is the first gate where I speak before the marketing slides.
It's my own rule from last autumn. I said capacity would be in the room from day one. Today we find out whether I meant it. Amanda — thirty seconds on where we are with Harlow's first.
Five weeks into Richard's twelve. Average availability on the oat bar: ninety-eight point six. One depot week flagged, accepted by his team. And the director's lunch went well — he asked more questions about our forecasting than about our prices.
Amanda gives exactly what was asked: where we are against the target, the one exception, and one useful observation. A short status at the start reminds everyone what's at stake before a new decision.
Good. Keep that number in your heads, because everything we decide today has to protect it. Luca.
So. Can we make it? Yes. On Line Two, the recipe runs. But there's a catch. The protein blend contains milk, so after every protein run we need a full allergen clean-down — the same as the caramel multipack. That's half a shift lost every time.
Luca answers the main question first, clearly, and then names the condition. Starting with the catch would sound like a no; hiding it would be dishonest. "Yes, but" in that order is the most useful answer a technical expert can give.
So we run it in long blocks, not little batches.
Exactly. Two weeks on, then a clean-down, then back to the oat bar. Never one day here, one day there. Small batches would cost us more in cleaning than they make in bars.
And while Line Two runs protein for two weeks, what happens to the oat bar?
That's the real question. Today, Line Two plus the co-packer's reserve keep us above ninety-eight. Take two weeks of Line Two away, and I can't promise you ninety-eight. So my honest answer is: we can make the protein bar, but not on Line Two, not until Richard's twelve weeks are finished.
Luca links the new product to a commitment the company has already made. A new opportunity is never judged alone; the real question is always what it does to the promises you're already keeping.
Thank you. That's exactly what I wanted to hear before anyone shows me a pack design. Amanda — demand.
Number, confidence, trigger — the way we agreed. The number: two hundred cases a week, in a test of one hundred stores. The confidence: low. Honestly low. Nobody has bought this product yet, and protein is a different shopper from oats.
Amanda uses the same three-part forecast as before, and she isn't afraid to say her confidence is low. A forecaster who admits uncertainty is trusted more when she says she's sure.
Where does two hundred come from?
Three sources. What similar protein bars already sell at Harlow's. How our oat bar did in its first weeks. And a discount on both, because we're new to protein and shoppers don't know us there. If anyone wants a more precise number, they'll have to wait for real shoppers.
Amanda explains her forecast in three short sources, including the adjustment she made and why. A number with a visible method is easier to trust — and easier to challenge in a useful way.
And the trigger?
Two of them, after eight weeks. If at least a quarter of first-time buyers come back and buy again, we scale up. If fewer than fifteen percent come back, we stop. Anything in between, we keep testing for another eight weeks.
Amanda sets the go and stop lines now, while nobody is emotionally attached to the result. Agreeing the rule in advance stops people from reading the numbers the way they hope to.
Stop at fifteen? That's brave. Most new products get three chances before anyone says the word "stop".
That's exactly why I want it in writing today, while we still like it.
Then here's my side. Richard told us: if we hold ninety-eight, the second facing is ours at the September reset, and then "a different conversation" about what else goes on his shelf. That's the protein bar. The September reset is the door. If we miss it, the next one is February.
Edmund explains why timing matters, using the customer's own calendar. An opportunity with a fixed date — and a clear cost of missing it — is much easier for a room to weigh than "sooner is better".
Luca, can September work without Line Two?
Only at the co-packer. They already make the oat bar for us, but a new recipe has to be qualified — trials, audits, the allergen controls. That takes twelve weeks. Last autumn, the co-packer was a spring answer to an autumn question. This time, it's a spring question, and the answer arrives just in time for autumn.
And what does qualifying cost us?
Trials, audits, a few days of their line. Less than one week of last month's penalty. I checked, because I knew you'd ask.
So if we start qualifying now, we're ready in July.
If nothing goes wrong. Twelve weeks is the plan, not the promise. I'd want a date in July that says: if we're not qualified by then, we don't launch in September. We don't squeeze it.
Luca separates a schedule from a guarantee and asks for a cut-off date in advance. A pre-agreed deadline protects everyone from the temptation to "just squeeze it" when the date gets close.
And in September, I'd want all two hundred of Richard's top stores. If we go small, we look like we don't believe in it.
And if we go big with low confidence, and it doesn't sell, we've used Richard's trust on a guess. That trust took us a crisis and a penalty to earn.
Edmund and Amanda want different things, and both explain the risk they're trying to avoid: looking weak versus spending trust. Arguing about risks keeps a disagreement professional — and makes it solvable.
Fair. But a hundred stores isn't a test Richard will notice.
Then let's make it one he will. Edmund — which hundred stores?
If I chose them? The hundred best of his top two hundred. The ones his director visits.
I can work with that. The best stores give the cleanest data anyway. And if the repeat rate passes twenty-five, we're in the other hundred by the new year.
So it's a test in the stores that matter most, with a clear path to all of them. Edmund, can you sell that to Richard?
I can sell anything that comes with a number and a date. He taught me that.
Helen doesn't choose between "small test" and "big launch". She asks one practical question — which stores? — and the answer satisfies both sides: a test that is visible, with a clear route to scale.
Decision time. At a gate we have three choices: go, no-go, or go with conditions. One word each, and if it's conditions, one condition each. Luca.
"Go, no-go, or go with conditions" gives everyone the same three options and forces a clear position. Asking for one condition each keeps the list short enough to actually manage.
Go with conditions. If the co-packer isn't qualified by the end of July, the launch moves to February. No squeezing.
Go with conditions. We start with a hundred stores, and we scale or stop on the repeat rate — twenty-five and fifteen, written down today.
Go with conditions. Nothing about the protein bar is allowed to touch Line Two until Richard's twelve weeks are done. If we lose ninety-eight, we lose the facing — and then there's no September conversation at all.
The commercial person, who wanted the biggest launch, adds the most protective condition. A good gate decision works because each person guards the risk they understand best — even when it limits what they want.
Then it's go with three conditions. The co-packer by July or we move to February. A hundred stores, scale at twenty-five, stop at fifteen. And Line Two belongs to the oat bar until the twelve weeks are over. Owners: Luca, Amanda, Edmund, in that order. I'll sign it today. Anyone not behind that?
Behind it.
Behind it. And for the record — this is the shortest gate meeting I've ever been in.
One last thing, Edmund. Richard asked us who makes it, and how many. You can answer both now.
The co-packer, and two hundred a week in a hundred stores. But I'll tell him after the twelve weeks, not before. Right now, the only thing I want him to see from us is ninety-eight percent.
Edmund has a strong answer for the customer, but decides to wait. Timing is part of the message: talking about a new product in the middle of a test could look like the team is distracted.
Good instinct. And before we go. Last autumn, we found the capacity problem five weeks before a launch, and we spent a winter paying for it. This time, capacity was in the room before the product even had a name.
Helen connects today's meeting to the mistake that created the rule. Naming the lesson at the moment it works is how a team turns one hard experience into a habit.
The Chair's own rule. I did say I'd quote it back.
Speaking of names — does it have one yet?
Marketing want something with "power" in it.
Then it can wait for Gate 3.
Well. If it's dull, it worked.
This one was dull. Thank you, everyone.
In February, the team wanted a "very dull" feature — and got a crisis. This time, the meeting is dull because the hard thinking happened early. A dull decision meeting is usually a sign that the preparation was done right.