Beneath MarketingSeason 1Episode 3
The campaign worked. The branch collapsed.
In short: A campaign fills the hall but adds no chairs. Little’s Law, the slowest station, the psychology of waiting, and the 80% capacity rule before any campaign.
Why a campaign should ask the kitchen’s permission before it launches
Abstract
A campaign raises the number of people who turn up. It doesn’t raise the speed of service. Once a branch nears its limit, every extra customer makes everyone wait longer. This episode covers three simple ideas that make marketing ask operations before it hits “publish”.
Before you sign: The same rule sits in the Branch Gate: a branch above 80% of its peak capacity does not need a campaign, and a branch below it does not need a new neighbour. So before you sign a campaign budget or a lease nearby, look at the peak.
When to come back to this episode: Before any campaign on a branch close to its peak limit, and before the seasons: Ramadan, Eid and summer.
After isha
Composite example, not a specific client
Thursday night, after the isha prayer. A Snapchat influencer posted about the new burger that afternoon. By 9 pm the drive-thru line spills into the street. In the kitchen, one screen takes both car orders and app orders, and delivery riders are standing at the door. A family in the dining room has been waiting half an hour and watches the riders collect their orders first.
At 11 the manager pauses the app “due to high demand”. A week later, the rating drops.
At the Sunday meeting, marketing says: “The campaign worked, demand jumped.” Operations says: “The branch collapsed.” Both are telling the truth.
The wedding hall
We’ve all been to a wedding where the family sent out more invitations than there were chairs. The invitations worked: everyone came. But guests stood by the door, dinner ran late, and the coffee never reached everyone.
On the drive home, nobody talks about how lovely the invitation card was. Everyone talks about the crush.
The campaign is the invitation; the branch is the hall. The number of chairs doesn’t grow because the invitation was beautiful. The better the invitation works, the more people end up standing.
Three ideas beneath the crowd
One: the arithmetic of the queue. In 1961, John Little proved a simple relationship: the number of people in a queue equals the rate at which they arrive multiplied by the time each one spends in it (Little’s Law).1 Illustrative arithmetic: if two cars arrive every minute and each spends ten minutes from joining the lane to leaving the window, you have twenty cars in the lane on average. A campaign only raises the first number. The kitchen decides the second.
ChartIllustrative arithmetic
Little’s Law in the drive-thru lane
Data as a table
| Value | |
|---|---|
| 2 cars arrive per minute | 2 |
| 10 minutes from joining the lane to leaving the window | 10 |
| 20 cars in the lane on average | 20 |
And waiting doesn’t grow gently. When a branch runs close to full capacity, a small delay has no slack in which to recover, so it piles up, and the line grows faster than demand itself. That’s why the doctrine sets the limit at 80% of peak capacity, not 100%.2
SimulatorIllustrative
The wedding hall: invitations vs chairs
Send more invitations to a hall with fixed chairs: the branch’s dot climbs the waiting curve, and near 80–100% the line explodes from two minutes to half an hour.
Result at the default values (turn on JavaScript to move the numbers).
- Utilisation before / after the campaign
- 75% → 97.5%
- Average wait before / after
- 2.3 min → 29.2 min
- Orders in the line after the campaign
- 38.0
- Orders still waiting when the peak ends
- none
- To get back to 80%
- +17.5 orders/h, or redirect 14
Watch 97.5% at peak: add capacity before the campaign (+17.5 orders an hour at the slowest station), or redirect 14 orders an hour to quieter hours, a nearby branch or pickup.
Rule source: Revenue Engineering Doctrine: Principle 3, scale follows capacity
Single queue with random arrivals (M/M/1): the shape is general, your branch’s minutes will differ. The 80% and 100% lines are the Doctrine’s rules.
Data as a table
| Utilisation | Wait (min) | Orders waiting |
|---|---|---|
| 50% | 0.8 | 0.5 |
| 60% | 1.1 | 0.9 |
| 70% | 1.7 | 1.6 |
| 75% | 2.3 | 2.3 |
| 80% | 3.0 | 3.2 |
| 85% | 4.2 | 4.8 |
| 90% | 6.8 | 8.1 |
| 95% | 14.2 | 18.0 |
| today · 75% | 2.3 | 2.3 |
| after the campaign · 97.5% | 29.2 | 38.0 |
How it’s calculated
- Utilisation = orders per hour ÷ what the slowest station serves per hour
- Wait before service = U ÷ (1 − U) × (60 ÷ capacity) minutes
- Orders waiting = U² ÷ (1 − U); Little’s Law: L = λW
- At 100% and above the line never settles: it grows by (demand − capacity) every hour
Two: the slowest station runs the whole branch. In his novel The Goal (1984), Eliyahu Goldratt showed that a production line moves only as fast as its slowest station, which he called the constraint.3 In a branch it might be the grill, the payment window or the riders’ pickup shelf. Every order a campaign adds ahead of that station doesn’t add sales; it lengthens the line in front of it. And every hour lost there is an hour lost for the whole branch.
Three: waiting is a feeling before it’s minutes. In 1985, David Maister wrote about the psychology of waiting lines: unoccupied time feels longer than occupied time, uncertain waits feel longer than known ones, unexplained waits longer than explained ones, and unfair waits longer than fair ones.4 A family watching a rider collect before them isn’t counting minutes. They’re counting the unfairness.
That’s good news, because part of the fix is cheap: a screen showing the expected time, a riders’ shelf out of the families’ sight, a staff member walking the drive-thru line and taking orders early. None of these adds a minute of capacity, but they can make the wait itself feel lighter.
James Heskett and his Harvard colleagues tied the whole chain together: the service experience builds customer loyalty, and loyalty builds profit.5 In other words, a crush isn’t just an operations problem. It’s a deferred marketing cost.
And delivery isn’t a side channel. A large share of orders in the major cities now comes through the apps,6 and all those orders go over the same grill at the same peak. That’s why we count delivery inside capacity, not outside it.7 In Ramadan the peak comes twice: before iftar, and again late, around suhoor. And after Eid, when restaurants and cafés were the only category with a positive change in spending in the week after Eid al-Fitr 2025,8 the temptation is to launch a campaign into what is already the busiest week.
Where the money leaks
In the roundabout episode we called this place the closed door: advertising sends people to a branch that can’t serve them. Now we know why the door shuts: the hall filled up and one station choked.
ChartIllustrative
The same campaign, three branches
- before
- after
- 80%
Illustrative: the same campaign, +15% orders; the slowest station serves 100 orders an hour; M/M/1 model
Data as a table
| Branch | Utilisation before | After | Wait before (min) | After (min) |
|---|---|---|---|---|
| Relaxed branch | 60% | 69% | 0.9 | 1.3 |
| Busy branch | 75% | 86.3% | 1.8 | 3.8 |
| Nearly full branch | 85% | 97.8% | 3.4 | 26.1 |
Demand you can’t serve, you pay for twice: once in the campaign that brought it, and again in the second visit that never happens.2 But there’s a third, less visible cost. The campaign brings new customers on your worst night. Their first experience of you is the longest line, the late order and the paused app. A campaign doesn’t just choose your customers; it chooses the night they meet you.
The rule is simple: when a branch hits 80% of peak capacity, add capacity or redirect demand before any campaign, to quieter hours, a nearby branch or pickup. Above 100%, stop campaigns for that branch.2
The Sunday-morning question
One question, one metric, one rule
| The question | “Can the branch’s slowest station handle the demand this campaign will bring?” |
|---|---|
| The metric | Peak-hour utilisation: orders in the hour divided by what the branch can serve in an hour, delivery included. Plus average wait and hours the app was paused |
| The decision rule | At 80%, add capacity or redirect demand before any campaign. Above 100%, no campaign for that branch. |
This week
- Stand in the branch on Thursday after isha with a stopwatch: count cars and orders every fifteen minutes, and time from order to pickup.
- Find the slowest station: where do orders pile up? Then separate the riders’ wait from the families’ wait, and show the customer an expected time.
- Make every upcoming campaign need the operations manager’s sign-off on the list of target branches before it goes live.
Chairs before invitations
Marketing can fill the hall. It can’t add a chair. So campaign planning starts from the slowest station in the kitchen, not from the ad.
None of this is new: Little (1961), Goldratt (1984), Maister (1985), Heskett and colleagues (1994). The only new thing is reading them in a drive-thru line after isha.
And yet some customers waited in that line and didn’t drive to the place next door. Why? Why do people pass three cafés to come to you? That’s the next episode.
Sources and notes
John D. C. Little, “A Proof for the Queuing Formula: L = λW”, Operations Research 9(3), 1961, pp. 383–387. doi.org/10.1287/opre.9.3.383 (opens in a new tab)
Revenue Engineering Doctrine: Principle 3, scale follows capacity. 123
Eliyahu M. Goldratt & Jeff Cox, The Goal: A Process of Ongoing Improvement, North River Press, 1984.
David H. Maister, “The Psychology of Waiting Lines”, in J. A. Czepiel, M. R. Solomon & C. F. Surprenant (eds.), The Service Encounter, Lexington Books, 1985. davidmaister.com/articles/the-psychology-of-waiting-lines/ (opens in a new tab)
James L. Heskett et al., “Putting the Service-Profit Chain to Work”, Harvard Business Review 72(2), 1994. hbr.org/2008/07/putting-the-service-profit-chain-to-work (opens in a new tab)
Rest of World (2025) and Redseer, on the growth of app-based delivery in Saudi Arabia. restofworld.org/2025/delivery-app-keeta-hungerstation-saudi-arabia/ (opens in a new tab) and redseer.com/articles/keeta-has-gained-10-market-share-in-ksa-whats-next/ (opens in a new tab)
Revenue Engineering Doctrine: channel rules, delivery is part of capacity.
Arab News, citing SAMA data, week ending 5 April 2025. arabnews.com/node/2596397/business-economy (opens in a new tab)
Cite this
English
Alhajj, S. (2026, September 29). The campaign worked. The branch collapsed.. Beneath Marketing, season 1, episode 3 (BM S1.E3). seiflden.com. https://seiflden.com/insights/wedding-hall-capacity/
Arabic
الحاج، سيف الدين (29 سبتمبر 2026). «الحملة نجحت والفرع انهار». «تحت التسويق»، الموسم 1، الحلقة 3 (BM S1.E3). seiflden.com. https://seiflden.com/ar/insights/wedding-hall-capacity/