※ This article is part of the Field Letters series, based on real inquiries from saltnfire.net readers. All names and situations have been anonymized and modified with permission.
Dear Saltnfire.net,
Hey bro. I run a small gastro pub. Recently, I turned down a reservation from a group of 15. I simply couldn’t afford to hire an extra server on short notice, and I was worried it would break the “production rhythm” you always talk about. Am I crazy? Or did I make the right call? Just wanted to hear your thoughts.
Sincerly
Dear Bro,
You made an excellent decision — one I stick to myself. While big groups are often celebrated in Korean food culture, they only work under specific conditions. And in most small pubs, those conditions don’t exist. Let’s break down why.
1. Breakdown of Heijunka – the rhythm killer
Heijunka refers to stabilizing the production flow (A → B → C) within a standard lead time. The basic principle: If you increase speed too suddenly or jam too many items into the line, you get overflow and quality defects. [See: Toyota pub principles- Heijunka]
To keep lead time short, you need parallel cooking, standardized actions, and minimal switching time between tasks. But when a party of 10 or 20 shows up, your hands can’t magically move faster. Bottlenecks form instantly. In The Goal, the character Eli Goldratt says that a factory’s true speed is set by its slowest bottleneck. And every other process must align to it.
But our restaurant is not a huge factory. For example, if pizza takes 20 minutes, everyone else starts asking, “Hey, when’s that pizza coming out?” If you don’t have a 1-equipment-to-1-menu setup, the entire kitchen slows down. That’s why my “5 dishes in 15 minutes” article worked — each dish had its own tool.
But most restaurants still use line cook–based divisions. That system causes idle labor and half-prepped ingredients when you don’t know what’s going to be ordered next. Toyota’s Taiichi Ohno would’ve lost it: “Special orders go to the back of the queue! You know that!” In restaurants, quality matters more than “hospitality.” And group orders create quality risks, idle workers, and piles of in-process food. Not ideal.
2. Labor Waste – A Financial Simulation
Large parties hit Toyota-style pubs harder than big kitchen restaurants. Why? Because small Toyota Pubs run ultra-lean: Minimal investment in equipment, fewer staff — and instead, the owner produces 2x or 3x value personally. That also means no industrial dishwasher. When a group walks in, you suddenly have 30+ dishes to wash, and now you need a dish staff and extra FOH too. Your variable cost just exploded.
From my experience, a 10-top brings in about $200. Let’s say it’s 50% food, 50% drinks — that’s typical casual dining in Korea, where limit of dining spend per income (Food Limit Index) is low. If two servers show up early, at $10/hour x 2 hours x 2 staff = $40 labor.
Now let’s model it:
| Item | Amount |
|---|---|
| Gross Sales | $200 |
| Food COGS (15%) | -$30 |
| Drink COGS (30%) | -$60 |
| Labor (2 FOH) | -$40 |
| Net Profit | $70 |
Take out taxes, card fees, and incidentals? Maybe you’re left with $50–60. And this doesn’t include your early-morning prep or the sheer physical toll. Honestly, working two hours early just to make $50 isn’t worth it. After taking 3 big parties in 3 days, I was so burned out I skipped the gym. Physically, it crushed me.
You’re working the line too — you know what I mean. A steady flow of small tables is way more sustainable. Business is a long game.
3. Why the U.S. Has It Slightly Easier – Service Charge + Tip Culture
U.S. restaurants have a bit more flexibility thanks to service charges and tipping. But it’s no picnic. From the FOH’s point of view, tips often drop with big tables — especially when checks are split or service feels rushed. Just browse Reddit — most servers say they’d rather turn tables of 2 than chase a flaky 20% on big parties. For BOH? Worse. They get slammed with 20 tickets in a flash, and then maybe — maybe — get a cut.
Imagine this:
- 10 guests spend $500
- 30% service charge = $150
- Split among 3 BOH = $50 each
For that dinner rush? I’d say no thanks. I’d rather run lean — just two chefs, stable volume, no burnout.
4. When Should You Actually Take Large Groups?
Not saying you should never take them — just be strategic.
(1) When customers cook for themselves
Korean BBQ, hot pot, malatang… these formats use bulk-prepped ingredients and guests do the cooking. Perfect for groups — the labor is outsourced to customers. But let’s be honest — most readers of saltnfire.net aren’t running those kinds of places.
(2) When the request is post-dinner — likely alcohol-focused
A party of 10 booking at 6:30pm? They want food. Avoid. But a 9:00pm group? Probably beer-focused. Beer has no labor cost, and if they ask for a pizza, just say it’s not available now. From my experience, night groups just want to drink and chat. Offer finger food: fried chicken, potato wedges, currywurst, dry snacks. I’ve never had a complaint when I quickly steered them to easy options.
(3) When business is slow or staff is idle
If your kitchen is quiet and food’s about to spoil — sure, take them. Just do it smart. Talk with staff, prep in bulk, push hit-to-serve items. Things like:
Schnitzel, goulash, chili cheese fries, nachos, mini sliders, sausage platters.
Even if cooked fresh, they require less attention.
(4) Once per week — no more
What burns out your staff isn’t the workload — it’s the rhythm change. If the team knows it’s coming, they adapt. But daily chaos? Morale killer. Limit large parties to once a week. Let staff mentally prepare. “Tomorrow won’t be like today” is a powerful thing.
5. Conclusion
You made the right move. If someone asks, “Don’t you want the money?” — just say,
“There’s no margin in it.”
If you take groups without knowing your system can handle them, You’ll burn out — fast. Running a higher-skill kitchen? Then group parties aren’t revenue. They’re risk. Even with $1,000 in revenue, I’d rather earn it over three days at $330/day than have a single blowout night.
Stable rhythm = staff retention. In my experience, when sales spike and crash like a rollercoaster, people get cranky and quit.
Sincerely,
from saltnfire.net
If you enjoyed this article, you can support the project – thank you!