How do I calculate my kitchen throughput cap?

Throughput cap = (productive kitchen hours × dishes per hour per station) limited by the bottleneck station. Example: 2 tandoor + 4 burner range, 8 productive hours, bottleneck is tandoor at 40 naan/hour = 320 naan max. If average order needs 2 naan, cap is 160 orders. Most kitchens never measure this — they discover it when orders fail.

Should I stop taking aggregator orders when at capacity?

Yes — temporarily snooze or extend prep time on aggregators before you fail orders. A 45-minute delay notification retains the customer; a cancelled order after 60 minutes loses them permanently and hurts your aggregator ranking.

What is menu simplification during peak?

Cut to 40-60% of SKUs during peak demand windows. Keep highest-margin, fastest-prep items. A 60-SKU menu at 3× demand fails; a 25-SKU set menu at 3× demand succeeds. Festival set menus exist partly for throughput, not just marketing.

How much extra staff actually increases throughput?

Adding FOH does not increase kitchen throughput. Adding a trained cook on the bottleneck station increases throughput 20-40%. Adding untrained help in a packed kitchen decreases throughput 10-15% (congestion, errors, rework).

Kitchen throughput cap: when your forecast exceeds kitchen capacity

Every festival season, someone posts a revenue screenshot — 4× normal day — and every kitchen manager quietly knows half those orders failed or went out cold. Demand forecasting tells you how many customers want to eat; kitchen throughput cap tells you how many you can actually serve. When forecast exceeds capacity, the playbook is not 'work harder' — it is throttle, simplify, and manage demand honestly.

Forecast vs capacity: the gap

A demand forecast (from POS history, festival calendar, or cuisine demand signals) tells you expected covers. Kitchen throughput cap tells you maximum covers at quality. The gap between them is where restaurants fail:

Scenario	Forecast	Capacity	Gap	What happens without a plan
Normal Saturday	120 covers	140 covers	None	Smooth service
Diwali evening	280 covers	160 covers	+75%	60-90 min waits, cold food, staff quit
IPL final (bar + kitchen)	200 covers	120 covers	+67%	Aggregator cancellations, 1-star reviews
Ganesh visarjan day	350 covers	180 covers	+94%	Kitchen meltdown, society complaints

Find your bottleneck station

Kitchen throughput is limited by one station, not average capacity. Identify it:

Tandoor-heavy menu — tandoor is almost always the bottleneck. 30-50 naan/hour per tandoor; 15-25 tandoori portions/hour.
Biryani-focused — dum cooking limits batch size. One 50-portion handi every 45-60 minutes.
Fry-heavy QSR — fryer basket cycle time. One basket = 4-6 portions every 4-6 minutes.
Wok / tawa (Chinese, South Indian) — usually highest throughput station. Rarely the bottleneck unless menu is 80%+ wok items.

Time your bottleneck station for 30 minutes during a busy service. Count output. Multiply by productive hours. That number is your real capacity — not what you wish it were.

The 5-step throttle playbook

Set a hard cover cap — when forecast exceeds capacity by >20%, cap reservations and walk-ins at capacity × 0.9 (10% buffer for timing variance).
Simplify the menu — cut to a set menu or 40-60% of SKUs. Remove bottleneck-heavy items first (see substitution maps in crisis playbooks).
Throttle aggregators — increase prep time, snooze high-complexity items, or pause the platform 30-60 minutes during peak. Better than failing orders.
Pre-batch the bottleneck — par-cook tandoor items, hold biryani handis, pre-fry base batches. Reduces per-order time by 30-50%.
Communicate wait times — FOH tells walk-ins the honest wait. 45-minute wait with a complimentary drink retains customers; silent 90-minute waits do not.

Menu simplification matrix

At capacity	Action	Example
<80% capacity	Full menu	Normal operations
80-100% capacity	Remove lowest-margin, slowest-prep items	Drop 10-15 SKUs
100-130% forecast	Set menu or festival thali only	25 SKUs → 8-item set menu
>130% forecast	Set menu + cap covers + throttle aggregators	Diwali 8-item thali, max 160 covers

Festival playbooks like Diwali and Ganesh Chaturthi recommend set menus partly for this reason — set menus are a throughput tool, not just a marketing device.

Staffing for throughput (not just headcount)

Add at the bottleneck — second tandoor operator, extra fryer, parallel biryani handi. This is the only staffing move that increases output.
Two-shift kitchen — for festival weeks, run 6 AM-2 PM prep shift + 4 PM-midnight service shift. Extends productive hours without overtime burnout.
Do not crowd the kitchen — extra untrained bodies in a packed kitchen slow everyone down. Better to have 6 skilled cooks than 10 with 4 watching.
Pre-assign stations — fixed station assignment for peak service. No mid-service station shuffling.

When to turn away revenue

Turning away covers feels wrong but is correct when the alternative is failed service. A restaurant that serves 160 covers well earns more repeat revenue than one that serves 280 covers badly. Failed aggregator orders cost ranking, refunds, and permanent customer loss.

If your forecast is 2× capacity or more, do not run promotions, aggregator discounts, or marketing pushes. You are already oversubscribed. Marketing into a throughput wall is how festivals become reputation disasters.

Build throughput into your forecast loop

Every demand forecast should be paired with a throughput check before you staff, prep, or market. The salary week demand guide covers the opposite problem (excess capacity); this post covers the peak. Together they bracket the monthly operating range.

Forecast demand vs your kitchen capacity →