Designing Prep Sheets That Cut Restaurant Waste 15–20%

A prep sheet is the most under-designed document in most restaurant kitchens. It is usually a list of items, sometimes with quantities, sometimes with par levels, almost never tied to forecasted demand. The prep cook works through the list, fills the containers, dates the labels, and leaves. Whatever is unused at the end of the day either rolls forward, gets tossed, or quietly migrates to staff meal.

The waste that flows through a poorly-designed prep sheet is invisible on a daily basis and devastating on a monthly basis. A typical full-service independent in the DMV wastes 4–6% of food cost through over-prepping. Three points off COGS at a $2.4M restaurant is $72,000 a year. The prep sheet — properly designed — is one of the highest-ROI single interventions in restaurant operations.

This post is the prep sheet redesign we install when starting an operations engagement. The structure works at independents from QSR to full-service. The discipline takes 60 days to install and pays for itself in less.

What a prep sheet is actually supposed to do

The prep sheet has four jobs:

1. Translate forecasted demand into prep quantities for the upcoming service period
2. Sequence prep tasks in an order that respects equipment constraints (oven space, cooler space, mixer availability)
3. Allocate prep responsibility to specific staff members on the shift
4. Generate a record of what was prepped and how much was used

Most prep sheets do job 3 (loosely) and not the other three at all. Job 1 — translating forecast into quantity — is the highest-value job and the one most consistently missing.

A prep sheet that does not start from a forecast is a record of what someone thought might happen. A prep sheet that starts from a forecast is a tool for deciding what to make. The two are not the same.

The forecast — the foundation everything else sits on

A prep sheet that does its first job needs a daily forecast. The forecast can be sophisticated or simple. Both work; neither one is "no forecast."

The simple version, suitable for any independent operator:

For each menu item, calculate a 4-week trailing average of unit sales for the corresponding daypart and day of week. A Saturday dinner forecast for the salmon entrée is the average number of salmon entrées sold on the last four Saturday dinners. A Tuesday lunch forecast for the chicken bowl is the average of the last four Tuesday lunches.

Adjust the trailing average for known anomalies:

• A weather forecast above 90°F or below 30°F adjusts cold-item or hot-item demand by ±15–25%
• A known event nearby (game, concert, conference) adjusts walk-in demand
• A school calendar change (back-to-school week, spring break) adjusts family-dining concepts
• A reservation count above or below normal adjusts the entree mix by 10–20%

The forecast is not perfect and does not need to be. A forecast that lands within 15% of actual is a substantial upgrade over no forecast at all.

The sophisticated version uses POS-integrated forecasting tools that pull data automatically. Several DMV operators run on Toast, Square, or a custom POS overlay that produces daily forecasts. These work. They also cost more than a 4-week trailing average. For an independent operator the trailing average is fine.

Translating the forecast into prep quantity

Every menu item has a yield ratio between sold quantity and prep quantity. The salmon entrée is one 6oz portion off a 7oz raw portion (after trim). The pasta primavera uses 4oz of mixed prepped vegetables. The Caesar salad uses 5oz of romaine and 1oz of dressing.

The prep sheet translates forecast quantity into prep quantity using the yield ratio:

• Forecasted Saturday dinner salmon sales: 42 portions
• Prep quantity needed: 42 × 7oz = 294oz = 18.4 lb raw salmon
• Yield buffer (15% for portion variability, returns, comps): 21.2 lb
• Existing on-hand from Friday: 6.5 lb
• Order/prep quantity for Saturday: 14.7 lb

That math, run across every prepped item every day, is what produces a working prep sheet. The math is unsexy. The savings are not.

The prep sheet template

The redesigned prep sheet has eight columns per row:

1. Item name
2. Par level (the quantity that should be on hand at service start)
3. On hand (the count at the start of the prep shift)
4. Forecasted sales (today's expected unit sales for this item)
5. Yield-adjusted prep (the quantity that must be prepped today)
6. Assigned to (which prep cook owns this task)
7. Time slot (when this task should be completed within the prep shift)
8. Used at end of day (the actual quantity consumed during service)

Columns 1–5 are filled in before the prep shift starts, by the manager or chef. Columns 6–7 are filled in by whoever is sequencing the prep. Column 8 is filled in at the end of the service day.

The eighth column is the column most prep sheets don't have. It is the column that produces the feedback loop. Without it, the prep sheet is fire-and-forget. With it, the prep sheet becomes a learning instrument.

The Sunday and Wednesday reviews

Once the prep sheet has column 8 data for a few weeks, two recurring reviews happen.

Sunday review (15 minutes). The manager or chef pulls the prep sheets from the previous week and looks at the difference between "yield-adjusted prep" (column 5) and "used at end of day" (column 8) for every prep item. For each item, the question is: did we over-prep, under-prep, or hit the target?

Items consistently over-prepping by more than 10% need either:

• A lower par level (the par was set wrong)
• A lower yield buffer (the buffer was set too generously)
• A lower forecast (the trailing average is overstating demand)

Items consistently under-prepping mean either:

• A higher par level
• A higher yield buffer
• A higher forecast

The Sunday review is the discipline that gets prep quantity right. Most kitchens have prep par levels that were set years ago and never revisited. Sunday review revisits them with current data.

Wednesday review (10 minutes). The Wednesday review looks at the prep sequencing. Was the cold prep done before the hot prep? Did the oven get used efficiently? Did anyone wait for equipment? The prep cooks themselves are the best source of this data. A 10-minute conversation produces sequencing changes that compress the prep shift by 10–20% over 6–8 weeks.

The prep sheet that gets reviewed every Sunday and adjusted every Wednesday is a prep sheet that gets better every week. The prep sheet that gets printed every day and never reviewed is a prep sheet that stays exactly as wrong as it was the day it was designed.

Specific waste-reduction tactics

Several specific techniques compound on top of a well-designed prep sheet.

Tactic 1: Tiered par levels by day

Most kitchens run one par level per item. A 15-quart sauce par on a Tuesday is the same 15-quart par on a Saturday. The volume isn't the same, so the par shouldn't be.

The fix is tiered pars: a low par for Monday/Tuesday, a medium par for Wednesday/Thursday, a high par for Friday/Saturday/Sunday. This reduces the rollover/dump cycle on slow days and reduces the run-out risk on busy days.

Tactic 2: Just-in-time for high-perishable items

For items with a sub-24-hour usable life (cut produce, certain seafood preps, certain dairy mixes), shift the prep from once-daily to twice-daily — small morning prep and small afternoon prep. The afternoon prep is sized to actual reservation data for the night, which is known by 2:00pm.

This adds 25–35 minutes to the prep schedule but reduces waste on high-cost perishables by 30–50%. On a high-volume protein it can pay for itself in the first week.

Tactic 3: Cross-utilization tracking

For items that can roll forward into a different application (today's roast chicken into tomorrow's chicken salad; today's bread into tomorrow's croutons), the prep sheet should explicitly track the roll-forward. Items that get used in a roll-forward application are not waste; items that don't get rolled forward and get tossed are.

The tracking is simple — a "rolled to" column on the prep sheet end-of-day section. Over 30 days the data shows which roll-forward applications work and which produce no usage. The ones that produce no usage get eliminated; the prep doesn't roll forward, it doesn't get prepped.

Tactic 4: Server-driven mid-service prep adjustments

On long dinner services (more than 4 hours of seating), give the lead server access to the mid-service "what's running low" view from the kitchen. If a specific dish is running ahead of forecast, the server team is the earliest signal — they have reservation counts for the second seating and can see the ordering pattern.

A 4:30pm forecast adjustment is too late. A 7:15pm forecast adjustment based on first-seating data, with one more hour of prep capacity, is exactly the right moment to top up the items running hot.

What this looks like at 60 days

Operators who install this prep sheet discipline and hold it for 60 days typically see:

• 2–4 points off food cost (the headline number)
• Faster prep shifts: prep cooks finish 30–45 minutes earlier on average because the prep work is sequenced and sized to actual need
• Cleaner walk-ins: less leftover prep means less rollover means less first-in-first-out confusion
• Fewer 86s mid-service: forecast-driven par levels track demand more closely than legacy pars, so run-outs are less common

The food cost savings is the most visible. The cleaner walk-in is the most operational. The fewer 86s is the most guest-facing.

See weekly COGS variance reporting for the financial-side discipline that surfaces whether the prep sheet changes are actually moving the COGS line, or whether the waste is migrating somewhere else (most often: portion creep at the line).

Common implementation problems

Problem 1: The chef resists the forecast

A chef who has set prep pars by intuition for years often resists the idea that a trailing-average forecast should drive prep quantity. The resistance is understandable — intuition has real information in it, especially about menu items with unusual demand patterns.

The fix is to use the chef's intuition as the override, not as the base. The base prep is set by forecast. The chef can override any item by 15–20% for known reasons (a particular preparation that needs longer cure time, an event the forecast doesn't capture). The override is documented in the prep sheet.

This structure preserves the chef's expertise while creating a data-driven baseline that becomes the discipline.

Problem 2: The prep cooks resist column 8

End-of-day "used" data requires someone to track what was used during service. The most accurate source is the closing cook's count. Prep cooks who finish at 3:00pm aren't there to capture it.

The fix is that column 8 is filled in by the closing kitchen manager during the post-service walk, not by the prep cook. The data flows back to the prep sheet binder for Sunday review.

Problem 3: The Sunday review keeps getting skipped

The single most common reason this discipline fails is that the Sunday review is treated as optional. It is not. It is the entire reason the prep sheet exists.

The discipline that prevents skipping: the review is on the calendar at the same time every week. The output is a one-page summary (par changes for the upcoming week, items that needed adjustment, sequencing notes). The summary goes to the prep cooks Monday morning.

The summary is the artifact. The summary is what makes the discipline visible. Without it, the review erodes within 3–4 weeks.

When the prep sheet is the wrong place to start

In two cases, the prep sheet is not the right first project.

Case 1: The forecast doesn't exist at all

If POS sales data is not captured at the menu-item level, or if the data is in a system that cannot produce historical daypart-by-day-of-week summaries, the forecast cannot be built and the prep sheet cannot do its first job. The first project is then data integration — getting the POS data into a form that can produce a daily forecast. Prep sheet work follows.

Case 2: The kitchen leadership is not stable

A prep sheet redesign requires a chef or kitchen manager who will be in the role for at least the 60 days of installation. If the chef just left, or is about to leave, the redesign is wasted effort. Stabilize the leadership first.

Getting started

Three steps in the next two weeks.

Week 1: Pull the last 8 weeks of menu-mix data from your POS. Calculate a 4-week trailing average for each menu item by day of week. This is the forecast.

Week 2: Build a prep sheet template with the 8 columns described above. Apply it to one daypart for one week. Fill in columns 1–5 from the forecast; let the kitchen fill in 6–8.

Week 3: Run the first Sunday review. Adjust the par levels and yield buffers based on the first week's column-5-vs-column-8 comparison. Roll out to the full menu the following Monday.

By week 8 the discipline is installed. By week 12 the COGS line should reflect it.

If you want help with the forecast model or want a second set of eyes on the prep sheet redesign at your operation, book a discovery call. Bring a sample current prep sheet and 8 weeks of POS mix data. We will walk through the redesign on the call and tell you which items to start with for the highest first-quarter ROI.

The prep sheet is one piece of paper. It is also the cheapest 2–4 point COGS improvement an independent operator can install in a quarter. Most kitchens treat it as a checklist. The kitchens that treat it as a forecasting instrument run measurably better.