How to Create a Restaurant Sales Forecast for Inventory and Production Schedules

Have you ever run out of ingredients on a busy day, only to have too much stock left over the next day? This often happens when purchasing and production decisions rely solely on guesswork.

A restaurant sales forecast helps businesses estimate demand more accurately and connect sales data with purchasing, inventory, and kitchen production decisions.

Forecasting Is Not Just About Revenue; It Supports Operational Decisions

A sales forecast is often seen as an estimate of revenue over a given period. For restaurants, however, its value goes well beyond revenue projections.

A useful forecast should answer practical questions: How many portions are likely to sell? Which ingredients need to be prepared? And when should production begin?

For example, a daily sales target of IDR 15 million may help track business performance, but it does not tell the kitchen how much food to prepare.

The team still needs to estimate the number of portions of fried rice, grilled chicken, and beverages to prepare. That is why restaurant sales forecasts should be broken down by product or menu category.

Read also: OLIN by ESB: An AI Application for Culinary Businesses That Can Increase Efficiency and Sales

Choose the Right Level of Detail: Day, Outlet, and Menu

Source: iStockphoto

Before creating a forecast, decide how detailed the data needs to be. Avoid using one number for the entire business when outlets and menu items have different demand patterns.

For a single-outlet restaurant, daily forecasts by menu category may be a reasonable starting point. Multi-outlet businesses should also separate forecasts by location.

For instance, an outlet in a business district may be busiest during weekday lunches, while one in a shopping mall may attract more customers on weekends.

Menu-level detail also matters. Main dishes, beverages, desserts, and seasonal items have different demand patterns. Combining them can make it harder to translate forecasts into ingredient requirements.

Build a Baseline from Historical Sales

Once you have chosen the level of detail, create a baseline using historical sales data.

A simple approach is the moving average, which calculates average sales over recent periods. For example, grilled chicken sales over three days were:

The average is:

(80 + 90 + 100) ÷ 3 = 90 portions.

The initial forecast for the following day is therefore 90 portions. This is a starting point, not a final production target.

Sales are influenced by many factors. Demand on Thursday may differ from demand on weekends.

A menu promotion or an unexpected increase in customers may also affect sales.

Use the average as a baseline, then adjust it to reflect operational conditions.

Adjust for Weekdays, Weekends, Seasonality, Promotions, and Events

A simple average may not reflect actual demand. Restaurants should examine sales patterns across different days and circumstances.

Differences between weekdays and weekends are often easy to identify. If weekend sales are consistently higher, use a separate forecast for Saturdays and Sundays.

Seasonality also matters. Restaurants in tourist areas may experience a surge in customers during holidays, while outlets near offices may see demand shift when more employees work from home.

Promotions and local events can affect demand as well. Discounts, new menu launches, concerts, exhibitions, and nearby activities may increase orders.

However, do not automatically raise the forecast whenever a promotion is scheduled. Review results from similar campaigns, if available. For example, a previous lunch promotion may have increased rice bowl sales by 20%.

That result can inform the estimate, but it should be adjusted for the conditions of the new campaign.

Turn Sales Forecasts into Ingredient Needs and Production Plans

This is where forecasting becomes useful for daily operations. Expected sales must be translated into production quantities and ingredient requirements.

Suppose the restaurant expects to sell 120 portions of grilled chicken tomorrow. Under the standard recipe, each portion requires 150 grams of chicken. The required amount is:

120 portions × 150 grams = 18,000 grams, or 18 kilograms.

With an additional 5% operational buffer, the total requirement becomes 18.9 kilograms. The calculation is shown below:

Next, check the stock on hand. If 5 kilograms of chicken remain in storage, the restaurant needs approximately 13.9 kilograms more to meet this requirement.

Repeat the calculation for other ingredients, such as rice, vegetables, seasonings, and packaging. Purchasing can then plan orders while the kitchen schedules preparation and production.

The buffer should reflect the characteristics and risks of each ingredient. Perishable products require more careful planning than items with a longer shelf life.

Read also: The Role of Artificial Intelligence in Business: 15 AI Applications You Should Know

Measure Forecast Error by Comparing Predictions with Actual Sales

Source: iStockphoto

Actual sales will not always match the forecast. Restaurants therefore need to measure the difference between predicted and actual results.

For example, a restaurant forecasts 120 portions of grilled chicken but sells only 108. The forecast was 12 portions higher than actual sales.

To express the absolute error as a percentage of actual sales, use:

Forecast Error (%) = |Forecast − Actual| ÷ Actual × 100%

Do not stop at the size of the error. Investigate its causes so the next forecast can be improved.

When Should a Forecast Be Updated?

A forecast should not be created once and left unchanged. Business conditions can shift within days or even hours.

Update forecasts when demand changes significantly—for example, when group reservations increase, promotions outperform expectations, suppliers are delayed, or a menu item suddenly becomes popular.

For daily operations, review the forecast before purchasing and production begin. In multi-outlet businesses, managers should also adjust estimates for each location.

If actual sales begin to diverge substantially from the forecast, do not wait until month-end. Review the latest data and adjust stock needs and production schedules.

Do Not Use Last Year’s Sales Data Without Adjustments

Historical data from previous years can reveal seasonal patterns, but applying it unchanged may produce inaccurate estimates.

Suppose the restaurant sold 5,000 portions last December. Since then, it has opened two new branches, expanded its menu, and changed its operating hours. These changes can alter sales patterns.

The same applies to price changes, outlet renovations, or adjustments to kitchen capacity. Review older data before using it in a new forecast.

Use historical sales as a reference, then adjust for current business conditions. This keeps the restaurant sales forecast relevant to inventory and production planning.

Manual Forecasting vs. Data-Driven and AI Forecasting

At first, a spreadsheet may be sufficient for forecasting. Teams can enter sales data, calculate averages, and adjust estimates for different days or campaigns.

However, manual forecasting becomes more demanding as the number of outlets and menu items grows.

Data from each branch must be collected, checked, and consolidated, which takes time and increases the risk of recording errors.

A data-driven approach uses structured transaction records to identify sales patterns more consistently.

AI forecasting can help analyze more complex relationships, including demand trends, seasonality, and changes in customer behavior.

Even so, AI does not guarantee accurate forecasts. Data quality, consistent records, and an understanding of actual outlet conditions remain essential.

For multi-outlet businesses, a system that can analyze sales history across locations may provide more practical decision support.

Managers can use those insights as a starting point and adjust decisions using the latest information from each outlet.

Restaurant Sales Forecast FAQs

1. How Often Should a Restaurant Update Its Sales Forecast?

For inventory and production planning, forecasts can be updated daily or in line with the purchasing cycle. Weekly reviews also help identify demand patterns.

2. Do Small Restaurants Need Sales Forecasts?

Yes. Small restaurants can use simple methods, such as averaging sales over recent days, to estimate ingredient requirements.

3. What Is the Difference Between a Sales Forecast and a Sales Target?

A sales target is the amount a business aims to achieve, while a sales forecast estimates likely sales based on available data and current conditions.

4. What If Actual Sales Differ Significantly from the Forecast?

Check for changes in demand, promotions, events, stockouts, or operational disruptions. Then update the forecast and revise the production plan.

Conclusion

A useful sales forecast does more than estimate revenue. It helps restaurants plan ingredient purchases, set production schedules, and reduce the risk of both excess inventory and stockouts.

Start with available sales records. Separate the data by outlet and menu, establish a simple baseline, then adjust for weekdays, seasonal patterns, promotions, and events.

Compare each forecast with actual sales to improve future estimates. As the business grows, restaurant sales forecasting increasingly depends on connected, accessible data.

Managing separate spreadsheets can make it difficult to review sales history across multiple branches.

To support data-driven decision-making, explore OLIN by ESB. This solution can help restaurants turn operational data into more actionable insights.