Optimized F&B menu for a luxury hospitality chain to maximize revenue

A luxury hospitality chain that manages a portfolio of hotels, resorts, jungle safaris, palaces, spas, and in-flight catering services wanted to optimize its food and beverages (F&B) menu to increase sales and improve guest loyalty.
TransOrg analyzed guests’ transactions on F&B by using advanced analytics techniques and machine learning models to:

• Find opportunities of bundling, cross-selling and up-selling menu items as combo deals and to prepare a recommendation guideline for the servers to suggest items to customers while dining.
• Find insights in revenue and quantity trends and suggest actionable strategies for increasing revenues, reducing costs and optimizing menu pricing.
• Identify opportunities of restructuring menu layout with reference to transactions.
• Identify opportunities to create offers based on the time of transaction and seasonality.

The project commenced with a premium hotel property situated in one metro city and eventually scaled up the solution to other hotel properties across India.

TransOrg analyzed the Point of Sales (POS) data for a period of one year comprising of approximately 30,000 transactions and 90,000 ordered items.

Transactional analysis was done on multiple parameters such as:

• Frequency of item ordered vs. item name
• Revenue of item ordered vs. item name
• Overall revenue vs. DOW, TOM, and season*
• Revenue change per item vs. DOW, TOM, and season.
• Average, Overall and Daily revenue of top items vs. DOW, TOM, and season.

**DOW: Day of the week (weekdays: Mon-Thu and weekend: Fri-Sun); TOM: Time of the meal (Breakfast, Lunch and Dinner); Season: Summer, winter, spring, autumn and monsoon.

Market Basket Analysis:

TransOrg used the Apriori association method to find frequent item sets and derived association rules to uncover meaningful correlations between different products according to their co-occurrence in a data set. The following measures were used to evaluate the strength of association:

• Support for the rule indicates its outcome in terms of overall size
• Confidence determines the operational usefulness of a rule
• Lift ratio indicates how efficient is the rule in finding consequences, compared to random selection of transaction

Key Impacts

• Increased revenues through combos and in-demand items
• Targeted campaigns to up-sell and cross-sell
• In-depth understanding of top revenue generating items across seasons and days

Key Insights

• Dishes with regional names and restaurant specific names have contributed significantly to revenue.
• Client earns highest revenues during summer.
• Buffet breakfast is the revenue driver for weekdays in all seasons except the monsoon.
• Product R dropped out of Top fifteen revenue contributors in Autumn and Winter.
• In Autumn, seasonal products P and Q that are added to menu for 15 days contributed approximately USD 7,000.
• Spikes in order frequency align with suggestions by food aggregator apps such as Swiggy or Zomato.
• Demand for products X, Y and Z surge in summer.
• Combo X has negligible demand in monsoon.