Menu Engineering with Data Analytics for Indian Restaurants

Why Menu Engineering Is More Important in India Than Operators Realize

A typical Indian restaurant menu — especially in the casual dining and QSR segments — carries between 40 and 120 active SKUs across multiple categories. The intuitive assumption is that more menu options means more customer choice and therefore more revenue. The data-driven reality is almost always the opposite: most menus have 15-25% of dishes that generate 60-70% of revenue, while a significant tail of low-volume items creates operational complexity, increases food waste, ties up inventory capital, and slows kitchen throughput.

Menu engineering is the discipline of using sales data, cost data, and customer preference data to make deliberate decisions about which items to promote, which to reprice, which to remove, and which to develop further. In India, where food cost inflation, aggregator commissions, and thin operating margins make profitability per dish more critical than ever, menu engineering is not a luxury — it is a fundamental management practice.

The Menu Engineering Matrix: Stars, Plowhorses, Puzzles, and Dogs

The foundational framework for data-driven menu analysis was developed by Michael Kasavana and Donald Smith in 1982, and it remains the most practically useful tool for Indian restaurant operators today. The matrix classifies every dish on two dimensions: popularity (relative sales volume compared to the menu average) and profitability (contribution margin compared to the menu average).

Stars — High Popularity, High Margin

Stars are your best-performing dishes. They sell frequently and they make money when they do. In an Indian restaurant context, a perfectly positioned biryani, a signature curry, or a bestselling dessert that uses low-cost ingredients and sells at a premium price would qualify. Stars should be featured prominently in menus, both physical and digital (on Swiggy and Zomato, they should appear in recommended sections and top-of-category positions). The strategic goal is to protect and grow Stars — ensure consistent quality, maintain prominent placement, and never discount them unless strategically justified.

Plowhorses — High Popularity, Low Margin

Plowhorses sell well but do not make much money per unit. For Indian restaurants, this is often the case for staple dishes like dal makhani, plain naan, or jeera rice — items customers order reliably but that are priced low (often because they are perceived as commodity items) and/or have high ingredient costs. Plowhorses are the most actionable category: slight price increases (Rs. 20-30 on a Rs. 180 dish rarely reduces demand meaningfully), ingredient cost reductions through supplier renegotiation, or portion optimization can shift Plowhorses toward Star territory without losing their volume advantage.

Puzzles — Low Popularity, High Margin

Puzzles are dishes that make excellent margins when ordered but are not ordered often enough. This may be because they are not visible enough in the menu, not well-described, or simply not known to customers. For Indian restaurants on Swiggy and Zomato, Puzzles are candidates for featured placement, better photography, or inclusion in combo offers that expose them to customers who might not discover them independently. A tandoori specialty dish with a 65% gross margin that only 3% of customers order could become a significant revenue contributor if its ordering rate doubles through better merchandising.

Dogs — Low Popularity, Low Margin

Dogs consume menu space, kitchen prep effort, and inventory without meaningful contribution to revenue or profit. The appropriate action for most Dogs is removal from the menu. However, this decision requires judgment in the Indian context — some items have cultural significance (a restaurant might keep a regional specialty on the menu for identity reasons even if it sells rarely), and some apparent Dogs are seasonal items misclassified due to limited data windows. Remove Dogs systematically, but investigate before cutting any item with potential strategic value.

Calculating Contribution Margin Per Dish

The profitability axis of the menu engineering matrix requires calculating contribution margin (CM) for each dish, not just gross margin. Contribution margin is defined as: selling price minus variable food cost minus variable packaging cost (for delivery items). It is distinct from the gross margin shown on financial reports because it isolates the variable cost per dish rather than allocating fixed overhead.

The formula: Contribution Margin = Menu Price − (Raw Material Cost per Serving + Packaging Cost per Delivery Order)

To calculate this accurately, you need a recipe costing database that links each menu item to its ingredients and standard portion sizes, combined with current ingredient purchase costs. In Indian restaurant chains, this data typically lives in the inventory management system (if one is used) or in a chef's cost card spreadsheet. Pulling this into your analytics system alongside sales volume data from the POS enables the full Stars/Plowhorses/Puzzles/Dogs classification at scale.

Regional Menu Performance Differences in India

One of the most practically important dimensions of menu analytics for Indian restaurant chains with multi-city presence is that the same dish performs very differently across regions. A North Indian chain expanding to South India will quickly discover that certain staple items that are high-volume Stars in Delhi and Indore become low-volume Dogs in Chennai or Bengaluru — not because the dishes are poor, but because local palate preferences, familiarity, and competing local cuisine options create structurally different demand patterns.

A paneer butter masala at a North Indian chain might sell 80 units per day in a Jaipur outlet and 22 units per day in a Hyderabad outlet at the same price. The Hyderabad outlet's Top-5 selling items may have completely different composition from the Jaipur outlet's Top-5. Managing menu strategy with a single national lens — applying the same featured items and promotional priorities chain-wide — systematically underperforms compared to a regional menu strategy informed by outlet-level sales data.

For chains above 15 outlets, building regional menu performance reports that cluster outlets by city or market segment and rank items by contribution margin contribution within each cluster is a high-ROI analytics investment. This enables regional operations managers to make menu decisions appropriate to their markets rather than following a national mandate that was calibrated for a different customer base.

A/B Testing Menu Changes on Swiggy and Zomato

Delivery platforms offer a significant advantage for menu experimentation that dine-in restaurants do not have: the ability to test menu changes on a subset of customers and measure the impact before committing to a chain-wide change. Both Swiggy and Zomato allow restaurants to update menus at the outlet level — meaning a chain can run different menu configurations at different outlets and measure comparative performance.

A practical A/B test structure for Indian restaurant chains: select two similar outlets (similar market size, similar customer demographics, similar baseline revenue) and implement a menu change at one outlet while keeping the other unchanged. Run for 4-6 weeks minimum to account for day-of-week and weekly demand variation, then compare revenue per order, item-level sales volumes, and AOV between the test and control outlets.

Menu changes worth testing include: price increases on specific Plowhorse items (does a Rs. 30 price increase on dal makhani reduce order volume by more or less than the margin gain?), featuring Puzzle items in the recommended section (does improved placement increase their order rate?), and category ordering changes (does moving desserts higher in the menu sequence increase dessert attachment rates?).

What to Measure in a Menu A/B Test

• Item-level order volume for the changed items — did the intervention increase or decrease ordering?
• Category-level order volume — did changes in one category affect adjacent categories?
• Overall AOV — did the change increase or decrease average order value?
• Customer rating trends — did quality-sensitive changes (portion size, recipe adjustment) affect post-order ratings?
• Revenue per outlet day — the ultimate measure of whether the change was net positive

Seasonal Menu Optimization

Indian restaurants have strong seasonal demand patterns that are often managed by instinct rather than data. Mango-based dishes spike in demand from April to June and then disappear from customers' consideration entirely. Soup and warm beverage items see significant demand spikes October through February in North and Central India. Wedding and festive season (October-November) drives large-format and party menu demand. Monsoon season in Mumbai and Pune creates a predictable demand shift toward comfort foods.

Analytics enables proactive rather than reactive seasonal menu management. By looking at prior years' item-level demand by month and correlating with seasonal or calendar events, chains can predict which items to feature, which to remove, and which new items to introduce during each season — rather than discovering demand shifts from customer complaints or inventory pile-ups. Building a seasonality calendar informed by two or more years of sales data is one of the highest-value menu analytics exercises an Indian restaurant chain can conduct.

How Restrologic Powers Menu Engineering for Indian Chains

Restrologic's restaurant analytics platform includes a dedicated menu performance module that automates the Stars/Plowhorses/Puzzles/Dogs classification for your entire menu, updated weekly as sales data flows in from your POS and aggregator platforms. Our contribution margin calculations integrate with your recipe costing data to provide true margin-based classification rather than revenue-volume proxies. For multi-city chains, we provide regional menu performance breakdowns and outlet-cluster analysis that reveals how the same items perform differently across your estate. Menu engineering decisions that once required a week of manual analysis can be made in a daily menu performance dashboard review — with the confidence that the data is current, complete, and structured for action.