Restaurant Analytics and Performance Dashboards: What to Track
Restaurant operators who track the right metrics make better decisions. Here is a guide to the key analytics and KPIs for restaurant performance management.
Data-driven restaurant management requires tracking the right metrics — not just revenue and cover count, but the leading indicators that explain performance and inform decisions.
Revenue and Volume Metrics
- Revenue per day/week/month: The baseline. Trend more important than absolute.
- Cover count: Number of diners served. With revenue, gives revenue per cover.
- Average transaction value: Drives revenue leverage beyond footfall.
- Revenue per labour hour: Efficiency metric particularly important for labour-heavy operations.
Online Ordering Metrics
For restaurants with online ordering: orders by channel (platform vs direct), average order value by channel, delivery completion rate, order-to-kitchen time, kitchen-to-delivery time.
Food Cost and Margin Metrics
- Food cost percentage: (Food cost / Revenue) × 100. Target typically 28–35%.
- Menu item profitability: Revenue vs food cost by item — identifies high-margin stars and margin drags.
- Waste percentage: Waste cost as a percentage of food purchases.
Customer Metrics
- Review scores by platform: Google, TripAdvisor, delivery platforms — tracked weekly.
- Repeat customer rate: Percentage of online orders from returning customers.
- Allergen incident rate: Number of allergen-related customer contacts per period.
Manage online ordering and dietary filtering for your restaurant on Shopify with Smart Shop at saltai.app.
Staff and Labour Analytics
Labour is typically the largest controllable cost in a restaurant operation, often sitting between 28% and 35% of revenue. Tracking labour cost as a percentage of revenue — broken down by day part, day of week, and service type — gives operators a clear picture of where staffing decisions are adding or destroying margin. A Monday lunch service running at 45% labour cost tells a fundamentally different story from a Saturday dinner at 22%, and your dashboard should surface that contrast automatically.
Scheduling accuracy is a metric worth building into any performance dashboard. Compare scheduled hours against actual hours worked each week, and track the variance over time. A consistent pattern of over-scheduling on quiet shifts or under-staffing on busy ones points to a forecasting problem that costs money in both directions — unnecessary wage spend on slow days, and compromised service quality on high-demand ones.
Beyond cost, staff productivity metrics matter. Revenue per labour hour, tracked by role or by shift, helps managers understand which team configurations deliver the best commercial outcomes. Pairing this with customer satisfaction scores from the same period begins to answer the more nuanced question: not just how cheaply can we staff a shift, but how do staffing decisions affect the customer experience and repeat trade?
Table Turnover and Capacity Utilisation
For dine-in operations, table turnover rate is one of the most important volume metrics available. It measures how many times each table is seated during a service, and improving it — even marginally — can significantly increase revenue without any change to footfall. A restaurant turning tables 2.1 times per service versus 1.8 times is extracting meaningfully more revenue from the same physical space and the same number of walk-ins.
Capacity utilisation sits alongside turnover as a paired metric. Tracking the percentage of available covers filled across each service, by day of week and time of day, reveals demand patterns that should directly inform your marketing, promotional calendar, and staffing plan. A restaurant consistently running at 40% capacity on Tuesday evenings has a different problem to solve than one that is full Thursday through Saturday but unable to build midweek trade.
Time-to-table and dwell time are operational metrics that feed into both turnover and guest experience. If average dwell time is climbing, the cause could be a positive one — guests enjoying themselves — or a negative one — slow service or kitchen delays. Cross-referencing dwell time with review sentiment and kitchen ticket times helps distinguish between the two. The goal is not to rush guests out, but to ensure that pacing is deliberate and operationally controlled rather than accidental.
Menu Performance and Engineering Metrics
Menu engineering is a well-established discipline, but it requires consistent data to be actionable. The core framework plots each menu item on two axes: popularity (order volume) and profitability (contribution margin). Items that are both popular and profitable are stars — protect them. Items that are popular but low-margin are plowhorses — candidates for ingredient substitution or price adjustment. High-margin but rarely ordered items are puzzles — they may need repositioning, better menu placement, or stronger staff recommendation.
Tracking attachment rates adds another dimension to menu performance analysis. An attachment rate measures how frequently a given item is ordered alongside another — a particular starter with a main, a dessert after a specific dish, a drink upgrade during online checkout. High attachment rates reveal natural pairings that can inform upselling scripts for front-of-house staff and recommendation logic for online ordering platforms. Low attachment rates on high-margin add-ons represent untapped revenue.
Seasonal and time-based menu performance data is often underused. Tracking which items over- or under-perform during specific periods — holiday menus, summer vs winter, lunch vs dinner — enables smarter menu planning ahead of each season rather than reactive adjustments after margin has already been lost. A dashboard that surfaces this data year-on-year gives operators genuine institutional memory, rather than relying on informal recollection of what sold well last Christmas.
Building a Useful Dashboard: Principles and Pitfalls
The most common mistake in building a restaurant analytics dashboard is tracking too many metrics without a clear decision-making framework behind them. A dashboard that displays forty numbers simultaneously creates cognitive load without clarity. The better approach is to organise metrics by decision type: operational decisions (daily), tactical decisions (weekly), and strategic decisions (monthly or quarterly). Each layer should surface only the metrics relevant to the decisions made at that frequency.
Leading indicators deserve more dashboard real estate than lagging ones. Revenue last month is a lagging indicator — it tells you what happened. Online order conversion rate, repeat customer rate, and average preparation time are leading indicators — they tell you what is likely to happen. Building your dashboard to foreground leading metrics means problems are visible while they can still be acted on, rather than after they have already affected the bottom line.
Integration between your point-of-sale system, online ordering platform, and inventory management tools is what determines whether your dashboard is live intelligence or a manual reporting exercise. Restaurants running Shopify for online ordering have a structural advantage here: Shopify's data infrastructure connects naturally with analytics tools, and apps built specifically for hospitality use cases can surface the restaurant-specific metrics — food cost percentage, allergen incident tracking, channel-by-channel order performance — that generic e-commerce dashboards do not provide. If your current setup requires manual data exports to produce a weekly performance report, that is the first operational problem worth solving.
Benchmarking and External Context
Internal trend data tells you whether performance is improving or declining, but benchmarking against external data tells you whether your improvement is keeping pace with the market. Industry benchmarks for food cost percentage, labour cost percentage, and average transaction value vary by cuisine type, service model, and geography — a fast-casual operation in London has a different cost structure to a full-service restaurant in a smaller market. Using the right peer group for comparison matters.
Review platform scores are a useful proxy for relative standing within your local competitive set. Tracking your average score on Google and TripAdvisor alongside your position relative to comparable local operators gives context that an absolute score alone cannot provide. A 4.2 average sounds reasonable in isolation; it looks different if every comparable competitor in your area is sitting at 4.5 or above.
Economic and demand context should also inform how you interpret your own performance data. Revenue declining during a cost-of-living squeeze is a different problem from revenue declining in a buoyant consumer environment. Keeping a simple external context log alongside your internal data — noting significant local events, competitor openings or closures, and macroeconomic conditions — makes historical performance data far more useful when you return to it for planning purposes six or twelve months later. Context is what turns raw data into genuine business intelligence.
Try SaltAI Agent for Shopify free at saltai.app — no credit card required.
SaltAI Team
SaltAI builds focused Shopify apps for food merchants and general merchants. Every app is tested in production at a real food store — including Vanda's Kitchen — before it ships.