Restaurant analytics dashboard setup in India

Many Indian restaurant chains hold plenty of data and still decide by gut feel. This guide walks through the steps: define the right KPIs, choose a BI tool, connect your data sources and build a dashboard that operators will open every day.

Why do restaurant analytics projects fail early?

The conversation often starts in the wrong place. A chain sees a competitor talking about data-driven decisions, hires an analyst, and three months later has a handsome dashboard nobody opens. The reports are right but irrelevant. They show what the analyst found interesting, not what an outlet manager needs at 11 AM on a Tuesday when deciding whether to run a lunch promotion.

The project starts from the tool instead of the decision. Before you choose a BI platform or connect any data, agree which decisions you want to make with data and how often. Every KPI on the dashboard should match a decision someone makes regularly.

Step one: how do you define your KPI framework?

Track metrics at three levels: outlet operations, chain business and channel performance for aggregators and direct ordering. Mixing them in one dashboard gives reports that overwhelm and do not help.

Which outlet-level KPIs matter?

• revenue per outlet per day, split by dine-in, takeaway and delivery
• average order value (AOV), tracked per channel, since Swiggy and dine-in AOV differ in structure and should not be blended
• order volume by hour, which drives staffing and prep
• rejection rate for aggregator orders, which affects Swiggy and Zomato visibility
• table turn time for dine-in outlets
• food cost percentage against theoretical food cost, weekly

Which chain-level KPIs matter?

• EBITDA per outlet, which needs accounting data
• revenue channel mix across dine-in, Swiggy, Zomato, direct ordering and ONDC
• repeat order rate within 30 and 90 days
• outlets ranked by revenue and profit, which shows underperformers and stars quickly
• ramp rate of new outlets towards target revenue, a sign of expansion quality

Chains that settle the KPI framework before choosing a BI tool tend to get better adoption among their operations teams.

Step two: how do you choose a BI tool?

The right tool depends on your team’s technical skill, budget and the complexity of your questions.

Is Metabase a good fit?

Metabase is open source, can be self-hosted to control cost, and has an interface non-technical operators can use. Complex joins need SQL knowledge. It is a good, low-cost option if you have an analyst who can build and maintain the data models.

What about Power BI?

Power BI is widely used in Indian enterprises and works well with Excel and Microsoft 365. It is strong on rich visuals. It is Windows-centric, and editing in the browser is less capable than on the desktop.

Where does Grafana fit?

Grafana is a time-series and monitoring tool. It suits live operations dashboards for kitchen throughput, order queues and aggregator performance at peak. It is not a general BI tool and is a poor fit for financial reporting or complex segmentation. Consider it alongside a traditional BI tool.

What about Looker Studio?

Looker Studio is free and works natively with BigQuery, which suits chains on Google Cloud. It is less powerful than Power BI for complex modelling but easier to learn. The enterprise Looker product costs much more and mostly suits very large chains with a data engineering team.

Step three: which data sources should you connect?

A dashboard is only as good as its data.

Why start with the POS?

Transaction data from your POS is the base. It needs to reach the analytics layer reliably and on a set schedule. Our POS integration guide explains the technical side of connecting these systems.

What about Swiggy and Zomato data?

Both offer reporting access for partners, though the data may be less detailed than what you see in their partner portals. You can pull order-level data, ratings and performance measures into your analytics layer. The platforms change their interfaces from time to time, so connectors need upkeep. Scheduled report downloads can serve as a fallback for history.

Why connect accounting?

Without accounting data, a dashboard shows revenue but not whether you made money. Connecting Tally or Zoho Books enables food cost tracking, rent and overhead allocation and true EBITDA per outlet.

How does inventory help?

Inventory data joined with POS sales allows theoretical against actual food cost reconciliation, one of the most useful ways to find waste, theft and recipe non-compliance.

Step four: what should the first dashboard look like?

Build a daily operations summary that a general manager or cluster manager opens each morning. It should answer four questions without filtering. How did each outlet do yesterday? Which outlets are on track for weekly targets? Which had operational issues, such as high rejection, unusual voids or low order volume? How does this week compare with last week?

Do not build a 20-tab suite first. A one-page daily digest that eight people open every morning is worth more than a thorough report nobody reads. Add depth for questions people actually ask.

Which mistakes are common?

• Mixing revenue and order count without context. An outlet with many low-value orders and one with fewer high-value orders look very different in an order-count chart. Show revenue next to volume.
• Ignoring day-of-week patterns. Compare same weekdays. This Tuesday against last Monday means nothing.
• Treating aggregator gross value as restaurant revenue. Commissions mean only part of each order value reaches you. Dashboards that skip this overstate delivery profit.
• Ignoring data freshness. A dashboard that shows yesterday’s data labelled “today” destroys trust. Show the last-updated time.
• Building for the analyst, not the operator. Managers in India typically open dashboards on a phone during service. Mobile-friendly views are essential.

Step five: how do you drive adoption?

Even a technically strong setup fails if operators do not use it. Get buy-in from operations leaders, train people in practical terms and show early wins that the dashboard gives useful insight and not just what managers already know.

Find one or two power users, such as a COO, head of operations or a data-minded cluster manager, and build the first version around their workflow. When they quote dashboard numbers in review meetings, adoption spreads through the team.

How does Restrologic help?

Our restaurant analytics and BI service covers the full stack: connecting POS systems and aggregator feeds, building the data layer and delivering dashboards set up for your KPI framework. We start with a core operations dashboard and a channel performance view. It works with Restrologic POS and with other systems. To plan yours, contact us.