The whole month, from the export to the brand’s report.
MenuMetrics reads the settlement files you already download, rebuilds every payout from the file itself, and turns what does not tie into a claim you can send. Here is all of it.
The exports you already download
No partner-portal login, no integration to wait for. You download the file as you do today and drop it in; the software works out the rest.
Both platforms, every generation of their exports
Swiggy's weekly settlement, its monthly annexure and its consolidated CSV. Zomato's weekly and monthly settlement reports. CSV or XLSX, whichever the portal gave you.
Dine-in as well as delivery
Zomato Dining's payout report settles table bookings and walk-ins on a statement that shares no column with the delivery one. It is read on its own terms, and the booking channel is kept so a prebooked table is never counted as a delivery.
The platforms' own business-metrics reports
Daily, weekly or monthly. A portfolio export covering forty restaurants is split to every branch it names, and any restaurant it cannot place is asked about by name rather than filed somewhere approximate.
Files place themselves
Every export carries the platform's own restaurant ID. The upload reads it, matches it to the branch that trades under it, and files itself — brand, outlet and platform, without being asked.
Columns matched automatically
Each platform's column names are written down and checked against real exports, and the file's own glossary settles the rest. Each file's columns are matched and checked against the payout arithmetic before anything is imported — there is nothing to map.
A re-upload replaces, it does not duplicate
Overlaps are judged on the orders in the file, not only on its dates, so a week that brushes last month's is not mistaken for a repeat. The original file is archived and can be downloaded back from the imports list.
Every payout, rebuilt from the file
The arithmetic you would do by hand for one order, done for every order — and shown, so a claim goes out with its working rather than with an assertion.
The payout equation, order by order
What the customer paid, less the platform's fees and charges, against what actually settled. The share of orders that tie is stated before anything is imported, so a mapping that is wrong is caught before it reaches the ledger.
Commission checked two separate ways
Whether the platform computed its own stated rate correctly — provable from the file alone — and whether that rate is the one you negotiated. They are different claims with different strengths, and they are never merged into one number.
Cancellations handled per platform
Swiggy keeps crediting a cancelled basket and recovers it; Zomato drops it and settles a refund. The convention is read off the file rather than assumed, and an order the file's own arithmetic cannot explain is reported as unchecked instead of claimed.
Deductions taken outside the order rows
Advertising, penalties and adjustments billed against the statement rather than against an order are counted too — including a settlement week with no orders that still carried a charge.
Lines that ran above their own normal
Measured as a share of gross against what this branch actually does, not against a fixed threshold — because eighteen percent is normal on one brand and a breach on the next.
Where every figure came from
A reference page, generated from the same declarations the importer runs on, saying which column of which export became which number. It cannot go stale, because it is not written by hand.
From a finding to money back
Findings are raw material, not claims. You decide what is worth raising; the software carries it through and remembers where it got to.
Grouped the way a claim is filed
Findings arrive sorted by the portal the claim goes into and by what they are worth, so one pass through a page is one claim.
A claim you can paste and send
Copy out the orders, the amounts and the arithmetic behind each one, ready for the partner-portal ticket.
Status through to recovered
Raised, in progress, won, written off. Recovery lands on the brand's report as a figure you can point at.
Nothing chased from memory
Tasks per brand and per branch, with an assignee and a due date, on the portfolio and on the brand's own page.
The month, and what moved in it
A monthly total is where a problem hides. Every view here exists because someone was building it in a spreadsheet first.
Gross to money received
The full bridge: what customers paid, commission, fees, funded discounts, advertising, taxes — and what reached the bank. One specification, so the brand's page, a single import and the report cannot disagree.
Trends, one row per period
The same twelve lines against day, week or month. Sortable on any column, filterable by order type, and downloadable as a workbook that is the table you were looking at rather than a second gathering of it.
The real cost, not the headline rate
Commission, payment fees, funded discounts and ad spend as one all-in rate per branch — with the spread behind it, because one averaged rate can hide orders that sit well above it.
Breakfast is not dinner
The day split into breakfast, lunch, snacks, dinner and the small hours. A branch whose lunch is thriving while its dinner bleeds looks perfectly healthy in a monthly total.
Order size and delivery distance
Counted in brackets rather than averaged. Forty orders at ₹120 and twenty at ₹500 are two businesses wearing one number, and the fix for each half is different.
What the settlement cannot see
The platforms' own measurements alongside: how many people saw the listing, opened the menu, built a cart and ordered — and how many had ordered before. A settlement only ever describes orders that happened.
Swiggy against Zomato, line by line
The business metrics both apps publish, named once and set side by side with the gap between them — sales, cancellations, ratings, prep time, the funnel, new and repeat customers, ads and offers. Where one app gives a percentage and the other a count, both are worked out.
What the kitchen sold, dish by dish
Both apps' itemwise reports on one line per dish, matched across menus that spell it differently — “Biryani” and “Biriyani” are one dish, counted once. Sortable, searchable and downloadable.
Which branches have gone quiet
A listing that arrived every week for three months and then stopped is the one you will forget. The software notices, without you having to set a schedule.
The month you could have had
The profit sheet you would otherwise keep in Excel, worked from the brand's own month — and the exact change that would have broken it even.
A profit calculator on the real month
The month in detail from the brand's report, worked through the consultants' profit sheet: credited amount, unaccounted amount, dine-in price, menu price recovery, break-even and net profit — the same formulas, cell for cell.
Type the percentage, see the profit
Change the discount, advertising, delivery, refunds, penalties, packing, menu markup or food cost and everything below recomputes, beside the month as it was. The platform fee can be tried too, with a warning that it is the contract and hard to lower.
What would break it even
Each change that closes the gap on its own, worked out exactly and applied with one click — and which costs could not close it even at zero.
Money the platforms did not pay
Where the payout came to less than the month's entries add up to, the shortfall is named in rupees and pointed at Disputes.
The month, on paper
The output of the month, in your name and your colours — a document your team or a brand's owner can read without a login.
In your own branding
Your logo and your colours. Send it to an owner white-labeled — nothing on it sells this software.
Written in the owner's voice
The same bridge the person running the numbers reads as “borne by outlet” reads as “borne by you” for the owner. One specification, two voices — so the two can never drift apart.
Read it before you keep it
Preview the whole report, re-cut the range or the scope without leaving the page, and keep it only when it says what you meant.
Prints properly
Charts and all, with panels moved whole to the next sheet rather than sliced across a page break. The branch's registration details are on it where a brand's accountant expects them.
Profit, where the costs are agreed
With each outlet's menu markup and food cost set, the report states menu price, menu price recovery, food and packing, and profit after advertising.
Counts what you tell it to
Choose which order types a report includes, so a dine-in month is not reported as a delivery story — or a train-order channel an owner does not count as their own is left out.
Twenty brands without twenty spreadsheets
The part that is not analysis: your brands, your team, and the setup you only do once.
Brands, branches, platforms
Each brand's outlets, the restaurant ID each one trades under on each platform, the commission agreed there, and the markup the profit line is built on.
Every branch on one page
Every outlet at once, searchable, sorted by brand — and handed over as a workbook that is the list you were looking at, search and all.
Your team, with seats
Invite colleagues, set who is an admin, assign tasks to a person. A workspace can never be left with no admin at all.
One portfolio view
Every brand ranked by what it turns over, what its platform fees came to, what is outstanding and whose report has gone out.
Check this against your partner portal
Every one of these is read from a real export. A platform that is not listed still works, as long as its file has one row per order — the columns are matched by name.
Swiggy
- Weekly settlement report
- Monthly annexure
- Consolidated order CSV
- Business metrics
- Order report with items
Zomato
- Weekly settlement report
- Monthly settlement report
- Dining payout report
- Business metrics
- Itemwise sales report
Magicpin · ONDC · others
- Any order-level export
- Mapped by its own column names
- CSV or XLSX
Start with one brand and last month’s settlement file.
One branch, one upload — and you will see what the platform actually charged.