Retail analytics software for store chains.
Sales, margin and stock for every store, compared on one page by morning, with out-of-stocks and overstock flagged store by store. It works with the POS and ERP your stores already run and changes nothing inside them.
Network · demo chain
data as of 08:40 · 4 of 5 locations reporting
| location | revenue, yesterday | margin | vs its own normal |
|---|---|---|---|
| City Center | 4 820 € | 27.8% | +6% |
| Westside | 3 140 € | 24.1% | −11% |
| North Park | 1 960 € | 26.0% | −2% |
| Harbour | 2 470 € | 25.4% | +1% |
| Old Town | no data since 06:10POS offline — excluded, not counted as zero | ||
| networkfour locations | 12 390 € | 26.1% | |
network margin is revenue-weighted, never an average of percentages
Your stores keep their systems. The chain gets one.
Every location keeps the system it has. What marql cannot read, it says out loud instead of filling in.
Cloud POS, and the ERP behind the firewall
Cloud POS, accounting, the web shop and the marketplaces connect over their own APIs. A 1C or ecorg ERP that lives inside your network is read by a small on-prem agent instead. marql only reads your POS and accounting. With e:corg 2.0, e:corg creates an approved order or transfer as its own draft.
The chain becomes comparable
Two spellings of one product become one SKU, categories and VAT are mapped, and money is converted to one currency — so another country is a row in the same table rather than a separate report.
A location that sent nothing is named as not reporting, not shown as a complete total
Forty alerts become one decision on Today
Anomalies, margin drift and the problems that have been repeating for weeks are ranked by the money behind them and bundled into one decision on Today, rather than forty alerts nobody opens.
Every decision comes back with a number
Commit to a reprice, a reorder or a transfer and its baseline freezes; when the window closes the ledger writes what it was worth, with the comparison the number rests on.
Decision never carried out — Couldn't measure, and no money is written
The objection that decides this page
Different POS systems. One shelf.
Unify · demo chain
what the three systems sent
e:corg POS
receipt line · in-store
LAPTE ZUZU 1,5% 1L
3 × 1.25 €
Shopify
online order
Zuzu Milk 1.5% 1L
2 × 1.45 €
1C ERP
purchase invoice
Lapte Zuzu 1.5% 1L, bax 12
0.92 € / unit
one line in the chain's model
| product | category | units | revenue | cost | margin |
|---|---|---|---|---|---|
| Zuzu milk 1.5% 1L | Dairy | 5 | 6.65 € | 4.60 € | 30.8% |
| White bread 500g | Bakery | 18 | 12.60 € | — | — |
no cost in any source — margin is left empty, not estimated
in-store and online stay separable · the same product is one SKU across every store
The questions a multi-store week actually asks.
Which store, which category, which product — and what it costs to leave any of them alone for another week.
Store P&L, costs included
Revenue, cost of goods, gross, operating expenses and net — per location and for the network, with OPEX broken out by category and imported from the ledger where it exists.
One store's bottom line before the month closes, drawn below.
SKU and category margin
Product-level performance across every location: what sells, what sits, inventory turnover, dead stock and transfers between stores.
The tail of the assortment stops being invisible.
Reports that name the action
Dead stock, working capital, store health, assortment, category P&L and replenishment — scheduled, and exportable as PDF.
A week of analyst work arrives as a file.
Compared with stores of the same format
Each location is placed against stores of the same format — revenue scale, average check, how many days it is actually open — and, once five or more stores compare, the distance to their median is priced per month.
A kiosk stops being measured against a flagship.
Every decision that reaches you carries its cost
Robust statistics flag a store, a category or a single SKU that has broken its own pattern, and separately flag the ones that have been broken for weeks, with the cost accumulated so far. Only what is worth at least €50 a week reaches you.
A store that has been quietly wrong all month says so.
Five roles, including the franchise partner
Owner, COO, analyst, store manager and franchise partner each get their own scope; a manager sees their own locations and nothing above them.
This can go in front of the shop floor.
P&L · one location
month to date
- Revenue
- 94 200 €
- Cost of goods
- − 71 500 €
- Gross profit
- 22 700 €24.1%
- Operating expenses
- − 19 400 €
- Net profit
- 3 300 €3.5%
Operating expenses
- Rent7 600 €
- Payroll9 300 €
- Utilities1 400 €
- Other1 100 €
same statement for one location or the whole network · OPEX comes from the ledger, or is entered by hand where it does not
Built for your kind of store.
The same product reads a supermarket, a bakery and a pet shop differently. Each format has its own page: its numbers, its questions and where the line is.
Compare sales and margin across all your stores.
Which store is behind, against what, and by how much. Each store is read against its own past and against the stores in your network that trade like it. The same questions can be asked in plain words to marql's AI data analyst.
Every store in one table
Sales, change, gross margin, average check and share of the chain for every store, ranked by revenue, with the chain total under it. Compare with the previous period or the same period last year.
Against stores that trade like it
Each store's page places its revenue, average check and margin among its peers, and names the categories where it is stronger or weaker than its format. Give stores a format and they are compared only within it.
The gap in money
When there are enough comparable stores, a margin below their median is priced per month, so the stores can be ordered by what the gap is worth.
Pace against plan, today
Set a target per store, and the day's view shows how many stores are behind pace and how much they are short by at this hour.
Store-level analytics for every location.
Open any store and its own numbers are on one page: how it trades, when, on what, and how it stands against last year. What marql's AI employees then do with them is set out in AI for retail operations.
One page per store
Revenue, margin, guests, average check and items per check for that store, over seven days up to a year, with revenue and margin drawn day by day.
Against last year, new stores apart
Any period can be set against the one before it or the same period last year. A store that has just opened is marked new rather than given a growth figure, so it does not flatter the chain.
When and what it sells
Peak hours, the strongest and weakest day of the week, sales by part of the day, and the products growing or falling most in money.
A weekly score per store
Every week each store gets a score out of 100 and a short passport: revenue, gross and net margin, dead stock against the chain median, and what is still open for it.
Find out-of-stocks and overstock from till and stock data, without shelf cameras.
Stock management across stores comes down to out-of-stocks, overstock and dead stock, store by store. A chain total can show plenty of stock while one store has had nothing to sell since Saturday and another sits on products that stopped selling. marql reads sales and stock store by store, so both show up instead of hiding in the total. What the AI does with that stock is set out in AI inventory management for retail chains. The routines behind it, from reorder points to counts, are in our stock control guide.
Out of stock, next to what it sells
Every product's revenue, units and margin sit next to its stock, and a product with nothing left is marked out of stock, for the chain or for one store.
What runs out next
Revenue at risk on products with less than a week of supply, and each product's days of supply, grouped into low, dead and fine.
Where capital is frozen
Stock with no sale in 30 days is priced as frozen capital, slow and dead products are listed with the value they hold, and dead stock (no sale in 60 days) is set against the month before.
Excess, store by store
Stock holding more than 60 days of cover is marked as excess, so each store's frozen capital splits into dead or slow stock and excess.
The line between a recommendation and an order.
A chain runs on systems that already work. The useful version of this product sits on top of them and is plain about what it cannot see.
marql has read-only access to your POS and accounting and does not change a price, a stock level or an order there.
With e:corg 2.0, e:corg creates an approved order or transfer as its own draft.
Margin needs cost.
Where neither the POS nor the accounting system carries a cost price, the product keeps its revenue and units and margin is left empty rather than estimated.
A few POS systems — Toast, Fudo, Toteat and Lightspeed among them — connect at account level today, with a seven-day history window, so for those systems per-location history starts when you connect rather than reaching back.
Replenishment is a recommendation.
The report says what to move and how much; no order is placed automatically.
Comparison happens inside your own network.
marql does not rank your chain against other companies' data.
Whether anything has to be replaced.
No. Every location keeps the POS it has. marql reads it — over the vendor's API for cloud systems, or through a small on-prem agent for a 1C or ecorg ERP that lives inside your network — and changes nothing in your POS or accounting: no prices, no stock levels, no orders. With e:corg 2.0, e:corg creates an approved order or transfer as its own draft.
It is the normal case, and the figure above is what it looks like: products, categories, VAT rates and currencies are mapped into one model, so two spellings of the same milk are one SKU and two countries are rows in one table. 151 connectors are live today across POS, accounting, e-commerce, delivery, payments and invoicing, and anything without one is mapped from a REST API or a scheduled CSV export without custom code — which is how most first connections start.
Wherever a cost price exists. When the POS sends cost, or the accounting system carries it, margin is computed per SKU and per category, for one store and across the chain. Where no source has a cost for a product, that line shows revenue and units and its margin stays empty — we would rather leave a gap than print an estimate you would find out about later. Category margin usually lands first, because products get mapped before every cost source is connected.
A 1C or ecorg ERP inside your network is read by an on-site agent, or by a nightly read for older ERPs; cloud systems sync every five minutes; the all-locations view refreshes about once a minute and analytics runs on an hourly grain. Every view stamps when the data landed and how many locations reported, so a POS that stopped sending is visible rather than quietly averaged into the total.
Access is by role: the owner sees everything including billing, a COO everything except billing, an analyst has read-only access to the whole picture, a store manager sees their own locations, and a franchise partner gets a read-only portal for theirs. A manager can be shown where their store stands without being shown the network's finances.
Yes. Shopify and WooCommerce come in as their own channel, and in-store Shopify POS sales are attributed to the location rather than merged into online — so one product can be read on the shelf and on the web without the two cancelling out. eMAG is connected for Romania and Bulgaria.
Per location, on a graduated ladder: €200 a month for each of the first nine locations, €185 for locations 10 to 29 and €160 for 30 to 49. Each rate applies only to the locations inside its band, so twelve locations are €2,355 a month rather than 12 × €185. From 50 locations we map the stack first and then quote. The calculator is on /pricing.
Also built for
From the blog
Ready to see
the euros it found?
The walkthrough is built around the POS and accounting you already run. Leave your details and pick a time.