POS and inventory: why they need to talk to each other

The POS knows what was sold. The inventory system knows what was consumed. In theory, these two things should be reconcilable: the consumption should reflect the sales. In practice, they often are not, because the two systems do not share data.

This disconnect is one of the most common sources of unexplained variance in hospitality operations. Understanding what the connection between POS and inventory is supposed to produce — and what happens when it is absent — explains why building it matters.

What POS data tells you

The POS records every transaction: what was sold, when, in what quantity, at what price, through what payment method. This data is accurate and complete as a record of what left the till.

What it cannot tell you is what left the stock. The POS records that a burger was sold. It does not record that this burger contained 180g of beef, a brioche bun, a portion of chips, and two tablespoons of sauce. It does not deduct these quantities from the inventory.

For a retail operation selling discrete products, POS and inventory are effectively the same thing: each sale of a product deducts one unit. For a hospitality operation where each sale is the output of multiple ingredients, the POS and the inventory are recording different things.

What inventory data tells you

The inventory system records what was in stock at the start of the period, what was received from suppliers, and what was counted at the end of the period. The difference, adjusted for deliveries, is the consumption.

This consumption is the cost side of the GP calculation. It includes everything that left the stock: product sold, product wasted, product consumed by staff, product spilled, product stolen. The inventory calculation does not distinguish between these; it captures the total disappearance from stock.

The POS records sales. The inventory records disappearance. The gap between them is variance: the difference between what should have been consumed given the sales and what was actually consumed.

Why the gap matters

When POS and inventory do not share data, the variance calculation is manual. Someone exports sales data from the POS, exports consumption data from the inventory, and compares them. This process is time-consuming, error-prone, and usually happens too infrequently to be operationally useful. The wider picture of how POS, inventory, and finance work as three connected systems shows where this connection fits in the overall data flow.

When POS and inventory share data, the comparison is automatic. The system knows what was sold from the POS. It knows what was consumed from the stock take. The variance — what was consumed beyond what the sales would predict — is visible by category and by product, updated as data arrives.

This variance visibility is the operational tool. A category showing higher-than-expected consumption is worth investigating. A specific product with persistent variance identifies a specific issue. Neither is visible without the connection between the two data sources.

What "talking to each other" actually requires

For POS and inventory to exchange data usefully, two things are required.

First, a data connection. The POS needs to send transaction data to the inventory or back-office system automatically, not via a manual export process. This requires an API integration between the two systems. The guide to connecting the POS, accounting, and inventory triangle explains how all three systems fit together.

Second, a mapping layer. The POS records sales at the menu item or product level. The inventory tracks stock at the ingredient level. For the systems to compare like with like, there needs to be a mapping: this menu item contains these ingredients in these quantities. This mapping is the recipe or product specification layer.

Without the mapping, the POS data and the inventory data remain on parallel tracks that cannot be reconciled. With it, the sales from the POS can be converted into theoretical ingredient consumption, and that theoretical consumption can be compared against the actual consumption from the stock take.

The theoretical versus actual comparison

When the mapping is in place, the comparison between theoretical consumption (based on what the POS says was sold) and actual consumption (based on the stock take) produces the "variance to recipe" figure.

A product where actual consumption consistently exceeds theoretical consumption has one of a small number of explanations: the recipe is wrong (the yield assumption is too high, or the portion specification is too low), portion compliance has drifted, or there is loss that is not sale-related.

Identifying which explanation applies requires investigation. But the starting point — knowing that the variance exists, in which product, and by how much — requires the POS and inventory connection.

Hops connects POS sales data and inventory consumption data automatically, mapping sales to ingredient-level consumption via the recipe layer and surfacing variance by product. The comparison between what should have been consumed and what was actually consumed is available continuously, not assembled manually once a period.