What is a data warehouse?

A data warehouse is a database built for analysis rather than for running the business. It copies data from the systems that do run it, such as the ERP, the scale system and the production system, keeps its history, and organises it so that questions across the business get one consistent answer. IBM describes it as a store that "aggregates data from various sources into a central data store optimized for querying and analysis". The idea, it notes, "emerged in the 1980s to integrate disparate data into a consistent format for analysis".

Why it exists

Transaction systems are built to record today's work quickly. Many keep their own history, but each on its own terms. A warehouse lines up every system's history on the same definitions, so a question such as which stock hasn't moved gets one answer. From Quarri's own work with a lumber and millwork manufacturer: 34%+ of stocked SKUs had not sold in over twelve months.

In a timber business

The harder part is that timber history keeps changing. Scale tickets arrive days after the load, weights are corrected, and settlements are adjusted after a month is closed. Warehouse designers call these late-arriving facts. The Kimball Group, whose dimensional modelling techniques are a standard reference for warehouse design, defines one as a record whose context "does not match the incoming row", which "happens when the fact row is delayed". Its rule is to look up the context "that were effective when the late arriving measurement event occurred": the contract, price and customer as they stood on the day of the load, not today.

That leaves a choice. A late ticket can restate last month, or be booked in this month. Finance usually needs both: the figure as first reported, and the figure as it now stands.

What it isn't

A warehouse isn't the ERP, which runs transactions. It isn't a data lake, which stores raw files of any kind before they are organised. On its own, it isn't a whole data platform either.

What to check

Ask whether it keeps history or only the latest values, and where each measure, such as volume or margin, is defined. Above all, ask how late records are handled once a period has been reported, and whether you can see both versions.

How Quarri works explains the platform as a layer over existing systems, not a migration.

Quarri is an AI-native data platform for the timber supply chain. It connects buying, production, sales and inventory for forest management, sawmill, wood products and pulp, paper and packaging operators.