How is a timber data platform different from an ERP?

An ERP records each transaction as it happens: the purchase order, the receipt, the invoice, the payment. A data platform keeps those records over time and joins them with systems the ERP never sees, such as scale, harvest and production data. Then it answers questions across all of them. Being built for timber is not what separates them. Microsoft's documentation for Dynamics 365 keeps the two jobs apart, and says why. What makes the platform a timber one is the data it has to join.

What the usual answer says

The top results for this question are ERP vendors and ERP listicles. Their answer is that a good timber ERP does both. An ERP vendor for the timber trade writes that "Timber ERP systems offer real-time visibility into each stage of your operations, from logging to delivery," and that "Real-time data analysis and reporting capabilities of ERP systems also enable businesses to make more informed decisions." The argument turns on timber-specific features, such as scale tickets and stumpage, against generic ERPs that lack them.

Those features matter for recording transactions. Whether the same system is the right place to ask questions of five years of them is a separate matter.

What the ERP vendors do themselves

Microsoft's documentation is direct about it. Its Link to Microsoft Fabric feature, documented in July 2026 for "all Dynamics 365 apps, including Dynamics 365 Finance and Operations apps", sets up a separate analytical store. The documentation says the system "creates an optimized replica of your data" in a separate format. It does this "such that your operational workloads aren't impacted." The same page invites customers to combine that copy with data from "sensors, and signals from your warehouse and factories", which the ERP does not hold.

Microsoft's own analytics product for its ERP, Business performance analytics, states its limits in the documentation updated in September 2026. Customers "can currently report on up to eight quarters of data," and it "supports two data refreshes per day." Those are reasonable limits for a reporting tool attached to a transactional system, and Microsoft plans to raise both. They are also a description of what the ERP side is not built to do: long history, frequent refresh and data from outside.

Our reading of the reason is structural. An ERP's database is tuned to record each transaction correctly and quickly while people are working in it. A question like margin by customer and grade over three years reads a large share of the history across tables and systems. Microsoft's wording, "such that your operational workloads aren't impacted", suggests the same concern: the second job is kept off the first system.

Why timber businesses feel the gap sooner

Integration gaps are common even in large companies. In Salesforce's 2026 Connectivity Benchmark for MuleSoft, 1,050 IT leaders were surveyed at enterprises of "at least 1,000 employees". MuleSoft sells integration software, so it has an interest in the finding. The report found that "The number of apps in enterprises grew from 897 to 957 year over year, with only 27% of them integrated together." That figure does not transfer to a timber business running a handful of systems. What does transfer is the pattern: the scale house, the harvest plan, the optimiser and the ERP each hold part of the answer, and the ERP can only report on its own part.

A general analytical store over the ERP, such as Fabric, solves the load problem but not the timber one. The joins a timber business needs run between records in different units and from different sources. Scale tickets are in tons, harvest plans in volume and tallies in board feet, and some records exist only on paper. A timber data platform is one that makes those joins, and keeps them checkable.

What a data platform adds, and what it risks

A data platform holds the joined history and answers questions the ERP cannot. It also depends on joins, and every join between records is a rule someone wrote. From Quarri's own work with a sawmill: a join error was hiding about half of finished inventory from reporting. A wrong join doesn't announce itself. The report still runs, and can count less than there is, or count some records twice.

That sets the test for any data platform. Its numbers have to tie back to the ERP wherever the ERP holds the same fact, and it should show where they don't.

When it doesn't apply

A business whose questions all sit inside one ERP, and inside that ERP's reporting window, gains little from a separate platform. The ERP's own reports, or its vendor's analytics add-on, will do. The case for a platform starts when a question needs a second system, longer history than the ERP keeps for reporting, or answers faster than the refresh schedule allows.

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.