Natsana is the Düsseldorf company behind four leading German nutrition brands. Gemma Analytics built two things for Natsana on a shared Snowflake and dbt data platform: intelligent inventory management with predictive stock planning, and true profitability for every single product.
The starting point
Natsana sells natural supplements under Nature Love, Natural Elements, Gloryfeel and Feel Natural, with a strong position on Amazon, in drugstores and in its own online shops. Gemma has worked alongside Natsana's in-house data team since 2021, and two problems kept coming back. The first was stock. Warehouses, manufacturers and the planning sheet each reported their own number for the same product, so nobody could say with confidence how much was on hand or on its way. The result was stockouts, excess inventory and planning that never felt solid.
The second was profitability. Leadership saw one blended margin across the whole range. Whether a single product made or lost money remained an open question, because procurement, production and logistics costs never came together at product level.
What we built
For inventory, Gemma built a unified data foundation for sales and operations planning. Clean stock histories, in-transit tracking and manufacturer flows come together in one inventory model, accurate down to the batch, with one stock number per product. On top of it sits predictive inventory planning: a forward-looking model calculates out-of-stock dates in advance, accounts for when reorders will arrive and adjusts when reality deviates from the plan. Every stock risk also carries a value in euros, so decisions reach management with a clear price tag.
For profitability, Gemma built a granular profit and loss model that gives every SKU its own true margin instead of one blended number. Procurement, production and logistics costs sit in one view and are assigned to the product. Revenue and cost forecasts per SKU extend the picture forward, so margins are managed ahead instead of explained afterwards.
The result
Natsana now sees stock risks before they hurt: one trusted stock figure per product, sell-outs foreseen early enough to reorder, and the financial impact of each stock decision in euros. On the profitability side, the true margin of every SKU is visible and hidden loss-makers are flagged, so they can be dealt with. Product-level profitability delivered an EBITDA impact of EUR 0.7 million.
Tools and setup
Data from the shop and order systems, the Amazon marketplace, the ERP and the warehousing and fulfilment partners is loaded with Airflow into Snowflake, the central data warehouse. The inventory and profit and loss logic is modelled, tested and documented in dbt, and the results reach the business teams in Power BI dashboards.