Lokad
Paris-based supply chain optimization editor using probabilistic forecasting to make buying, inventory and pricing decisions from data
Overview
Lokad is a French editor at 83-85 boulevard Vincent Auriol in Paris, focused on quantitative supply chain optimization. Its hallmark is probabilistic forecasting: instead of a single future demand, the platform calculates a probability distribution across all scenarios, then derives direct decisions (order quantities, warehouse allocation, sale price, clearance of unsold stock) valued in margin euros rather than forecast error percentage. The engine runs on Envision, a programming language Lokad built to model real business constraints like minimum order quantities, volume discounts or expiration dates. The editor publishes no grid: billing takes a monthly subscription form, cancellable month-to-month.
Lokad's site, commercial docs and support exist in French, and as a French-registered company its contracts and customer relationship follow French law, simplifying personal data handling for a French SME significantly. The platform technically runs on Microsoft Azure, with file exchange via SFTP and FTPS and documented connectors to NetSuite, Brightpearl and Cin7 Core. Two serious limits must be stated. First, Lokad is not self-service software: every deployment is backed by a "supply chain scientist" who writes the model, and the editor prices implementation at six months of runtime cost. Second, the approach assumes clean data history and sufficient decision volume.
Our verdict
Best for French distributors, manufacturers and e-commerce managing thousands of SKUs who want purchase decisions quantified in margin rather than a simple forecast table, with a French-speaking contact and French contract law. Not for you seeking self-service software or managing a small catalog: implementation cost does not pencil at that scale.