Review of SKU Science, Supply Chain Forecasting Software Vendor

Go back to Market Research

SKU Science is a cloud-hosted Software-as-a-Service solution launched around 2017–2018 that targets supply chain and operations management professionals by delivering demand and sales forecasts through an ensemble of statistical models. Designed for rapid deployment on AWS, the platform automates forecast model selection from a set of 644 preconfigured statistical methods, provides interactive dashboards for real‐time adjustments, and supports sales & operations planning and master scheduling. In addition to its forecasting capability, SKU Science offers practical tools such as product lifecycle management for smooth product transitions and robust security and compliance built on modern cloud standards. Although its claims of leveraging “advanced” artificial intelligence are largely underpinned by established statistical techniques rather than cutting‐edge deep learning, the solution remains an accessible tool for organizations focused on rapid implementation and ease of use in managing forecast accuracy and business performance.

Overview

SKU Science is a cloud-hosted SaaS solution for demand and sales forecasting oriented toward supply chain and operations management. According to its own descriptions on the company’s website and related pages¹², the platform was developed following consultations with supply chain professionals and was launched between 2017 and 2018. Independent sources have noted support from initiatives such as BPI France’s French Tech Acceleration Fund³. Its architecture, built on AWS with modern security standards, emphasizes rapid deployment and ease of use.

What the Solution Delivers, in Practical Terms

• Demand & Sales Forecasting: The platform ingests historical sales data (typically 2–3 years of records) to automatically generate forecasts at various levels of aggregation—from individual SKU details up to customer-specific views⁴⁵. These forecasts are designed to support Sales & Operations Planning (S&OP) as well as master scheduling.
• Interactive Dashboards & Reporting: SKU Science provides intuitive operational dashboards that display demand curves, backlog comparisons, and key performance indicators (KPIs) such as forecast accuracy and absolute error. Users can review both the raw and platform-generated forecast data alongside consensus figures⁴⁶.
• Product Lifecycle Management: An integrated feature supports managing product transitions by inheriting historical data—with configurable conversion ratios—from discontinued products to their replacements, ensuring continuity in planning⁷.

How the Solution Works

SKU Science emphasizes ease of use through features such as automatic model selection. The system fits an ensemble of 644 statistical forecasting methods to the available data and selects the best-performing forecast model based on historical performance⁵. A robust graphical interface allows users to visualize raw and consensus forecasts and manually adjust values as needed. Hosted on AWS and benefiting from rapid cloud deployment (with some setups reportedly completed in two days), the platform combines automated forecasting with interactive user oversight to balance statistical rigor and practical decision-making⁶³.

Analysis of the Technology and AI/ML Claims

SKU Science promotes its solution as “advanced” by touting its use of artificial intelligence and a broad range of statistical methods. In practice, however, the core forecasting mechanism relies largely on an ensemble of preconfigured statistical models rather than employing novel machine learning or deep learning techniques. Although the platform also offers consulting services for building custom ML models that incorporate additional variables such as promotions and pricing, detailed disclosures about these models (including algorithm types and performance metrics) remain limited⁸. As a result, while the system delivers practical forecast outputs and user-friendly visualizations, its claims of state-of-the-art AI remain more rooted in effective aggregation and automation of well-known methods than in innovative AI breakthroughs.

Additional Operational and Technical Details

The design of SKU Science is oriented toward non‐technical users. The solution provides interactive dashboards, multi-level forecast editing, and custom report generation. Videos and help articles demonstrate its modern interface, which facilitates manual adjustments and rapid review of performance metrics. With its deployment on AWS (hosted in Ireland), the solution emphasizes robust security (including ISO 27001 and SOC certifications) and compliance with leading standards. Although discussions of the company’s technical stack hint at a conventional web technology base (involving JavaScript and C++ components), there is limited evidence of proprietary frameworks or pioneering architectural innovations in the documentation²⁸.

SKU Science vs Lokad

While SKU Science deploys an ensemble of 644 statistical forecasting models to deliver rapid, user-friendly demand and sales forecasting on AWS, Lokad’s approach—detailed in its extensive technical brief—is markedly different. Lokad has evolved from early cloud-based “forecasting as a service” to a highly programmable, end-to-end predictive optimization platform built on Microsoft Azure. Whereas SKU Science relies on automatic model selection from standard statistical combinations⁵, Lokad integrates probabilistic forecasting with deep learning and a domain-specific language (Envision) to enable bespoke decision optimization⁹¹⁰. In practical terms, Lokad’s platform emphasizes not only generating forecasts but also automating actionable recommendations (such as inventory orders, pricing adjustments, and stock allocations) through advanced optimization algorithms and differentiable programming concepts¹¹¹². Thus, while SKU Science offers a more out-of-the-box, lightweight solution focused on ease of use and quick deployment, Lokad targets organizations seeking customizable, technically robust, and deeply integrated supply chain optimization solutions backed by advanced artificial intelligence capabilities¹³.

Conclusion

SKU Science presents a comprehensive and user-friendly cloud-based solution for demand and sales forecasting that emphasizes rapid deployment, automated model selection, and intuitive interactive dashboards. Its reliance on an ensemble of established statistical models makes it a practical tool for organizations with immediate forecasting needs, even if its use of “advanced AI” remains more marketing-driven than technically revolutionary. In contrast to platforms like Lokad—which combine deep learning, programmable optimization, and a more sophisticated technical stack—SKU Science is best suited for companies seeking a fast-to-deploy solution with a gentle learning curve. Prospective users should therefore weigh the benefits of a ready-to-use forecasting tool against the need for more customizable and cutting-edge predictive optimization capabilities.

Sources