Toolio leverages a combination of statistical and machine learning models, including XGBoost and Triple Exponential Smoothing for time series forecasting, and K-Means for clustering. These models are used to optimize demand forecasting, inventory planning, and assortment decisions, ensuring both accuracy and explainability for end users. Toolio’s approach allows for flexibility and transparency in how forecasts are generated and adjusted by users.

Forecast accuracy is measured using metrics such as Mean Absolute Error (MAE) and Bias. Toolio regularly runs forecasting competitions between models and retrains them monthly or weekly, depending on the use case. Users can compare system-generated forecasts with actuals and user overrides, and Toolio provides clear visibility into model performance.

Yes, users have full control to override AI-driven recommendations. Toolio provides visibility into the key inputs and outputs influencing forecasts and recommendations, allowing users to understand the drivers behind the results and make manual adjustments as needed. Explainability is a core value, ensuring users are never left with “black box” outputs.

Toolio’s AI includes mechanisms for true demand calculation, anomaly detection, and data cleansing to remove the impact of stock-outs and outliers. While Toolio primarily focuses on internal retail data, it can incorporate external factors such as promotions, seasonality, and even weather or market data if provided by the customer.

Yes, Toolio’s AI models can be tuned and retrained based on customer data and feedback. The platform supports custom attributes, metrics, and even the integration of internally built models at the SKU x Location level, ensuring alignment with unique business requirements.

Toolio enables users to create multiple planning scenarios, simulate different demand, pricing, and inventory situations, and instantly see the projected impact on sales, margin, and inventory. This helps retailers evaluate risks and opportunities, supporting more informed decision-making.

Yes, Toolio supports multi-location and multi-channel operations, allowing for granular planning by store, region, or channel. The AI can tailor forecasts and allocations based on product attributes, store clusters, and regional demand patterns, supporting strategies like BOSS (Buy Online, Ship from Store) and channel-specific planning.

Toolio’s main machine learning models are retrained on a monthly or weekly basis, depending on the use case and data availability. This ensures that forecasts and recommendations remain relevant and accurate as business conditions evolve.

The strongest AI/ML use cases in Toolio include SKU rationalization (suggesting optimal assortment breadth and depth), demand forecasting, and store or product clustering. These capabilities help retailers optimize inventory, improve sales forecasting, and tailor assortments to customer needs.

Toolio incorporates an AI-powered chatbot (“Tooli”) to help users quickly surface key insights, exceptions, and recommendations. The platform’s intuitive dashboards, customizable workflows, and robust training resources further streamline decision-making and support collaboration across planning teams.