Following the success of the DataMine Machine Learning Service, CME Group identified an opportunity to extend QDT's forecasting capabilities to their institutional client base -- including major energy companies like Shell and Chevron, and agricultural giants like Cargill -- who required ready-made, institutional-grade commodity forecasts.

• Institutional clients needed turnkey forecasting without building custom models.
  

• Forecasts had to cover energy, agriculture, and metals traded on CME exchanges.
  

• The solution required seamless integration with CME's existing market data feeds.
  

CME and QDT launched a combined forecasting solution, packaging QDT's predictive models on top of CME's proprietary exchange data and delivering them directly to enterprise subscribers.

• Pre-built forecasting models covering commodities across CME's exchange ecosystem.
  

• White-label delivery through CME's existing enterprise data distribution channels.
  

• Integration with CME market data feeds for real-time model updates.
  

• Enterprise-grade access for clients including Shell, Chevron, and Cargill.
  

Based on the success of the DataMine Machine Learning Service, this second collaboration demonstrates the depth of the CME-QDT partnership. Institutional clients now access sophisticated commodity forecasts as part of their CME data subscription, bringing quantitative forecasting to enterprise decision-makers across the global commodities market.

Mars needed to optimize procurement strategies for their pet nutrition division by predicting price movements across key commodities used in pet food manufacturing.

• Complex commodity markets with high volatility affecting procurement costs.
  

• Need for division-specific forecasting tailored to pet nutrition supply chain.
  

• Integration requirements with existing internal data systems and workflows.
  

We developed a bespoke forecasting solution specifically designed for Mars's pet nutrition division, with custom predictors and seamless integration.

• Custom predictors built for the pet nutrition supply chain.
  

• Integration with Mars internal data systems for real-time data flow.
  

• Real-time forecasting dashboard for procurement teams.
  

• Automated procurement recommendations based on model outputs.
  

Mars achieved significant cost savings in procurement through precision commodity forecasting, enabling them to hedge against market volatility and optimize purchasing strategies across their pet nutrition division.

NSE Cogencis wanted to differentiate their market data terminal by providing subscribers with access to sophisticated predictive analytics beyond standard market data.

• Need for native integration within the existing Cogencis terminal interface.
  

• Requirement for benchmark index forecasting including the Nifty 50.
  

• Serving thousands of financial professionals across India's trading ecosystem.
  

We integrated QDT's forecasting models directly into the Cogencis terminal, providing a seamless analytics experience for their subscribers.

• Native integration with the Cogencis terminal for a frictionless user experience.
  

• Nifty 50 and sectoral index forecasting models.
  

• Real-time model updates reflecting latest market conditions.
  

• Accessible to thousands of terminal subscribers across India.
  

NSE Cogencis now offers a differentiated terminal product with integrated predictive analytics, providing their subscribers with forecasting capabilities that were previously available only to institutional quantitative teams.

McDonald's leadership needed real-time macroeconomic intelligence across every region they operate in to inform strategic decisions around expansion, pricing, and workforce planning.

• Monitoring inflation trends, consumer spending, and labor market shifts globally.
  

• Requiring clean, structured data flowing directly into existing PowerBI infrastructure.
  

• Enabling non-technical leadership to access complex economic insights.
  

We integrated QDT's macroeconomics data directly into McDonald's PowerBI dashboards, providing automated data flows across all key economic indicators.

• Direct API integration with McDonald's PowerBI environment.
  

• Automated ingestion of inflation, employment, and consumer spending data.
  

• Region-specific dashboards tailored to each operational market.
  

• Structured data delivery requiring zero manual transformation.
  

McDonald's leadership team now makes strategic decisions backed by real-time macroeconomic data, transforming how they approach regional pricing, expansion planning, and workforce strategy across their global operations.

The Hershey Company needed to improve demand forecasting accuracy across their entire portfolio of brands, spanning multiple product categories, regions, and seasonal cycles.

• Forecasting demand across dozens of brands with distinct consumer behavior patterns.
  

• Requiring predictions at multiple timeframes -- weekly, monthly, and quarterly -- for different planning functions.
  

• Existing models struggled to account for promotional effects, seasonal surges, and macroeconomic shifts simultaneously.
  

QDT deployed bespoke demand forecasting models for every Hershey brand, calibrated to multiple time horizons and enriched with external data signals.

• Brand-level demand models covering the full Hershey portfolio including Reese's, Kit Kat, and Hershey's.
  

• Multi-horizon forecasting delivering weekly, monthly, and quarterly demand predictions.
  

• External signal integration including macroeconomic indicators, weather patterns, and consumer sentiment data.
  

• Automated model retraining to adapt to shifting demand patterns and promotional calendars.
  

The Hershey Company now operates with a unified demand forecasting system across all brands and time horizons, enabling procurement, production planning, and retail distribution teams to align around a single source of predictive intelligence.

Campbell's needed a data-driven approach to setting optimal prices across their full range of soup products. Traditional pricing methods relied on historical rules of thumb and lagging market research, making it difficult to respond to shifting consumer demand and competitive dynamics in real time.

• Setting the right price point across a wide portfolio of soup products with varying demand profiles.
  

• Understanding the relationship between price changes and sales volume for each product.
  

• Accounting for external factors like competitor pricing, seasonality, and macroeconomic conditions.
  

QDT first ran demand forecasts for Campbell's soup portfolio, using price data as one of the key predictors. By analyzing the feature importance of price within the forecasting models, we extracted precise price elasticity coefficients and built a dynamic pricing model driven by the interplay of sales, price, and elasticity.

• Demand forecasting models with price as a core predictor across all soup products.
  

• Feature importance analysis to isolate the impact of price on demand for each product.
  

• Dynamic pricing engine built on sales volume, price sensitivity, and elasticity curves.
  

• Scenario modeling allowing Campbell's to simulate the revenue impact of pricing decisions before execution.
  

Campbell's now sets prices informed by real elasticity data rather than intuition. The dynamic pricing model enables their revenue management team to optimize margins across the entire soup portfolio while maintaining competitive positioning and volume targets.