Industry
Retail & Consumer Packaged Goods (CPG)
reduction in Transfer Engine runtime, from 240 minutes to 71 minutes
of extended planning capacity to support long-term strategic goals
purchase order generation and streamlined procurement workflows
prediction accuracy consistent with prior benchmark performance
A renowned premium apparel company relies heavily on merchandise planning and allocation to keep stores stocked with the right inventory at the right time. After moving off Anaplan to build its own PySpark-based Transfer Engine, called TIGER, for greater customization and flexibility, the client found the new system’s execution time falling far short of what the business needed.
Altimetrik partnered with the client’s internal team to analyze and optimize the Transfer Engine’s codebase, cutting runtime dramatically and extending the system’s planning horizon, while preserving the prediction accuracy the business depended on.
Despite closely mirroring Anaplan’s codebase, the client’s in-house Transfer Engine ran far slower than the system it was meant to replace, taking roughly 240 minutes to complete processes that Anaplan finished in about 60. That gap directly affected planning cycles and store operations.
The extended runtime delayed decision-making across inventory management, order planning, and store layout, with a direct impact on revenue potential and customer experience. The client needed a customizable, scalable solution that could match its operational ambitions.
Efficient merchandise planning is critical to ensuring the right products reach the right stores at the right time. By optimizing its custom Transfer Engine, the retailer significantly reduced planning delays, expanded its forecasting horizon, and strengthened its ability to make faster inventory and procurement decisions without compromising accuracy.
The Solution: Targeted Code Optimization and Architectural Refinement
Altimetrik worked alongside the client’s internal team to diagnose and resolve the Transfer Engine’s performance issues, starting with a focused pilot on a single high-impact category. The approach included:
Optimized the Transfer Engine to dramatically accelerate merchandise planning and allocation processes.
Increased planning capacity to 500 days, enabling more strategic inventory and merchandise decisions.
Automated purchase order generation to streamline procurement workflows and reduce manual effort.
Preserved prediction accuracy while delivering significant performance improvements.
Established architectural improvements to support continued performance optimization as planning needs evolve.
This engagement reflects how Altimetrik approaches performance challenges: not chasing a perfect number, but building a durable foundation for continuous improvement. While the original one-hour runtime target wasn’t fully reached, the 70% improvement and extended 500-day planning capacity gave the client a meaningfully faster, more capable system, without sacrificing the accuracy it depended on.
As the client continues refining its planning capabilities in a competitive, fast-moving retail landscape, this engagement laid the technical and strategic groundwork for the next round of enhancements.
