Grocery delivery app saw 10x faster time to market

By using Pecan, they turned raw sales and inventory data into an effective demand forecast model in days instead of months.

produce display in grocery store



faster time-to-market


partner stores

data analyzed


accuracy for top revenue categories

The challenge

Competing internal priorities and 27 cities of data

Competing internal priorities for the data science team made it difficult for this grocer to build and maintain a demand forecast model that could accurately predict the sales volume at the SKU level across more than 27 cities with over 12,000 suppliers. This complexity of the modeling challenge made it impossible to deploy a scalable solution in-house.

This resulted in overstock of certain products at certain stores, and understock of other products at various stores creating massive inventory cost and a poor customer service experience.


From raw sales and SKU data to accurate predictions

Given the scope and complexity of this client’s supply chain, the desired predictive analytics project seemed insurmountable. Pecan achieved it in just a few days, beating the existing models created over several months in both accuracy and actionability. The confidence levels associated with predictions in this model were hyper-accurate, partially due to data enrichment with regional weather, traffic and economic data.


Unprecedented demand forecasting—and no more stockouts

By using the Pecan platform, the customer was able to achieve 10x faster time to market for their predictive models without compromising on model accuracy and usability. Since the business team could rely on Pecan to solve their long-awaited demand forecasting needs, the internal data science team was able to focus on more custom business questions and use cases. 

Pecan saw 80-90% accuracy across models consistently, even with the client building dozens of demand forecast models across time horizons, configurations, business aspects, frequency and more factors over several weeks.

For the highest value product categories, Pecan was able to accurately forecast 50% of the SKUs with less than 20% prediction error. Additionally, by leveraging Pecan’s proprietary Data Enrichment assets, the customer was able to uncover unexpected business insights about the strongest predictors of demand for top-priority SKUs.

Ultimately, the company saw overstock instances reduced by several orders of magnitude by using Pecan to forecast demand. They became more effective in supply capability, and improved their customer experience my minimizing stockouts to historical lows for the company.

Want to get started with predictive analytics?