The Challenge
A major U.S. food company was struggling with data exchange, unable to maintain a line of sight on its inventory or sales data. The company had no way to automatically match the data it was receiving from retailers, who used SKU (stock keeping unit) numbers that didn’t match their own. Because of that, a 10 percent sample was used to estimate the company’s inventory in stores, and with only 10 percent accuracy of this inventory data, the company had to extrapolate the total inventory.
This company dealt in very large volumes, shipping massive quantities of product to major retailers, including Wal-Mart, Amazon, and Target. A single percentage point of inaccuracy could have major consequences for the company’s revenue, business costs, invoice handling, supply chain management, and product delivery.
What’s more, the company found itself having problems with the information it did receive from retailers. Depending on which retailer was supplying the data, the information arrived in various unstructured formats, complicating and delaying the data integration process. In addition, the data often had spelling errors, non-standard abbreviations, and other syntax issues, making it difficult to organize, normalize, and interpret. It was impossible for IT teams to write a script to handle the noise that was coming in with that data, given the large volume and sheer number of variables.
Adeptia’s Solution
Adeptia AI was able to match and map the partners’ SKU numbers with the company’s own internal SKUs, which resulted in automating 70 percent of its data exchange, along with inventory and sales reports that were 90 percent accurate – as opposed to the 10 percent accuracy of those reports before Adeptia’s involvement.
Adeptia’s AI-augmented map-and-match capabilities were also able to cleanse, organize, and enrich the data regardless of how it was received, putting it into a format the company could easily use while correcting spelling, standardizing abbreviations, and normalizing the data. This added visibility into inventory data enabled the company to reduce its annual bottom-line expenses by a half-million dollars, saving time and increasing accuracy, which made it easier to do business with its partners and be more cost-effective in managing its inventory.
Before
- No line of sight on inventory and sales.
- Retailers using different SKU numbers.
- Manual effort to match the data from stores.
- 10% sample size to estimate store inventory.
- 10% accuracy and forced to extrapolate the total inventory.
After
- Matching and mapping SKU numbers with partners, automating 70% of data exchange.
- Data from partners automatically cleansed and formatted.
- 90% accuracy in inventory and sales reporting.
- Annual bottom-line expenses cut by $500,000.
