Retail / Industrial supplies wholesale
Cromwell: AI product search and quote classification for B2B online ordering
- Company
- Cromwell
- Country
- United Kingdom
- Adoption stage
- In operation
- Source published
- Date basis
- The date the source was published. It can differ from the date adoption started.
- How the source was checked
- Read the full source text
The work problem
According to a case study by the e-commerce developer Elogic Commerce, Cromwell, a UK wholesale industrial distributor, sells a large specification-heavy catalog of tools, fasteners, abrasives, cutting tools and safety equipment to manufacturing, maintenance, repair and operations (MRO), engineering and automotive aftermarket customers. Its previous online ordering platform supported only standard product discovery, manual quote handling and rules-based recommendations. Repeat buyers spent a lot of time finding items they had ordered before, buyers searching by specification or manufacturer part number often could not find the right product through standard search, and quote teams spent much of their time manually classifying incoming quote requests and passing them to the right sales engineer.
Technology and data
Elogic Commerce added AI capabilities to the B2B ordering platform built on Adobe Commerce and connected it so that pricing, inventory and order history come from SAP S/4HANA and product attributes and specifications come from Akeneo PIM. Search adds LLM-based intent classification and extraction of items such as manufacturer, part number and specification to the existing catalog index, so it returns the right products even for partial part numbers, specifications or applications. For repeat buyers, it shows previously ordered items from SAP S/4HANA order history and predicts reorder timing, and recommended products are shown only after contract pricing and stock are checked. For incoming quote requests, AI determines the type, complexity, customer segment and processing route, automatically classifies and routes simple quotes, and prices many of them automatically using SAP S/4HANA contract pricing rules. Complex and exception quotes go to sales engineers with AI-generated summaries, and people handle them.
Results
According to Elogic Commerce, after the rollout the time to order for repeat buyers fell 34%, search-driven conversions rose 28%, and zero-result searches and abandoned searches also decreased. Automating 65% of the manual quote classification and routing work let sales engineers and quote teams spend their time on complex, high-value commercial work instead of routine quotes. Elogic Commerce said the AI outputs stayed consistent with SAP S/4HANA pricing, contract terms and stock, and with the product information in Akeneo PIM.
Limits and open questions
The case study was written by the vendor, Elogic Commerce, and does not disclose the measurement period or sample for its figures, the names of the AI models used, or when the system was introduced. Material published by Cromwell itself was not found in this research.
Sources
Compiled from public sources. These are not results from ATF Works customers.