How West Coast Batteries solved complex battery recommendations with Chatbase

At a glance

West Coast Batteries is a wholesale distribution company for industrial batteries based in Corona, California, serving cars, trucks, airplanes, boats, data centers and more. Owner Albert Diaz, a former senior executive at Intel, Microsoft and Symantec, acquired the 27-year-old family business in 2024 and set out to build a world-class direct-to-consumer ecommerce experience for battery buyers. The challenge: battery recommendations are complex, getting them wrong is expensive, and no existing tool could handle it.

A problem that databases alone cannot solve

When your mechanic tells you to replace your battery, most people have no idea which battery their car actually needs. There are industry-standard group sizes, multiple batteries that fit each size, and then a further layer of decision-making based on how you actually use the vehicle. For Albert, this was not just a customer experience problem. It was a financial one. Getting a recommendation wrong meant an expensive return, an unhappy customer, and a reputational hit for a business he had just invested in.

Six to eight weeks of nothing, then Chatbase

Albert had already spent months trying to solve this with another AI vendor. After six to eight weeks of calls, he had made no meaningful progress. The tool's instruction limit was 10,000 characters, far too small for the complexity of the problem. He decided to stop waiting and start testing himself. Already running his business on Claude, he built a prototype recommendation engine to prove the concept worked. It did. Then he went looking for a platform that could deploy it properly. Chatbase's 20,000 character instruction limit gave Albert the space he needed. The answer correction feature also helped: as he worked through debugging, he could review the chat log and revise any answer that did not meet his standard.

The results

• Live in week one. After six to eight weeks of no progress with a previous vendor, Albert deployed Rafa, the WCB Battery Recommendation Expert, by himself, without any external help, within the first week of switching to Chatbase.
• Accurate recommendations across a complex product catalogue. Rafa correctly identifies the right Standard Size battery for vehicles going back to the 1960s and then recommends the right product, from the options WCB carries, based on how the customer actually uses their vehicle.
• The team uses Rafa too. Instead of looking up answers across four or five different systems, the WCB team now plugs customer requests into the assistant on their own website.
• A favourite feature: model comparison. The ability to compare Sonnet versus Opus side-by-side lets Albert pick the right model for the task at hand.

What's next

Albert is actively exploring how to use Chatbase for internal knowledge management, giving employees a single place to ask questions and get instant answers.