1,840 to 2,640 qualified leads a month.No new ad spend.

The client sells hydraulic fittings, adapters, and assemblies. Roughly 8,500 SKUs across 150-plus categories. By mid-2024, qualified inbounds had flattened at about 1,840 a month, and the cost of buying growth back through paid search kept climbing.

You could see why on the results page. AI Overviews had started answering the engineering queries that used to land on the client’s category pages, and they cited the manufacturers — whose spec sheets were the cleanest structured source around. In AI answers, a distributor sits downstream of its own suppliers until its catalog gives the engines something better to cite.

Most of the catalog copy was manufacturer-supplied. The same descriptions ran on every competitor’s site, and the product schema had gaps wide enough that engines had no reason to pick the client’s pages over the manufacturer’s. The brief had one hard line: grow qualified inbounds without adding ad budget. Whatever moved had to come from the catalog itself.

AI engines cite whichever page makes the answer easiest to assemble. The manufacturers won by default because their spec data was structured. The schema rewrite handed that advantage to the distributor, which had the same specs plus stock, pricing, cross-compatibility, and the application context a manufacturer page never carries.

Once the category pages and answer hubs gave the engines a better source, the citations followed. And the traffic those answers send arrives pre-qualified. An engineer who clicks through from a thread-sizing answer already knows what they need. That’s why the lead count climbed faster than raw sessions did.