El uso: cómo Canals implementa Elastic Cloud
Canals, which builds AI for wholesale distributors, uses Elastic Cloud to run a hybrid product-matching engine that combines keyword, dense vector, and sparse vector retrieval in a single index. Customer requests such as emails, PDFs, and spreadsheets are matched to the right catalog product, with a downstream ranking layer narrowing candidates to the top results.
Otros resultados publicados
Cifras publicadas en el caso de estudio de Elastic Cloud. usedby no las mide de forma independiente.
In roughly 90% of cases the system is confident enough that it does not show a list at all
serving on the order of 400,000 queries a day
On the products that do exist, recall reached 98%
Precision starts above 75% from the first interactions and climbs over a couple of weeks of usage
En palabras del cliente
Elasticsearch never blocked anything or slowed anything down. It let us run all of our experiments and all of our iterations and get to where we are today, and we're still iterating. Elasticsearch makes that easy. We could focus on building search rather than maintaining clusters.Otras empresas que utilizan Elastic Cloud
Clientes de distintos sectores, con pruebas públicas.
Ver todos los clientes de Elastic Cloud
Consultar la fuente
Basado en el caso de éxito publicado por Elastic Cloud. Verificamos las pruebas públicas; los resultados son los que declara el proveedor.
elastic.co- Última verificación
- 13 nov 2026
Abrir la captura Lo que dice la historia, y lo que verificamos
Lo que comparamos con la página.
- La cifra: 63% to 96%Verificado
La cifra está en la página; su etiqueta es redacción nuestra.
- El pasaje citado arribaVerificado
Copiado palabra por palabra de la página, cerca del nombre de Canals.
- Canals usa Elastic CloudVerificado
Línea de nivel Confirmado.
- El resultado en síNo verificado
Lo citamos; no lo medimos.
Canals, which builds AI for wholesale distributors, runs a hybrid keyword-and-vector product-matching engine on Elasticsearch that lifted top-3 match accuracy from 63% at launch to 96% today, with roughly 90% of matches now resolved without a human ever picking from a list.