Skip to main content
usedby
DatabricksusedbyAdidas

How Adidas uses Databricks

ConfirmedVendor case study · Checked Nov 5, 2026
View captured proof

The use case: how Adidas deploys Databricks

adidas built a RAG-based GenAI chatbot that analyzes sentiment across its product reviews and returns insights to product decision-makers, including nontechnical users. It uses Databricks for embedding, vector search, model serving, governance and experiment tracking.

More reported results

Figures published in Databricks’s case study. They are not independently measured by usedby.

98.5%

More efficient prompt engineering and smaller context windows reduced token input size by 98.5%, from 200,000 to just 3,000 tokens per query.

30–40%

Through GenAI, adidas improved review analysis efficiency by up to 30–40%, cutting down extensive manual workloads across global teams.

91.67%

91.67% cost savings

In the customer’s words

The GenAI infrastructure we built is already enabling use cases beyond product reviews,
Rahul Pandey, Senior Solutions Architect, Adidas — Source

Other AI tools used by Adidas

Other public relationships recorded on usedby.

Insider OneConfirmedAmplitudeObservedHighRadiusObservedJasperObservedLovableObservedSonarObserved
Explore Adidas’s AI stack

Databricks users in Fashion & Apparel

Companies in the same sector, with public evidence.

ColumbiaConfirmedBurberryObservedPrada GroupObserved
See all Databricks customers

More stories

Go straight to the source

Based on Databricks’s published case study. We check the public evidence; the results are reported by the vendor.

databricks.com
Last checked
Nov 5, 2026
databricks.comCaptured Databricks case study about AdidasOpen source capture
What the story claims, and what we checked

What we compared with the page.

  • The figure: 60%Checked

    Printed word for word on the page, near the name of Adidas.

  • The passage quoted aboveChecked

    Copied word for word from the page, near the name of Adidas.

  • Adidas uses DatabricksChecked

    Confirmed line.

  • The result itselfNot checked

    We quote it; we did not measure it.

Latency dropped by 60%, cutting average response time from 15.5 seconds to six. More efficient prompt engineering and smaller context windows reduced token input size by 98.5%, from 200,000 to just 3,000 tokens per query.
From the page. databricks.com