South KoreaRetail and e-commerce2026 – ongoingAbly
NABI natural-language data assistant
Ably employees asked NABI at least 1,000 data questions in about 2 months
Problem
Repeated requests for simple SQL queries interrupted Ably's analysts. Other employees waited for answers before making decisions.
Approach
Ably built NABI to translate employee requests into SQL. The assistant consulted metadata consolidated in dbt before generating queries and supplied Redash links for checking results. Uncertain requests triggered clarification questions.
Results
Takeaway
Building NABI required Ably to consolidate metric definitions. Employee feedback then helped analysts correct gaps in that metadata.
Categories
- Industry
- Retail and e-commerce
- Business function
- Internal work and knowledge management
- Innovation type
- Process innovation
- AI technology
- Generative AIAI agents
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Ably's data team described NABI's metadata controls and query verification. More than 100 employees submitted at least 1,000 questions in about 2 months, while simple SQL requests to analysts declined without a published percentage.
Full details are in the original sources.
