Lightdash Data Apps building custom data visualisations from natural language is the product analytics capability teams have been asking for

Natural language prompts building custom data applications that maintain a live connection to the data warehouse via the Lightdash semantic layer is the combination that makes the output both accessible and governed. Non-technical analysts building custom views from natural language descriptions rather than waiting for a data engineer to build them is the self-service data analysis capability.

The live connection to the data warehouse via the semantic layer being what ensures the metrics and dimensions are governed rather than ad-hoc is the quality control mechanism that distinguishes Lightdash from a tool that lets anyone build anything with raw data access. The semantic layer defines what the numbers mean and the Data Apps use those definitions.

The decoupling of front-end from back-end allowing metric definition changes in the semantic layer to propagate automatically to all Data Apps is the maintenance efficiency that changes how sustainable custom analytics are over time as underlying data structures evolve.

For data teams: how many current custom analytical questions are waiting for engineering resources to build and would natural language data app creation change that backlog?