Holistics AI Roadmap
Here's what we're building for Holistics AI over the next 1–2 quarters. Priorities may shift based on customer feedback and usage patterns.
Chat experience
Chat history
Access your previous AI conversations anytime. Pick up where you left off without losing context or repeating questions.
Dashboard intelligence
AI-powered dashboard summaries
Automatically generate and save dynamic summaries on dashboards. Summaries highlight key trends, notable changes, and actionable insights — giving readers a narrative alongside the data. This helps stakeholders who don't have time to interpret every chart get the story at a glance.
Extensibility & integrations
OAuth for MCP Server
Connect to the Holistics MCP Server using web-based AI agents (like Claude.ai or ChatGPT) without manual token setup. OAuth support reduces configuration friction and makes MCP accessible to non-technical users — so more of your team can benefit from AI-powered workflows without IT involvement.
Local development with AI agents
Use local AI agents like Claude Code, Cursor, or GitHub Copilot alongside Holistics through MCP. Local agents bring access to richer development tools — git actions, diffs, auto-complete, and inline code suggestions — enabling a more productive modeling workflow. This lets data teams use the development tools they already know while staying connected to Holistics.
Evaluation & continuous improvement
Organization-level AI memory
Allow admins to promote individual user memory insights to the organization level — surfacing suggestions to improve the semantic layer and dashboards based on real usage patterns. Over time, this makes AI smarter for everyone in your organization, not just individual users.
Usage analytics
Give admins visibility into AI adoption across the organization: who is using AI, what types of questions are being asked, and how usage trends over time. This helps you measure ROI, identify training opportunities, and understand which teams benefit most from AI.
Answer quality feedback loop
Let users rate AI answers to feed a continuous improvement cycle. Ratings are used to identify weak spots and improve AI accuracy over time. The more your team uses and rates AI, the better it gets at answering your organization's specific questions.