Organization-level AI Memory will let admins promote useful insights from individual users' AI memory up to the organization level — surfacing suggestions to improve the semantic layer and dashboards based on real usage patterns.
Today, AI memory is per-user: what one analyst teaches the AI stays with that analyst. Promoting it org-wide means the whole team benefits. Over time, AI gets smarter for everyone, grounded in how your organization actually analyzes data.
AI Usage Analytics will give admins visibility into AI adoption across the organization: who is using AI, what types of questions get asked, and how usage trends over time.
This helps you measure ROI, identify training opportunities, and understand which teams get the most out of AI — so you can double down where it's working and unblock teams where it isn't.
Custom embedded app theme is planned for the Embedded Portal. Admins will be able to fully customize how Holistics looks inside their host app, matching brand colors, dark mode, and visual identity across the entire embedded experience.
Previously, embedded Holistics always rendered in light mode regardless of the host app's theme or brand colors. Now you can pass visual context via the embed payload and Holistics will adapt accordingly.
What's included:
AI Block turns your dashboard's data into a written narrative - so viewers get the "so what," not just rows and bars to interpret themselves.
Write a prompt once, and use it to:
The result persists as a saved block, and refreshes on demand or automatically when a source visualization changes.
Learn more: AI Block on dashboard
AI chat attachments let you add images and plain text or code files directly into your AI chat conversations for extra context.
With attachments, you can:
Supported file types:
Learn more: Conversation Context
AI Tasks let you define a reusable unit of AI work once, then run it automatically - on a schedule or on demand.
Use an AI Task for work that's repetitive, doesn't need your presence, and has a clear definition of "done." Because you set explicit acceptance criteria, every run is automatically checked pass or fail, not just generated.
Use it to:
Learn more: AI Tasks
Working long sessions in Holistics with a fixed light interface causes real eye strain.
This is why we're introducing dark mode as a personal preference. Users will be able to switch at any time, and the change will take effect across the entire app at two levels:
Timeline: Est. End of Sep 2026
Markdown documentation makes every .md file in your project a first-class object. Previously, a Markdown file was only text: frontmatter appeared as raw characters, and docs were disconnected from the analytics they explained. Now, frontmatter renders as a Properties panel, while Markdown links connect docs to related documents, models, datasets, and dashboards.
This keeps business context, modeling logic, and decisions alongside the analytics they explain, rather than buried in separate wikis.
What you get:
The same metadata becomes the groundwork for AI retrieval: agents will be able to narrow to approved context before answering, instead of scanning the whole project.
Learn more: Markdown for analytics context
AI Anomaly Detection scans a time-series metric for statistically unusual data points, so you can quickly tell whether a spike or dip is worth investigating before a meeting.
Click Analyze in the chart toolbar and select Detect anomaly. Holistics AI:
From there, you can save the generated chart directly to a dashboard to keep monitoring that metric over time.
This is part of Smart Analysis — and it pairs with Analyze Changes: once you know a movement is statistically unusual, you can dig into exactly what drove it, without leaving your dashboard.
You can now redact data for AI.
Admins and Customer Success reps often need PII access to do their jobs - but that shouldn't mean AI gets it too. Previously, AI inherited the full permissions of whoever asked it a question, with no way to hold it to a stricter standard.
Now, you can configure your semantic layer so sensitive data stays visible to the people who need it and out of reach for AI, no matter who's asking.