64 posts tagged with "new-feature"

AI dashboard building has been available for a while, and we've now made it more accessible, more reliable, and easier to use.

1️⃣ Now available in Reporting​

You can now use AI to create new dashboards and edit existing ones directly in Reporting. Previously, AI dashboard building was only available in Development. With this update, Explorers can build dashboards with AI for the first time.

2️⃣ A better, more reliable building experience​

• Specialized build skills: Holistics AI now uses dedicated built-in skills for different actions, such as creating a dashboard, applying a theme, or building a custom chart, instead of relying on one general-purpose step.
• Plan before building: AI proposes a plan first, so you can review the approach before it starts building.
• Start from what you already have: kick off a build from a plan doc, sketch, or other attachment shared in AI chat.
• Fewer timeouts: builds are now less likely to stall or time out midway.

There's more happening under the hood to make the overall experience smoother.

Learn more: AI dashboard building

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:

• Summarize a dashboard: "Summarize this dashboard and call out key insights" turns your data into a short narrative viewers can read at a glance.
• Analyze customer reviews: turn open-ended reviews into a sentiment trend and common themes, instead of reading through every one yourself.
• Brief different viewers differently: use AI Skills in the prompt so the same block gives the Head of CS a strategic brief, Support Agents an ordered work list, and everyone else a plain summary.

The result persists as a saved block. It refreshes automatically when the dashboard opens (if Auto-run on open is enabled), and asks before refreshing when a viewer changes a source visualization.

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:

• Rebuild something that lives outside Holistics: attach a screenshot of an external dashboard and ask the AI to recreate it.
• Build out new metrics from a spec: attach a screenshot of the metrics your product team shared in Slack, and the AI drafts them for you to review and approve.
• Analyze against a list or a number that isn't in your warehouse: attach this quarter's budget and compare it to actuals.

Supported file types:

• Images: PNG, JPG, WebP, GIF, SVG
• Plain text or code files: TXT, MD, HTML, CSV, JSON, XML, SQL

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:

• Automate a recurring report - summarize last week's sales by region under your team's conventions, and flag any region that moved more than 15%, every Monday morning.
• Roll up AI Chat Insights - aggregate the past week's AI Chat Insights into a ranked list of the context gaps affecting the most conversations.

Coming soon: AI Tasks will be able to send their results to Slack and email.

Learn more: AI Tasks

Cost Analytics provides a breakdown of your AI spend, so you can see where it's coming from instead of just the total. Find it under AI Monitoring → Cost analytics, with three views:

• Cost Overview: your AI spend at a glance.
• Cost by User: spend broken down by user.
• Cost by Chat: spend broken down by individual chat.

Break the numbers down further by time, environment (Development or Reporting), origin (which AI feature triggered the cost), and more.

Learn more: AI Billing

If a Databricks metric view already holds governed dimensions and measures, you can sync it into Holistics to build dashboards, implement embeds, and enable users to self-serve, without maintaining a duplicated definition.

At a high level, a Databricks metric view syncs into Holistics as a Table Model. Once synced, you can add the model to a dataset to build dashboards and embedded analytics.

Learn more​

• Databricks Metric Views

If a Snowflake semantic view already holds governed dimensions and measures, you can sync it into Holistics to build dashboards, implement embeds, and enable users to self-serve, without maintaining a duplicated definition.

At a high level, a Snowflake semantic view syncs into Holistics as a Table Model. Once synced, you can add the model to a dataset to build dashboards and embedded analytics.

Learn more​

• Snowflake Semantic Views

Following the launch of dark mode, we're introducing light & dark mode dashboard theme.

How it works​

Each viewer sees Holistics in either light or dark mode, depending on their user settings. When you assign a theme to a dashboard, you can now choose how that theme responds to each viewer's mode:

• One theme for both modes: Everyone sees the same theme, regardless of their app mode.
• A theme per mode: Each viewer gets the theme matching their mode.

Learn more: Create & apply a theme | Syntax reference

Working long sessions in Holistics with a fixed light interface causes real eye strain. We've added dark mode as a personal preference:

• Switch between light, dark, or system mode at any time from your profile settings.
• The entire app updates instantly, including navigation, backgrounds, dialogs, and controls.

Your assigned dashboard theme stays static, but you can set separate themes for light and dark mode with light & dark mode dashboard theme.

Learn more: Switch to light/dark mode

Trace AI-triggered queries and jobs is now available across the Jobs API and warehouse query comments. You can now tell whether a query was triggered by AI, through which surface, and how the caller authenticated.

Previously, query metadata had no way to distinguish an AI-triggered query from a normal human-driven one, making it hard to attribute warehouse cost or audit usage by origin as AI usage grows.

Learn more: Trace AI-triggered queries and jobs