November 2026: Faster AI Dashboards, Easier MCP Access, and Mosaic Mission Control

Quick Answer

• Build AI dashboards from existing designs. Auto Dashboard improves HTML imports and bulk editing; Auto Summary now covers layered pages.

• Bring more certified cubes into AI workflows. MCP can use certified in-memory and live Intelligent Cubes without separate AI enablement on each cube.

• See Mosaic operations in one place. Sentinel Mission Control brings risk, audit activity, usage, cost, and performance into a central view.

• Reuse more of what teams have built. Mosaic Studio unions, reusable Python metric functions, and in-canvas selectors reduce duplicated work and preserve context.

Enterprise AI works best when it builds on analytics teams already trust. A dashboard design should not need to be rebuilt from scratch. A certified Intelligent Cube should not need a second copy just to support an agent. And as AI workloads grow, platform teams need to see what those workloads are doing.

November’s updates address those practical needs. Auto Dashboard carries existing designs further into dashboard authoring. The Strategy MCP Server opens more certified cubes to AI workflows. Mosaic Sentinel brings operating signals together, while new modeling and BI improvements make established data and logic easier to reuse.

The common thread is less rebuilding, more governed reuse, and better visibility into what runs.

Start From the Dashboard Design You Already Have

A dashboard often exists as a design before anyone builds it in an analytics tool: an HTML mockup, a layout to adapt, or images and charts prepared for a new audience.

Auto Dashboard 2.0 now imports HTML layouts more reliably, including visualizations, rich text, images, containers, and filters. Authors can position uploaded images more precisely, and clearer messages flag unsupported files. That means less cleanup between the mockup and a dashboard ready for refinement.

The design becomes a useful starting point, not another rebuild.

Make Bulk Dashboard Changes in One Pass

A dataset may arrive with technical field names, inconsistent labels, or names that need to be translated for a local team. Renaming each object manually slows down work that should be straightforward.

Auto Dashboard can now rename attributes and metrics in bulk from one instruction. Authors can search by naming pattern and bring multiple objects, including derived objects, into the chat together. Teams can apply a naming convention in one pass instead of repeating the same edit across a dashboard.

Keep AI Summaries Working Across Layered Pages

Some dashboards place important details in panel stacks and info windows. Those layouts help people navigate complex analysis, but they have not always worked smoothly with page-level AI summaries.

Auto Narratives and Auto Summary now work across visualizations in panel stacks and info windows, including pages built entirely from layered layouts. Authors can set the summary’s initial size and position, and viewers can adjust it as needed.

The summary can follow the dashboard’s layout instead of dictating it.

Put More Certified Intelligent Cubes to Work Through MCP

Many enterprises already rely on certified Intelligent Cubes for reporting and analysis. Those governed assets should be able to serve an AI agent without being recreated or separately enabled for every new workflow.

The Strategy MCP Server can now access certified in-memory and live Intelligent Cubes without requiring separate AI enablement on each cube. In-memory cubes must still be published; live cubes do not have that publishing requirement.

For example, an agent answering a regional sales question can use an existing certified sales cube instead of asking a team to rebuild its data for AI. November expands which governed cubes can be used this way; MCP access to cubes itself is not new.

As more applications and agents use these assets, platform teams need to understand how usage, cost, performance, and risk change together.

Get the Operating Picture in Sentinel Mission Control

A rise in workload cost or a change in query performance rarely tells the whole story. Administrators need context to know what to investigate.

Imagine usage climbing after a new agent rollout. Is the change simply more queries? Is cost rising faster than volume? Are latency or audit signals changing too? Answering those questions is harder when each signal lives in a different place.

Mosaic Sentinel Mission Control now provides a central landing page for risk, audit activity, usage, cost, and performance, with alerts highlighting changes worth investigating. Administrators can review the operating picture and identify which signals or workloads warrant further investigation.

Mission Control does not replace investigation or automatically remediate issues. It helps teams decide where to look first and what deserves attention.

Combine Matching Tables Without Maintaining Another SQL Union

A retailer may keep the same sales columns in separate regional tables. Finance may store transactions by year. The business wants to analyze those records together without another custom SQL union to maintain.

Mosaic Studio now supports unions in its data-wrangling workflow. Modelers can combine two or more tables with matching columns from the same or different sources, retain all rows or remove duplicates, and later edit or split the union. The resulting logical table can be prepared, modeled, and published like other tables.

Use Python Functions Across Metric Calculations

A specialized calculation used in multiple reports should not produce different answers because each team maintains its own version. Reusable logic reduces that risk and makes updates easier to govern.

Workstation supports Python scripts deployed as reusable functions for metric calculations, providing an alternative to the legacy R integration. Administrators deploy the functions; architects and analysts can use them in standalone and derived metrics. The result is less duplicated code and more consistent business calculations across reports.

Compare Selected Categories Against All Others

Filtering a chart down to two regions makes the comparison easier to read, but it can also hide how much of the business sits outside those regions.

In-canvas selectors now let authors filter selected values or group unselected values into All Others when the selector and visualization use the same attribute. Select APAC and EMEA, for example, and the chart can show those regions alongside the combined remainder.

A regional manager can compare the selected regions without losing sight of the rest of the business.

Choose Dynamic Reporting Periods on the Canvas

Many dashboard reviews begin with the same question: how are we doing month to date, quarter to date, or year to date? Users should be able to change the reporting period where they see the result.

In-canvas calendar selectors now offer dynamic date choices already available in the filter panel, including month to date, quarter to date, and year to date. Viewers can switch periods without leaving the canvas or asking an author to rebuild the view.

What November Makes Easier

November makes more of the analytics customers already have available for new work. Dashboard authors can start from existing designs. AI agents can use more certified cubes. Modelers can reuse tables and custom calculations, while administrators gain a clearer picture of platform activity.

These improvements matter together. As analytics spreads across dashboards, applications, and agents, teams need to extend trusted definitions and workflows without multiplying copies, custom workarounds, or operating blind spots.

Enterprise AI should build on trusted analytics, not create another set of assets and workflows to maintain.

Learn more about the November updates on the Strategy Software What’s New page and in the product documentation.