Build a Living Team Dashboard in SharePoint by Describing Charts in Plain English

What you'll learn in this article:

• What the SharePoint AI Charts web part does and the license it requires
• The full setup to build a multi-chart dashboard page from connected data
• A reusable prompt pattern for describing any chart in one sentence
• How to lay out and refine charts without touching Excel
• The ROI math: time saved per chart and per dashboard rebuild

Internal dashboards rot because they are annoying to build. Someone exports data to Excel, builds a chart, formats it, screenshots or embeds it, and repeats that for every metric on the page. When the numbers change next month, the whole cycle runs again. So the dashboard goes stale, and the team falls back to asking the one person who has the spreadsheet open.

Microsoft's SharePoint AI Charts web part, rolling out across May 2026 and reaching general availability mid-to-late May, removes the build step entirely. A page author describes the data and chart format in plain English, and SharePoint generates an interactive chart on the page. No spreadsheet editing, no manual configuration. It requires a Microsoft 365 Copilot Premium license.

This article walks through building a real dashboard page from connected data, the prompt pattern that gets clean charts on the first try, and the ROI that makes it worth the 25-minute initial setup.

What Shipped and What You Need

The AI Charts web part is part of a broader SharePoint refresh that includes a Copilot-powered page creation workflow. The specifics that matter:

• License: Microsoft 365 Copilot Premium is required to use AI Charts.
• Rollout: Targeted Release began early-to-mid May; General Availability runs mid-to-late May 2026.
• Capability: describe the data and chart type in plain language and SharePoint builds an interactive chart on the page.
• No spreadsheet step: the chart is generated and configured by the AI, not built manually.
• Related feature: AI Citations Analytics shows how often your SharePoint documents are referenced in Copilot answers across the organization.

The practical requirement is that your data lives somewhere SharePoint can read, typically a SharePoint list or a connected file. Get the data in first, then build the visuals on top.

Step 1: Prepare the Data Source

• Put the numbers in a SharePoint list or a file the page can connect to. A list named clearly, for example RegionalRevenue, is easiest to reference in prompts.
• Make sure column names are plain and descriptive (Month, Region, Revenue), because you will name them in the prompt.
• Confirm you have a Microsoft 365 Copilot Premium license on the account building the page.

Clean column names are the single biggest factor in getting a correct chart on the first prompt. The AI maps your plain-language request to the columns it can see.

Step 2: Build the First Chart

• Open the SharePoint site page in edit mode.
• Add the AI Charts web part where you want the visual.
• Connect the list or file that holds your numbers.
• Describe the chart in one sentence using the pattern below.

AI CHARTS PROMPT PATTERN

Build a grouped bar chart showing monthly revenue by region for the last six months. Put months on the horizontal axis, revenue on the vertical axis in whole dollars, and use a separate color per region. Add a title that reads "Revenue by Region, Trailing 6 Months" and show data labels on each bar. Use the data from the connected list named "RegionalRevenue."

The pattern that works every time names five things: the chart type, the two axes, the formatting (colors, labels), the exact title, and the data source. Specify all five and the first output is usually the one you keep.

Step 3: Add the Rest of the Dashboard

Repeat Step 2 for each metric, changing only the chart type and the described data. A working executive dashboard usually needs three to five charts:

• A trend line for the headline metric over time.
• A grouped bar chart for a breakdown (by region, product, or team).
• A single-number or gauge-style visual for a target-versus-actual.
• A stacked bar for composition over time.

Add each as its own AI Charts web part, then arrange them into a clean grid on the page. Because each is described in plain language, a non-analyst can build the entire page.

Step 4: Refine by Re-Prompting

If a chart is close but not right, do not rebuild it. Re-prompt it. Change the one thing that is off:

REFINEMENT PROMPT

Update this chart: change it from a bar chart to a line chart, keep the same data and title, and add a dotted horizontal line at the target value of $[target].

Refinement by re-prompting is what keeps this fast. You are editing with sentences, not menus.

The ROI Math

Here is the reusable framing to justify the setup time. A single chart built the old way costs 20 to 30 minutes: export, build, format, embed. The AI Charts approach costs under 5 minutes per chart. For a five-chart dashboard, that is roughly 100 to 150 minutes saved on the first build.

The larger saving is the rebuild. A traditional dashboard has to be manually refreshed when the data changes, often monthly, costing another chunk of that time every cycle. A SharePoint page connected to a live list updates from the source, so the recurring rebuild cost drops toward zero. The reusable template here is the prompt pattern: name the chart type, axes, formatting, title, and source, and reuse it for every visual you ever build.

Action Steps Summary

• Prepare a clean data source: Put numbers in a clearly named SharePoint list with plain column names before building any charts.
• Build the first chart with the full pattern: Name the chart type, both axes, formatting, exact title, and data source in one sentence.
• Assemble the dashboard: Add a chart per metric and arrange them into a single page grid.
• Refine by re-prompting: Edit any chart by changing one detail in a follow-up prompt instead of rebuilding it.
• Bank the recurring savings: Connect charts to a live list so the page updates from the source and the monthly rebuild disappears.