A fundraising team wants a single chart showing cash donations received during October, not pledges promised or cumulative giving since January. Microsoft documents using Copilot in Excel with an attached Power BI report to create summaries, charts and editable workbook elements. That connection provides a useful source relationship, but it cannot settle an unclear financial definition for the organization. The amounts used here are invented teaching values, not organizational results or evidence of a live Power BI session.
Agree on one measurable reporting question
Write the question before opening Copilot: ‘What cash donations were received in October, by approved campaign category, in US dollars?’ Decide whether the report’s date is transaction date, deposit date or accounting recognition date. A fundraising dashboard can contain pledges, gross donations, refunds, and net receipts that sound similar but are not interchangeable. Record the approved measure name, currency, fiscal calendar and exclusions in a short definition sheet. If the team has no agreed basis for the chart, stop and ask finance to define it. A language model cannot decide accounting policy from a prompt alone.
Confirm the Power BI source and permissions
Open the governed Power BI report directly and inspect its title, workspace owner, dataset refresh date and visible slicers. Ensure you have access to the appropriate report rather than a copied screenshot or obsolete export. Microsoft describes adding Power BI content to an Excel Copilot session from the prompt’s Add content menu and Power BI tab. Availability depends on account, platform, license and organizational permissions. Do not copy an administrator’s credentials or use a different employee’s access to reach restricted reports. If the Power BI option is absent, perform the analysis in the approved reporting tool and document the limitation.

Ask Copilot for a narrow, traceable view
In Excel, ask a specific question about the attached report: ‘Make a chart of cash receipts by campaign for October, using the report’s approved cash-received measure, with all currency values in USD. Identify filters and do not combine pledges.’ Microsoft states that Copilot can create editable tables and charts from a Power BI report. Review the proposed measure names and groupings rather than assuming the model selected the intended field. Avoid asking for ‘our fundraising performance’ without specifying the unit, date and scope: the answer could mix targets, commitments and payments into a persuasive but inaccurate story.
Reconcile totals to the governed report
For a fictional October, imagine the report shows $12,400 of cash received and $17,900 of pledges. The chart should not show $30,300 under a single ‘donations received’ label. Open the underlying report and compare totals at the same date range, filters and campaign categories. Check whether a refund appears as a negative value and whether cancelled pledges are excluded. If the extracted workbook has additional rows from an unfiltered year, do not simply edit the grand total to make it match. Find the source or filter discrepancy and record how it was resolved.
Inspect chart axes and denominators
Even correct numbers can be presented misleadingly. Label the vertical axis in dollars; do not use a percentage axis for raw receipts. If showing ‘percent of goal,’ confirm the target denominator and its approved period before computing it. An October-only actual cannot fairly be compared with a full-year target without explicit context. Check that stacked bars do not inadvertently combine cash received and pledges into one total, and that color choices remain distinguishable in grayscale. Add a note explaining any omitted or provisional campaigns. Compare the chart’s displayed number labels with the numeric table, not with the illustration alone.

Compare filters before exporting
Inspect the reporting period, campaign, currency, date basis and any personal filters on the Power BI report. Changes to slicers can alter the visible measures without changing a chart title. If the Excel visualization was generated before a filter change, request a new calculation and verify it again. Keep a small reconciliation table with the source value, generated workbook value, difference and reviewer initials. For a cross-currency report, obtain an approved conversion method rather than asking Copilot to invent a foreign-exchange rate. A visually identical chart shape can still conceal a scope mismatch.
Review privacy and narrative claims
Many organizations restrict named donor data, even when aggregate reports are shareable. Use a report view limited to the intended audience, and inspect the generated workbook for names, transaction identifiers or internal notes inadvertently copied into supporting tables. When drafting an explanatory sentence, separate observation from cause: a higher month does not prove that a campaign message produced the increase. Avoid identifying individual donors from small chart segments. Confirm the report’s sharing permissions before emailing the workbook or pasting the chart into a public deck. A traceable report is not the same as blanket authorization to redistribute its underlying data.
Perform a final reviewer handoff
Give the finance reviewer the chart, source report link, measure definitions, relevant filters and an explicit comparison table. Ask them to approve the title, date window, totals, and any caveats. Export only the approved sheet or figure; charts in a workbook may depend on data ranges that change later. If the report is subsequently refreshed, assess whether the chart needs to be regenerated and marked with a new reporting date. Store the reviewed version under a name that includes the period and approval status, but keep internal batch IDs out of public imagery or public marketing copy.

When an ordinary PivotTable is the better option
A simple fixed export may be safer to reconcile with an ordinary Excel PivotTable or a preapproved Power BI visual than with a natural-language chart request. Use Copilot with a report attachment when interpretation across multiple approved measures is genuinely useful and the reviewer can identify discrepancies. Do not present the fictional $12,400 comparison as a tested result or a benchmark. This tutorial provides an audit recipe built from Microsoft’s documented feature, while real source data, entitlements and output quality must be checked in the user’s own environment.
How We Research
Our research reviewed named official vendor help resources: Microsoft Support — Analyze Power BI reports with Copilot in Excel; Microsoft Support — Get data insights with Copilot in Excel; Microsoft Support — Get started with Copilot in Excel. We checked documented feature scope, entry points, limitations, plan or seat requirements, and what the vendor does and does not claim. The example organization, figures, and errors are fictional instructional scenarios, not measurements from a live account. No source datasets, customer materials, actual exports, user interviews, or product-generated outputs were available for hands-on testing. All illustrations are original diagrams, not screenshots of a vendor interface. Before publication or real-world use, an editor should confirm current documentation, privacy and accessibility policies, and review the approved original source against every significant claim.
Frequently Asked Questions
Microsoft documents attaching a Power BI report to the Excel Copilot prompt and creating answers, charts and other editable workbook elements, subject to access and plan.
No. You still need to specify the measure, period, currency and filters and compare results with the governed report.
Not unless the organization explicitly defines and labels a combined measure. The fictional example separates cash receipts from pledges.
Only after verifying the figures and removing or restricting confidential underlying data. Report access does not automatically authorize external distribution.
Check platform support, Microsoft 365 Copilot eligibility and organizational permissions. Use an approved manual reporting workflow when the feature is unavailable.
