Use Excel Copilot to Analyze QC Trends and Spot Rule Violations

What This Does

Excel Copilot analyzes your QC data to identify Westgard rule violations, calculate CV%, and spot systematic trends, the same work you'd do manually reviewing Levy-Jennings charts, but faster and more systematic. It can also create the Levy-Jennings chart from your data automatically.

Before You Start

• You use Microsoft Excel (desktop or online) for QC data tracking
• Copilot is enabled on your Microsoft 365 account (most organizations buy the Microsoft 365 Copilot add-on; smaller ones may have free basic access, check with IT if you're not sure)
• You have QC data in an Excel spreadsheet, at minimum: date, analyte name, control level, observed value, mean, SD

Steps

1. Open your QC data in Excel

Open the spreadsheet containing your QC results. The data should be in a table format with columns for date, analyte, control level, and observed value. If you have multiple analytes, they can be in separate sheets or tabs. Copilot can work with either.

2. Format as a table (if not already)

Select your data and press Ctrl+T to format as an Excel table. A table is no longer required, Copilot also reads plain ranges with a clean header row, but it keeps your ranges stable as data grows. Give each column a clear header: Date, Analyte, Level, Result, Mean, SD.

3. Activate Copilot

Click the Copilot icon in the lower-right corner of the Excel window. The Copilot panel opens on the right.

Microsoft moved Copilot off the Home ribbon in 2026; the corner button is now the one entry point. Press F6 to reach it from the keyboard, or right-click the corner button and choose Move to ribbon if you prefer the old placement.

4. Ask for a QC performance summary

Type: "Analyze this QC data. For each analyte and control level: calculate the CV%, identify any results outside ±2 SD, identify any results outside ±3 SD, and flag any systematic trends (7 consecutive values moving in one direction)."

What you should see: A table or list showing each analyte's CV%, count of 2s violations, count of 3s violations, and any trending flags.

5. Ask Copilot to create a Levy-Jennings chart

For any analyte of interest, type: "Create a Levy-Jennings chart for [analyte name], [control level]. Add horizontal lines at the mean, ±1 SD, ±2 SD, and ±3 SD."

What you should see: A time-series chart is inserted in the spreadsheet with the correct reference lines. If Excel can't automatically add the ±SD reference lines, ask: "Add four horizontal reference lines at these Y values: [mean+2SD], [mean+1SD], [mean-1SD], [mean-2SD]."

6. Generate narrative for your monthly QC report

Copy the Copilot output summarizing violations and trends, then go to a free chatbot (ChatGPT/Claude) and ask it to write a narrative paragraph from those statistics for your monthly QC report.

Real Example

Scenario: You have 30 days of chemistry QC data for 12 analytes across 2 control levels.

You ask Copilot: "Analyze this QC data. Calculate CV% for each analyte/level, flag any ±3 SD exceedances, and identify any systematic trends."

Copilot returns: "Glucose L1: CV 1.2%, 0 violations. Sodium L2: CV 0.8%, 1 result at +3.1 SD on 3/15, corrective action confirmed by comment. Potassium L1: CV 5.8%, exceeds typical acceptance criterion of 5%. Trend: Calcium L2 shows 8 consecutive increasing values (7T rule violation potential)."

What you do: Document the potassium CV% increase (reagent lot change on 3/12, investigate). Document the calcium trending for supervisor review. Monthly report takes 15 minutes instead of 45.

Tips

• The more structured your Excel table (consistent headers, no merged cells), the better Copilot performs
• Save your QC template with Copilot-friendly formatting so you don't need to reformat each month
• Ask Copilot to "create a pivot table showing average CV% by analyte for the past 6 months" to spot gradual performance changes

Tool interfaces change, if the Copilot button has moved, look for the Copilot icon in the lower-right corner of the window, or the "Analyze Data" button in older Excel versions.