Analyst is the quantitative sibling of Researcher. Where Researcher synthesizes documents and the web, Analyst crunches numbers — your Excel files, your CRM data, your dashboards. Statistics, charts, trend analysis, comparisons. It's "help me understand this data" for everyone who isn't a data analyst — and the skill is reading what it gives back with a critical eye.
Step 1 of 50% complete
The mental model
Insight, not just a chart
Hand Analyst a spreadsheet and it does the work you'd otherwise do with PivotTables and a stats refresher: win rates, correlations, outliers, significance — explained in business English, with charts. Researcher reads documents and the web; Analyst reads your numbers.
It's fast and genuinely capable. It's also confident — which is the catch.
Predict first
You hand Analyst a messy spreadsheet with a column simply labeled "value." What's the realistic risk?
Three ways to use it
What Analyst is great at
① Excel file → clear insight
Hand it a sales sheet; get top/bottom reps, win rate by region, outliers, and 3 exec-ready charts with a one-page summary in business English.
② Cross-source comparison
Point it at five quarterly files at once for a year-over-year trend per channel, in one combined chart — ending with "three things to ask the CMO."
③ Real statistics, explained
Significance tests, correlation, regression with controls — and a plain-English summary of what test it used and why.
Stats with explanationDoes our high-touch onboarding actually reduce churn? Control for company size, industry, and contract value. Tell me if the effect is statistically significant, the magnitude, and the confidence intervals — and explain the test you used in plain English for a non-statistical leader.
Analyst can run correct math on the wrong test, or misread an ambiguous column with total confidence. The output is a starting point you check, not a verdict you forward.
Do it · read the chart
Turn the chart into a decision
Analyst built this from your sales file and dropped it in a deck. A chart on its own isn't insight — the value is the takeaway you'd actually say in the meeting. Which one is it?
Win rate by region · Q1
closed / (closed + lost)
Do it · go or verify?
Self-serve, or check it first?
Analyst is for knowing when self-service works — not for replacing every analyst. For each task: hand it to Analyst and go, or verify / escalate first?
Call 1 of 6 · 0 right
📊
Lesson complete
You've got a data team in a prompt — and the judgment to use it
What you can do now
Use Analyst when you have data and need insight, not just a chart
Confirm it understood your columns before trusting the analysis
Combine multiple files for cross-period or cross-team comparisons
Ask Analyst to explain the test it chose for statistical work
Escalate high-stakes or messy-data analyses to a real analyst
Your move: replace one "quick analysis" request
Think of the last time you asked your data team for a quick analysis. Take that same data and request, frame it as an Analyst prompt, and compare the result. Then sanity-check the columns and the takeaway before you act on it.
Agents · Lesson 03 — Facilitator: meeting management in Teams
Facilitator runs alongside you in Teams meetings — taking notes, capturing decisions, moderating turns. The agent that means you never have to choose between participating and documenting.
Hey! I'm your AI Coach for this lesson on Copilot Analyst. Ask me how to frame a data prompt, how to sanity-check what it gives back, or when to loop in a human analyst. What data are you working with?
Free lesson coaching is limited to 3 questions. Upgrade to Pro for unlimited coaching on every lesson.