Analyst, the Built-In Agent

1Overview

Where to click

1. Copilot
2. Agents
3. Analyst
4. +
5. file 06

2Watch the demo video

3Watch where to click Real screenshots (consumer Copilot) + illustrations

The demo shows where to click, step by step. Steps marked Real screen are screenshots from Copilot (consumer version); your work Microsoft 365 Copilot may look slightly different. The other steps are drawings. The list under the player is the same steps as text.

Microsoft 365 Copilot (copilot.cloud.microsoft)Illustration - not the actual Copilot screen

Copilot Chat

Analyst - reasoning steps Show code ▾

• ✓ Reading 06_Service_Request_Data.xlsx (sheet Monthly Data)
• ✓ Writing Python to calculate the measures
• ✓ Running the code
• ✓ Checking for unusual values

import pandas as pd
df = pd.read_excel("06_Service_Request_Data.xlsx", header=3)
df["Completion Rate"] = df["Requests Completed"] / df["Requests Received"] * 100
df["FCR Share"] = df["Resolved on First Contact"] / df["Requests Completed"] * 100
df[["Month","Completion Rate","FCR Share","Avg Days to Resolve"]].round(1)

Copilot (illustrative response)

Month	Received	Completion %	FCR share %	Avg days
Apr	460	90.9	37.1	7.4
May	472	91.1	34.7	8.0
Jun	390	116.7	37.4	8.4
Jul	430	97.2	38.8	7.6
Aug	455	96.7	blank	7.2

Unusual: June completion is 116.7% (more completed than received). August first-contact data are blank, not zero.

Your hand check for June: 455 / 390 x 100 = 116.7 ✓ match

X06_Service_Request_Data.xlsx

+Message Copilot➤

⇪ Upload from this device

☁ Add from OneDrive (or type /)

Open a file (illustration)

X06_Service_Request_Data.xlsx

CancelOpen

📷 Screenshot from Copilot (consumer version). Your work Microsoft 365 Copilot may look slightly different.

Press Play to watch where to click.

▶ Play⟳ Replay◀ PrevNext ▶Speed SlowNormalFast

1. Real screenAnalyst appears under Pinned here (deck path: Agents > Analyst)
2. Real screenThe Analyst page opens
3. Real screenStarter cards like Analyze data are optional - you will paste the class prompt instead
4. Real screenSelect + to attach file 06
5. Real screenChoose Upload images and files (in work Copilot this may read Upload from this device; Attach cloud files = OneDrive)
6. Check 06_Service_Request_Data.xlsx
7. Select Open - the files attach above the prompt box
8. Paste the prompt from the Prompts section
9. Send - Analyst plans and writes code
10. Expand the reasoning steps to see the code
11. Read the results table
12. Optional: send the chart follow-up below, then look at the trend it charted
13. Check what it flagged as unusual - and that it did not invent causes
14. Compare one number with your own hand check

Annotated screenshots (consumer Copilot, opens the image): chat home empty analyst pinned plus button menu

4Do it: step by step

Check off each step as you go. Your progress saves in this browser.

1. Get the practice filesFiles already in your OneDrive (Step 0)? Pick them there when you attach; nothing to download. Otherwise: Path A: download file 06 (06_Service_Request_Data.xlsx) from the Files step. Path B: nothing to download. The Files step has the table as text.
2. Watch the demo videoSilent recording with step-by-step captions. This step checks itself off when you finish the video.
3. Watch where to clickPlay the demo, then open Copilot → Agents → Analyst → + → file 06.
4. Run the promptselect Copy prompt (it is also in the class chat), paste it into Copilot, and select Send.
5. Check one number by handJune: 455 completed divided by 390 received, times 100, is 116.7%. Does Analyst's June completion rate match? If you did Lab 6, compare with that hand check too.
6. Check the resultDo the numbers match your hand check?
7. Use the Five ChecksSource, Verification, Context, Resilience, Authorized Owner. Ask each one before you use the answer. The Toolkit page lists the five questions.
8. Pass the quizAnswer every quiz question correctly. This step checks itself off.

5Prompts: weak vs strong

Follow the brackets. The detailed prompt is built from the four C.H.A.T. parts: [Context], [How come], [Assign task], and [Type of output]. Keep the labels when you paste it; Copilot reads them fine, and they help you check that each part is there.

Compare a vague prompt with a detailed, source-checkable prompt. The activity below separates the request from the Guardrails (limits you add, such as "cite the source" or "do not decide for a person") you need before use.

Analyze this data.

[Context] I shared file 06, the monthly service request data (requests received, completed, and days to resolve by month).
[How come] I want to check its work against one number I calculate myself before I trust the rest.
[Assign task] Using the file 06 data I shared, analyze monthly changes in requests received, average days to resolve, completion rate, and first-contact resolution share of completed requests.
[Type of output] Identify trends and unusual values. Do not infer causes the data does not support.

[Context] Same Analyst chat, with file 06 and the Analyst results.
[How come] A chart makes the trend easier to see and to check against the numbers.
[Assign task] Make one simple line chart of Avg Days to Resolve by month from file 06.
[Type of output] One line chart, with one point for each month.

6Quiz

7Path B Text Pack and downloads

Download the practice files (Path A)

Path A (full access): these downloads are for you. Save them where Copilot can reach them (OneDrive for the Office app labs). Path B: nothing to download; the files are below as text, ready to paste. Path C: nothing to download; read the text below when you check sources.

Path B (Chat only): short steps

[Context] I shared file 06, the monthly service request data (requests received, completed, and days to resolve by month).
[How come] I want to check its work against one number I calculate myself before I trust the rest.
[Assign task] Using the file 06 data I shared, analyze monthly changes in requests received, average days to resolve, completion rate, and first-contact resolution share of completed requests.
[Type of output] Identify trends and unusual values. Do not infer causes the data does not support.

[Context] Same chat as the Analyst prompt, with the table I pasted from file 06.
[How come] A chart makes the trend easier to see and to check against the numbers.
[Assign task] Make one simple line chart of Avg Days to Resolve by month from the table I pasted.
[Type of output] One line chart, with one point for each month. If you cannot make a chart, give a month-by-month table instead.

FICTIONAL TRAINING FILE - The City of Fernhollow, its staff, residents and data are invented for the CPS HR Microsoft 365 Copilot workshop. Do not use real work data.

FILE 06 - SERVICE REQUEST DATA (text version of the spreadsheet)
Sheet: Monthly Data (Excel table name: ServiceRequests)
Blank = no data; 0 = measured and zero. Ending Backlog = Beginning Backlog + Received - Completed.

Month | Beginning Backlog | Requests Received | Requests Completed | Resolved on First Contact | Avg Days to Resolve | Escalated | Ending Backlog | Notes
Jan 2026 | 120 | 410 | 395 | 150 | 6.1 | 12 | 135 | 
Feb 2026 | 135 | 398 | 380 | 141 | 6.3 | 9 | 153 | 
Mar 2026 | 153 | 445 | 412 | 160 | 6.8 | 0 | 186 | No escalations recorded (confirmed zero)
Apr 2026 | 186 | 460 | 418 | 155 | 7.4 | 14 | 228 | 
May 2026 | 228 | 472 | 430 | 149 | 8.0 | 18 | 270 | 
Jun 2026 | 270 | 390 | 455 | 170 | 8.4 | 21 | 205 | Focused effort to clear older requests
Jul 2026 | 205 | 430 | 418 | 162 | 7.6 | 15 | 217 | 
Aug 2026 | 217 | 455 | 440 |  | 7.2 |  | 232 | Case-system migration: first-contact and escalation data not captured (blank = no data)
Sep 2026 | 232 | 468 | 452 | 181 | 6.9 | 13 | 248 | 

Lab 6 measures:
Completion Rate = Requests Completed / Requests Received x 100
First-Contact Resolution Share of Completed Requests = Resolved on First Contact / Requests Completed x 100