Last week I explored the Boston Crime Statistics dataset (~260,000 rows) using Excel Agent Mode , which lives within Excel. This week I revisited the same dataset with the same question using Microsoft’s M365 Analyst Agent - it is a completely different experience. Both tools analyze data and generate insights, but they differ in how you interact with them and how they talk , and show their work. One keeps you grounded in the familiar grid of Excel; the other lifts you into a conversational workspace (real Conversational Data Analytics) that feels more like working with a colleague than a formula bar.
What are they?
Excel Agent Mode (part of the Frontier experience and Excel Labs Frontier) sits directly inside Excel. You type natural-language prompts into a sidebar, and it runs analysis right inside your workbook - creating tables, formulas, charts, and summaries you can audit. It shows every step of the process. M365 Analyst Agent , on the other hand, is more of an AI analyst integrated into the Microsoft 365 ecosystem. Instead of working cell-by-cell, you hold a natural conversation: “Please summarize these statistics and help me understand and predict future occurrences.” or “Based on this data what recommendations do you have?” - the same prompts I used with Excel Agent Mode. The Analyst agent retrieves, analyzes, and explains results, then produces an analysis, visualizations and recommendations you can refine through follow-up questions. In short (the TL;DR):
- Excel Agent Mode = embedded, visible, auditable.
- M365 Analyst Agent = conversational, cross-app, contextual.
What does it mean from a business perspective?
It’s no surprise that given the interface and way of working with the two approaches that it’s a contrast.
- Transparency vs. Abstraction : Excel Agent Mode gives you visibility into formulas, tables, and logic. You can audit the results. The M365 Analyst Agent abstracts those details into a dialogue, prioritizing speed and interpretation over the mechanics. Both are powerful - but likely suitable for different compliance cultures.
- Hands-On vs. Conversational Workflows : The Excel Agent rewards people who like to see how the analysis happens. The Analyst Agent suits leaders who want to talk through insights and get executive-level summaries fast.
- Governance and Auditability : Because the Excel Agent logs its operations directly in cells, it provides traceability. The Analyst Agent’s conversational mode may need extra documentation for decision trails - e.g. copy/paste the output from the analysis.
- Adoption Path : Excel Agent is a natural first step for business analysts and power users. The M365 Analyst Agent seems to represent the next stage - conversational data analytics.
What do I do with it?
There are a few concrete steps to take - including building confidence in the output.
- Pilot Both, Side-by-Side : Run a comparative test on a dataset your team knows. You could even time how long each takes to generate accurate insights, and compare the audit trail, user experience, and business value.
- Choose by Use Case : Use Excel Agent for structured analysis, forecasting, or QA where traceability matters. Use M365 Analyst Agent for briefings, scenario discussions, or quick synthesis across files.
- Respect Data Provenance : Treat the Analyst Agent as an “insight generator,” not a data warehouse.
- Train Teams Differently : Analysts need to learn prompt-plus-formula hybrid skills for the Excel Agent; managers benefit from prompt-plus-interpretation coaching for Analyst Agent.
- Track Efficiency & Confidence: Measure time saved, report accuracy, and user trust. The best tool is the one your people actually use, and can defend it’s results and analysis. Both of these tools reveal a direction that we are moving in - where we have conversations with our data. When used together, they represent the next evolution in business analytics - both visible where it matters and conversational when needed.