Finance

Building and scaffolding financial models with AI

Use AI assistants (e.g., Claude in Excel, Copilot, ChatGPT) to scaffold integrated financial models and build valuation components—including three-statement structures, DCF analysis, scenario projections, sensitivity tables, and driver-based forecasts—following investment-banking and FP&A conventions and dramatically reducing build time.

Why the human is still essential here

A human analyst must define the modeling intent, business assumptions, and structural standards, then rigorously audit formulas and outputs before any investment or business decision is made.

How people use this

Three-statement model scaffold in Excel

Prompt AI to create the full model shell (tabs, timeline, income statement, balance sheet, cash flow, and core linking logic) using investment-banking layout conventions.

Microsoft Copilot for Excel / Claude

FCF build, terminal value, and debt schedules

Use AI to draft the free-cash-flow waterfall, terminal value calculations (Gordon growth and/or exit multiple), and a debt schedule with amortization and revolver sweep logic feeding the statements.

Claude / ChatGPT

Scenario projections and sensitivity tables

Ask AI to populate multi-scenario revenue forecasts across five years and generate two-way sensitivity tables (e.g., WACC × terminal growth) with a valuation summary output page.

Microsoft Copilot for Excel / ChatGPT

DCF model with sensitivities and outputs via Agent Mode

Agent Mode builds a multi-year DCF (FCF, WACC, terminal value) and generates sensitivity tables/charts in a standardized, presentation-ready layout.

Microsoft Excel (Copilot Agent Mode) / Macabacus

Valuation model seeded with market data inputs

Agent Mode structures the valuation workbook while the team pulls comparable company multiples and market inputs into the model for faster iteration.

Microsoft Excel (Copilot Agent Mode) / S&P Capital IQ

3-statement model from driver assumptions via Agent Mode

Agent Mode creates an integrated P&L/BS/CF model with revenue and cost drivers, working capital schedules, and basic integrity checks for analyst review.

Microsoft Excel (Copilot Agent Mode)

Driver-based forecast model build-out

AI scaffolds a driver-based 3-statement model (revenue drivers, headcount plan, opex categories) with checks and reconciliation tabs for FP&A analyst review.

Microsoft Excel Copilot / ChatGPT

Scenario and sensitivity tables for runway and covenants

AI creates scenario toggles and sensitivity tables (price, volume, churn, FX) and summarizes the impact on runway and covenant headroom.

ChatGPT / Microsoft Excel

Model formula debugging and integrity checks

AI reviews a workbook for broken links, inconsistent assumptions, and circular references and suggests fixes with a change log.

Microsoft Excel Copilot / Claude

Community stories (3)

LinkedIn

Microsoft released Agent Mode in Excel.

Microsoft released Agent Mode in Excel.

Think Copilot, but instead of just suggesting, it acts inside your workbook.


It builds, edits, and reasons step by step.


I put together a full guide for CFOs, Finance, and FP&A teams on how to use this new AI capability for finance:


https://lnkd.in/eaFuD7vH


And a video guide: https://lnkd.in/eWjVxGrG


It also includes a step by step on how to "download" it.


This way you can see what has been tested and know if it can help you to automate reporting, forecasting, and even build valuation models!


If you want me to send the Excel file with the results just comment "AI Agent Mode" and I can send!


Here’s what Agent Mode can do live in your workbook:


✅ Automate budget roll-forwards across tabs

✅ Merge messy data into clean Pivot-ready tables

✅ Generate full variance analysis with commentary

✅ Build multi-tab models like DCFs & 3-statement forecasts

✅ Iterate until the output matches CFO-level standards


Also remember before submitting a prompt, you can select which model you prefer to use, Agent Mode in Excel supports the latest Anthropic and Open AI models available to you.


To switch between models, use the model picker dropdown.


And the best part? Everything stays editable.


Some sample prompts you can try right now:


"Build a 5-year DCF model with revenue, OPEX, EBITDA, FCF, and terminal value. Format outputs in CFO-ready style."


"Consolidate actuals from multiple sheets into one clean variance-to-budget report with conditional formatting."


"Generate a revenue forecast with 3 growth scenarios (Base, Optimistic, Downside) and plot a chart of outcomes."


"Draft a month-end financial summary with key drivers, risks, and opportunities based on this dataset."


"Reshape this dataset into a management-ready P&L with standard financial formatting and subtotals."


Hope this guide helps!


Some notes from Microsoft's announcement:


How to try it


1. Use https://excel.new to quickly create new Excel workbooks.


2. Select Home > Copilot and open Copilot chat.


3. Select the Tools menu and choose Agent Mode.


4. Start with an outcome-based simple prompt, like “Build a loan calculator that computes monthly payments based on user inputs for loan amount, annual interest rate, and term in years. Generate a schedule showing month, payment, principal, interest, and remaining balance. Present the results in a clear, formatted table.”


Tell me in the comments if you have questions or what was your result! :)


Notes from Microsoft's website on Availability


Agent Mode in Excel requires either a Microsoft 365 Personal or Family subscription (with an AI credits plan), a Microsoft 365 Premium subscription, a commercial Microsoft 365 Copilot subscription, or a Copilot Chat-eligible Microsoft 365 or Office 365 business or enterprise subscription.

CM
Christian MartinezFinance Senior Manager
Feb 26, 2026
LinkedIn

Question I get often: How can I use AI in my business?

Question I get often: How can I use AI in my business? Here's how.

I just built a full 5-year DCF valuation model inside Excel using Claude.

Revenue projections across 4 scenarios. Free cash flow waterfall. Terminal value. Sensitivity analysis.


This is not so much about the DCF. This is about what's now possible with AI and how it can help achieve business outcomes.


This is part of the series on:

How You Can Use AI in Your Business. Follow along if you want practical examples.

EO
Ev OputaFounder, Veloent
Mar 1, 2026
LinkedIn

Investment Banking: Ranking the Best AI Tools for Financial Modeling (2026) 🤯

Investment Banking: Ranking the Best AI Tools for Financial Modeling (2026) 🤯

We extensively tested OpenAI's ChatGPT, Anthropic's Claude, Microsoft Copilot Agent Mode, and Shortcut on a real three-statement Excel model using investment banking standards.


Here’s what actually works, and what doesn’t.


Quick Answer: The Rankings 👇


Best Overall: Shortcut


Close Second: Claude (Opus 4.6)


Third Place: Microsoft Copilot (GPT-5)


Distant Fourth: ChatGPT (GPT-5.2)


The Testing:


Criteria 1: How It Feels Working With The Tool


Understanding the Assignment:


Winners: Claude and Shortcut


Claude and Shortcut asked thoughtful clarifying questions after receiving the prompt, about:


• Forecast preferences

• Revenue segmentation

• Share repurchases

• Layout decisions

• Schedule structure


That behavior closely resembles what you’d want from a good junior analyst.


Copilot and ChatGPT asked none.


Speed:


Winner: Shortcut


Shortcut and Claude completed the setup in ~15 minutes vs ~25 minutes for Copilot.


ChatGPT took close to an hour.


A decent analyst would have taken 1-2 hours to complete the assignment.


So all were faster than a human analyst at initial setup.


Criteria 2: Data Extraction, Formatting and Best Practices


Formatting:


Winner: Shortcut


Shortcut and Claude produced the most “investment-bank-like” outputs.


Shortcut was more consistent with input coloring and structure.


Claude missed several formatting conventions.


Copilot ignored IB formatting entirely.


Accuracy:


Winner: Copilot


Copilot won, but this was disappointing across the board.


Shortcut and Claude hallucinated significant portions of historical data.


In both cases, the errors were subtle enough to be dangerous, with slightly incorrect line items all adding up to correct subtotals.


Shortcut’s second attempt returned almost no mistakes.


Claude continued to generate bad data.


Fixing this would require careful cell-by-cell auditing that takes longer than just inputting the numbers yourself.


As a rule, analysts should not rely on these agents to find data and should instead upload PDFs and spreadsheets for the agents to work with.


Copilot and ChatGPT were more accurate.


ChatGPT’s presentation was the least polished, but its historical balance sheet was easiest to audit.


Had Shortcut and Claude used correct data, they would have won as they were attempting a more analytically rigorous presentation.


Shortcut was also going into the footnotes to break out certain items when appropriate.


Sourcing and Commenting:


Winner: Claude


Claude provided the best explanations of where data came from and why certain modeling decisions were made.


Copilot added no comments.


ChatGPT added too many.


Shortcut did some, but less consistently.


Claude was also the only tool to backsolve EBITDA correctly.


The Bottom Line:


Shortcut and Claude significantly outperform Copilot and ChatGPT.


But right now, even the best tool still underperforms a Junior Analyst.

MF
Matan FeldmanCEO & Founder, Wall Street Prep
Feb 23, 2026