Your session has been flagged for unusual activity.
Please confirm you're human to continue.
Upload XLSX, XLS, CSV, or TSV files. Tell the AI what you need, let it write Python/pandas logic, then download a real spreadsheet built from your file.
Use ChatGPT, Claude, Gemini, or Grok for the reasoning, while Python handles the spreadsheet execution.
Account statement 20250906141923.csv
57.9 KB · 321 rows · 11 columns
321
Rows
11
Cols
9
Dupes
1,461
Missing
AI wrote the code
Python ran it on your data
Your data stays private
Excel Studio is built for people who need usable spreadsheet work, not just chat answers about spreadsheets.
Start creating spreadsheetsAI processing
Upload XLSX, XLS, CSV, or TSV. AI receives metadata and 5 sample rows; Python applies the generated logic to the file.
Instant tools
Cleaning, formulas, formatting, reshaping, validation, and protection tools run without spending AI tokens.
Templates
Budgets, P&L, cash flow, invoice, sales pipeline, employee roster, inventory, OKRs, expense reports, and custom workbooks.
Formula assistant
Ask for the calculation you need and get a formula with explanation, example references, edge cases, and alternatives.
Project files
The app can list supported spreadsheet files from your projects so you can work from existing context instead of re-uploading every time.
Guardrails
Profiling and instant tools currently use a 10 MB sync guardrail. XLSX workbooks also pass a 500 MB uncompressed safety check.
A unique two-step architecture: the AI writes transformation logic, Python executes it on your actual file. No hallucinated cells. No guesswork.
Drop XLSX, XLS, CSV, or TSV files. AI processing accepts uploads up to 200 MB. Profiling and instant tools currently use a 10 MB sync guardrail and show row counts, column types, missing values, duplicates, and numeric ranges.
Write what you want in plain English. The AI receives column metadata, selected columns, data types, and 5 sample rows instead of the full dataset. It writes Python/pandas code to accomplish the task.
Python runs the code in a guarded sandbox on your uploaded file. Preview the result inline, iterate with follow-up instructions, then download XLSX, with CSV available for table outputs.
From routine cleanup to complex multi-step transformations — describe the outcome, and the AI figures out how to get there. Here's a taste of what's possible.
Add a column that categorises each row from context — transaction types, customer segments, product categories, sentiment. The AI reads meaning, not just keywords.
Let the AI scan for unusual patterns — statistical outliers, values outside expected ranges, duplicates that shouldn't exist, dates that don't make sense in context.
Pivot, unpivot, transpose, merge sheets, split data by a column into separate tabs — complex restructuring that would usually require careful manual work in Excel.
Add derived columns from complex logic — profit margins, running totals, YoY growth, weighted scores, or any multi-condition formula the AI translates from your plain-language description.
Translate headers, labels, and cell content into many languages. Reformat dates, numbers, and currencies to match regional standards. One prompt, fully localised output.
Go beyond trimming whitespace. The AI can normalise inconsistent company names, merge near-duplicate rows, infer missing values from context, and standardise messy free-text fields at scale.
Ask for a bar chart comparing monthly revenue, a heat map across a matrix, or a dashboard summary sheet — the AI builds it and embeds everything directly in the file.
One instruction, multiple operations. The AI understands intent and chains together everything needed — filter, calculate, format, sort, summarise — in a single pass over your data.
For deterministic tasks such as deduplication, date formatting, column merging, pivots, and validation rules, Python executes directly without an AI roundtrip. Pattern-based operations like filtering rows by number or sorting can be detected and run without model tokens.
27
precision tools
10 MB
tools/profile guardrail
0
AI tokens for tools
5
tool groups
Cleaning
Trim Whitespace
Strip leading/trailing spaces from all text cells
Remove Duplicates
Delete duplicate rows with smart column selection
Fix Case
UPPERCASE / lowercase / Title Case conversion
Remove Empty
Drop rows or columns that are entirely blank
Fill Blanks
Forward-fill or back-fill empty cells
Fix Dates
Standardize date columns to a consistent format
Split Column
Split by comma, semicolon, pipe, or custom delimiter
Merge Columns
Join multiple columns into one with a separator
Formulas
Summary Row
Add SUM, AVERAGE, COUNT at the bottom
Running Total
Cumulative sum column for any numeric field
% of Total
Each row as a percentage of the column total
Growth %
Period-over-period percentage change
Rank
Rank rows by value, highest or lowest first
Formatting
Auto Table
Convert to formatted Excel Table with filters
Color Scale
Heat map conditional formatting (red–yellow–green)
Data Bars
In-cell horizontal bar charts for numeric columns
Threshold Highlight
Highlight cells above, below, or between selected values
Fix Numbers
Auto-apply currency, percentage, date formats
Auto-Width
Fit column widths to content automatically
Freeze Panes
Lock header row while scrolling
Reshape
Pivot Summary
Group rows and aggregate values
Transpose
Flip rows and columns in one click
Split to Sheets
One sheet per unique value in a column
Validation & Protection
Dropdown Validation
Add controlled dropdown lists to selected cells
Range Validation
Restrict numeric input to safe minimum and maximum values
Protect Formula Cells
Lock formulas while keeping editable areas open
Named Ranges
Create named ranges for cleaner formulas and templates
Describe the spreadsheet you need, or pick a template. AI plans the full structure, Python builds the workbook structure.
Monthly income/expense categories with variance analysis
Revenue, COGS, expenses and net income with monthly columns
Operating, investing, and financing with runway calculation
Payment schedule with principal, interest, and balance
Line items with quantity, unit price, tax, and total
Deals with stages, probability, and weighted forecast
Staff directory with departments, salary, and tenure
SKUs with reorder points, unit cost, and total value
Tasks, dates, dependencies, and status tracking
Objectives, key results, targets, and progress
Options vs weighted criteria with scores and ranking
AI decides the best structure based on your description
What spreadsheet do you need?
e.g. Monthly budget for a 50-person startup, Sales pipeline tracker for a SaaS team...
Tell the AI what you're trying to calculate. It returns an Excel formula — with a step-by-step explanation, example cell references, edge case notes, and alternative approaches. For any level of complexity.
"Weighted average of column B using weights in column C"
=SUMPRODUCT(B2:B100,C2:C100)/SUM(C2:C100)
+ explanation, edge cases, and 2 alternatives
Any complexity
from IF to LAMBDA
Step-by-step
explanation included
Alternatives
multiple approaches
Describe the formula you need
Column headers & sample data (optional — improves accuracy)
Switch between models on any task. Pick a model per spreadsheet task and compare which reasoning style fits your file.
OpenAI
Strong at structured data tasks, code generation, and following precise step-by-step instructions.
Anthropic
Exceptional at nuanced instructions and complex, multi-condition data transformations.
Fast processing with excellent multilingual support — ideal for translation and localisation tasks.
xAI
Creative approaches to analysis — useful for unconventional transformations and exploratory data work.
Excel Studio is strongest when it becomes the start of a real business deliverable: clean the numbers, write the document, then present the result.
Upload XLSX, XLS, CSV, or TSV files. Ask AI to analyze, clean, format, validate, or transform the data with Python-backed execution.
Use Excel Studio →Turn cleaned data, notes, and source material into reports, proposals, summaries, briefs, whitepapers, DOCX, PDF, Markdown, or HTML.
Open Document Studio →Use your report, research, or spreadsheet findings as source material for an editable PPTX deck with layouts and speaker notes.
Create a presentation →Why this matters: many users do not want another chatbot tab. They want a finished file: a cleaned spreadsheet, a written business document, or a deck they can send to a client.
AI processing sends column metadata, selected columns, data types, and 5 sample rows so the model can write transformation code. Python applies the code to your uploaded file in a guarded execution flow.
Blocks unsafe imports, eval/exec, file/network access, and OS-level operations
Files stored securely with automatic 7-day TTL deletion
Rate limiting and CSRF protection on all endpoints
AI sees metadata and 5 sample rows, not the full spreadsheet
The practical details people usually need before they trust an AI spreadsheet workflow.
Excel Studio does not depend on a model rewriting a pasted spreadsheet. For AI processing, the model writes Python/pandas logic from file metadata and sample rows, then Python applies that logic to the uploaded file.
For AI processing, the model receives column metadata, selected columns, data types, and 5 sample rows. The full file is processed by Python rather than pasted into the AI prompt.
Inputs include XLSX, XLS, CSV, and TSV. CSV loading can handle comma, semicolon, and tab-separated files with common encodings. Processed outputs use an XLSX download flow, with CSV available for table outputs when produced.
Yes. Excel Studio includes 13 generation templates, including budget tracker, profit and loss, cash flow forecast, invoice, sales pipeline, inventory tracker, project tracker, KPI scorecard, comparison matrix, expense report, and custom workbooks.
It can help with cleaning, formulas, formatting, reshaping, validation, dashboards, summaries, and plain-language analysis. The Python toolset covers common spreadsheet tasks such as duplicates, dates, splitting columns, pivot summaries, dropdown validation, freeze panes, and protected formula cells.
Upload a file or start from a template. Ask for cleanup, formulas, dashboards, translations, validation rules, or business-ready Excel output.