LLMs are excellent at interpreting text, writing, summarizing, and generating content. But when it comes to calculation, statistics, and data analysis, “almost right” doesn’t cut it. For financial reports, sales spreadsheets, and business decisions, you need absolute precision.
That’s exactly what Deep Analysis is for: a tool that combines LLM natural language conversation with the accuracy of traditional computing.
What is Deep Analysis?
Deep Analysis is a safe, isolated environment (a “sandbox”) where, instead of the AI trying to “do the math in its head”, it:
understands what you want to do with the data
writes Python code to carry out that task
runs that code in the sandbox
returns the result with 100% correct mathematical calculations
In other words: you chat in Portuguese; Tess translates that into code, runs it, and gives you the finished result (tables, metrics, charts, segmentations, etc.).
How to use Deep Analysis in chat
Whenever you need some quantitative analysis, or quali/quanti, a report, HTML, or similar processing, turn on the Deep Analysis tool in the chat!
If you already have a base document, remember to upload your file, make your request in natural language, and mention the file and what needs to be done with it.
Use it whenever data accuracy is the top priority, for example:
Financial analyses
Calculate EBITDA, profit margin, average ticket
Compare periods (month over month, year over year)
Project scenarios (simulations and projections based on historical data)
Data visualization
Create bar, line, pie, scatter charts etc.
Visualize trends in sales, churn, engagement, costs
Sales and customer analysis
Identify best-selling products
Segment customers by value range or purchase frequency
Evaluate campaign or channel performance
Engineering, science and experiments
Run complex formulas
Process experiment data
Do statistical analyses (means, medians, standard deviation, correlations)
Common supported formats
Spreadsheets (XLSX)
CSV files
Other structured formats that can be read via Python (when applicable)
Prompt examples:
“Analyze this file vendas_trimestre.xlsx, calculate the total sales for each product category, and create a pie chart with each category’s share.”
“In this customer CSV, calculate the average ticket by region and show it in a table sorted from highest to lowest.”
“Generate a line chart showing the monthly revenue trend over the last 12 months.”
In this process, Tess will write the code, run it in the sandbox, and return the results (tables, explanations and, when requested, charts generated from the data).
Deep Analysis is the bridge between natural language conversation and the rigor of data science. It makes sure that the reports, analyses, and charts generated by Tess AI are not just smart — but mathematically correct.
