Analysis Recommendation
Upload your data, describe your goal in plain language, and the AI recommends the right methods from 60+ — then points you to the page where you can run them. No stats background required.
Overview
Reads your data
The moment you upload, the AI scans column types, sample rows, and variable roles — no setup needed.
Recommends, never guesses
It matches 60+ methods to your variable types, sample size, and question — so you don’t run the wrong test.
Points you to the right page
Each recommendation names the method and where to run it — you head to that page (Statistical Lab, Model Lab, …) to do the actual analysis.
What you need
What you can do
How it works
No stats background needed
Just describe what you want to find out in plain language. The AI matches your goal to the right methods from 60+ across the platform — so you don't have to know which test is correct. It recommends only; to run an analysis you open that method's own page (Statistical Lab, Model Lab, …).
Upload your data
Drop a CSV, Excel, or TXT file — it is processed locally in your browser. No file handy? Start from a built-in example like Iris or mtcars. The AI reviews the dataset right away.
Review your dataset and pick a direction
The Recommendation Chat opens with a dataset overview (row count, variables, types) and a variables panel you can click to ask about any column. Tap a suggested question, choose a goal (Basic Statistics, Relationships, Difference Testing, Predictive Modeling, Clustering, …), or just type your question.
Get methods recommended in chat
Ask in plain language and the AI replies in the chat with methods drawn strictly from this platform’s catalog — grouped by family (Descriptive & Exploratory, Comparison Tests, …), each with a one-line reason and the variables that fit. Everything it suggests is something you can actually run here.
