CSV & Excel
Ask questions about a CSV or Excel file
No database, no connection, no setup. Drag in a spreadsheet and start asking: ChatPlotDB works out the structure and answers with a table or a chart.
Most questions start life as a file somebody exported. ChatPlotDB queries CSV and Excel files directly: upload one, ask in plain English, and filter, aggregate, join and chart it conversationally, no formulas, no pivot tables, and no waiting for it to be loaded into a warehouse first.
What you'd otherwisebuild a pivot table for
The analysis that costs an afternoon of spreadsheet work, asked as a sentence instead.
“Which five regions had the highest total sales in this export?”
“What's the month-over-month trend, as a line chart?”
“Show me every row where the amount is above 10,000 and the status is open”
“Group by category and give me the average, median and count”
“Which entries appear in this file but not in the one I uploaded before?”
“Are there duplicate records in here, and how many?”
From uploadto answer in seconds
The point of a file is that it's already in your hands. Nothing about analysing it should require a project.
No connection to configure
There's no host, no port, no credential and no SSL mode. You drag the file in and ask your first question straight away.
Nothing to protect, nothing to configure
There's no production database on the other side of an upload, so none of the care a live connection needs applies. You ask, and the answer comes back.
Structure worked out for you
ChatPlotDB infers what the columns hold and how they relate, so you don't declare types or clean headers before you can ask anything.
Charts, not just numbers
Results come back as an interactive table or a Plotly chart, and you can keep refining the same question until the shape is right.
Excel as well as CSV
Both formats are queried the same way, so the file your colleague sent from Excel needs no conversion first.
Worth keeping, keep it
An answer you'll want again can be saved to a dashboard as a card rather than re-derived from the file next month.
The data that never made itinto the warehouse
Plenty of real analysis happens on files that will never be a table in production, and that's fine.
A one-off export from a vendor. A finance file that arrives monthly by email. The CSV a colleague pulled out of a system you don't have access to. None of it justifies a pipeline, so it usually gets analysed by hand in a spreadsheet, which is slow and quietly error-prone in ways nobody catches.
ChatPlotDB treats a file as a queryable source. You upload it, it infers the structure, and you ask the same kinds of questions you'd ask a database: filter, aggregate, group, compare, chart. Analysis that meant an afternoon of pivot tables and a formula you'd have to debug becomes one sentence and a follow-up.
Because there is no database behind the file, the flow is shorter than the database one: no connection to register, no credentials to store, and nothing to grant access to. Ask, and the answer comes back. When the same question turns out to matter every month, save it to a dashboard so it stops being a one-off.
Common questions
No. CSV and Excel files are uploaded and queried directly, with no connection to configure. If you never connect a database, the product still works: the file is the data source.
CSV and Excel. Both are queried the same way, so a spreadsheet a colleague sent needs no conversion before you can ask about it.
No. An upload is self-contained: there's no database behind it, no credential stored, and nothing granted access to your systems. It's the file you dropped in and the questions you ask about it.
No. ChatPlotDB infers the structure of what you upload, so you can ask your first question against the export exactly as it arrived.
Yes. Results come back as an interactive table or a Plotly chart, and you can ask for a different shape in a follow-up rather than starting over.
Any answer worth keeping can be saved to a dashboard as a card.
Connect it instead. The same conversation works against PostgreSQL, MySQL and Amazon DynamoDB, so the way you ask doesn't change when the data moves somewhere more permanent.
