Tabloy

Appearance

Open Tabloy

Upload a file ​

Excel, CSV and JSON files become datasets. Tabloy studies the table once, and then answers with exact numbers over every row.

Upload in a chat ​

  1. Drop the file into the chat, or press + and pick it.
  2. Tabloy asks how should this file be used? Choose Upload as dataset.
  3. The chip appears above the message box and shows the progress: Building Context.
  4. When it is ready, ask.

Building the Context takes from a few seconds to a minute or two for a large file. Tabloy reads the columns, works out their types and what each field means, so it can answer without you explaining the table.

Upload from Datasets ​

Open Datasets in the sidebar and press Upload. The file is prepared the same way. Click it in the list to start a chat with it attached.

Formats ​

  • Excel: .xlsx and .xls. Every sheet is read. Formula cells are read by their computed values.
  • CSV: comma, semicolon or tab separated. The first row should hold the column names.
  • JSON: an array of objects, one object per row.

What makes a file easy to read ​

You do not need a perfect file. These help:

  • One header row at the top, then the data. No title rows or merged cells above it.
  • Clear column names: order_date, region, revenue, not col1.
  • Dates that look like dates, numbers that look like numbers. A column with "$1,200" and "n/a" mixed in is harder to sum.
  • One topic per file. Sales and hiring in one sheet make every question ambiguous.

If an answer is off, check the column Tabloy used (it is named in the answer and in the chart), fix the file and upload it again.

Where files live ​

Datasets lists every dataset you uploaded, with its size and when it was added. Delete what you no longer need. Uploaded files count towards the storage of your plan: 5 GB on Trial, 50 GB on Pro, 150 GB on Pro+.

Update a file ​

Upload the new version as a new dataset and attach it to the chat; the old one stays until you delete it. For data that changes often, connect it as a source instead.

Try these requests

What are the columns and what does each one mean?

Open in Tabloy

Total revenue by month, as a bar chart

Open in Tabloy

Are there duplicate rows or missing dates?

Open in Tabloy