Skip to content

How to track expenses from a bank CSV

Export your statement as CSV, upload it in the web app, then ask where the money went: totals by category, a chart, a weekly report you re-run.

Most banks already give you the data; what is missing is a quick way to ask it questions. A bank export becomes a table you can question in plain language, and a report you build once keeps working as new months arrive. The same table is where the personal finance use case starts.

  1. Export the statement

    In online banking, look for transactions or statements and export CSV. Keep the header row; choose UTF-8 if your bank asks.

  2. Upload it in the web app

    Attach the file to a message: “bank export, table expenses”. The reply names the table, the row count and the columns.

    Ask your spreadsheet a question
    1. expenses-2026.csv

      Here's my bank export

    2. Imported 312 rows into the table expenses — 4 columns, dates from January to August.

    3. How much did I spend on restaurants in August?

    4. €412.30 across 17 visits. The top three:

      Restaurants, August 2026

      merchantvisitstotal
      Trattoria Nonna5€148.00
      Bao House4€96.50
      Café Lumen3€61.20

      14 more rows

  3. Ask where the money went

    “How much did I spend on restaurants in August?”, “top five merchants this quarter”. The answer comes as a number, a table or a chart.

    A chart from a sentence

    1. Chart my spending per merchant this quarter, top five

    2. Top five merchants, July to September.

      Spend per merchant, Q3 2026

      Rewe
      €612
      Deutsche Bahn
      €388
      Amazon
      €297
      Trattoria Nonna
      €214
      Spotify
      €36

  4. Make it a weekly report

    “Make me a weekly spend report by category, with a chart.” The report re-runs on your table, in your browser, every time you run it.

    A report the AI writes and your browser runs

    1. Make me a weekly spend report by category, with a chart

    2. Created “Weekly spend”. It re-runs on your data every time you run it — nothing leaves your browser.

      Weekly spend

      Ran in your browser

      This week

      €386

      −12% vs last week

      Groceries
      €142
      Restaurants
      €96
      Transport
      €58
      Subscriptions
      €47
      Other
      €43

  5. Keep it current

    Upload next month’s export to the same table and the rows are added. A single expense can be typed in any chat: “add: 12 Sep, Rewe, 43.20, groceries”.

Questions

Can I send the CSV in Telegram?Soon

Not yet — upload it in the web app. Questions and typed rows work in Telegram and WhatsApp.

Which files work?

CSV (UTF-8) or JSON with column names in the first row, up to 50 MB per file.

Can I compare months?

Yes — “spending per month this year” or “August against July” gives a table or a chart.

Related

Start with one message

Open the bot, say what's on your mind, and see it become a task.

Join in TelegramJoin with email

Closed beta: we let people in in batches. Sign up in Telegram or with your email, and we'll let you know when your access opens.