Use case · Analysts & Ops

The CSV that was supposed to take an hour.

You exported the data. The deadline is 4pm. Somewhere in those fifty thousand rows is the answer. Drop the file. Let AI pick the charts. Export the deck. Go home.

Upload your first file See pricing
Back to home

Spreadsheets are where analysis goes to die.

Excel opens the file. Three pivot tables later, you realise you need a different aggregation. Tableau wants a contract and a training day. The AI chat tools want you to upload customer records to their servers. Meanwhile, it's 3:47pm.

The usual Monday

  • Download the CSV from the data team
  • Open in Excel. Watch it choke on 500k rows
  • Three pivots later, realise you need a join
  • Copy-paste into Tableau Desktop (if you have a seat)
  • Screenshot the charts for slides at 11pm

With DuckViz

  • Drop the file in the browser
  • AI classifies the domain and picks charts
  • DuckDB runs the joins and window functions in-tab
  • Drag, drop, resize — dashboard in under a minute
  • One click: AI writes the report or the deck

How AI picks the right chart, every time

DuckViz classifies your data across 51 pre-trained domains — sales funnels, NPS surveys, finance ledgers, HR records, infrastructure costs, security events. Then it picks visualizations from a registry of 50+ chart types, tailored to the domain and the shape of the columns.

  • Domain-aware — AI recognises whether a column is a currency, a country code, a timestamp, or a sales-stage label, and picks accordingly.
  • 50+ chart types — bar, line, heatmap, Sankey, chord, treemap, word cloud, geo-map, sunburst. You are never stuck with "just bar or just line."
  • Ask in plain English — "show me revenue by region excluding returns" — AI writes the DuckDB SQL, runs it in-browser, and wires the result to a chart.
  • Multi-file joins — upload a folder. Each file becomes a DuckDB table. Cross-file SQL works out of the box.

Three minutes, start to finish

  1. Drop the file

    CSV, Excel (.xlsx), JSON, XML, or an entire folder. Auto- detection handles every format — no column mapping step.

  2. Let AI build the first dashboard

    In the time it takes to open Excel, you get 5–7 recommended charts covering the key dimensions of your data.

  3. Customize

    Drag, resize, swap chart types, add KPIs. Ask follow-up questions in plain English — new charts appear on the grid.

  4. Export

    One click: AI writes the executive report (PDF or Word) or generates a full slide deck (PowerPoint). Export. Send. Done.

Frequently asked

My CSV has customer PII. Is this safe to upload?

There is no upload. The file is read and parsed inside your browser tab. Our servers never see a single row. AI features only see column names and types — never the values.

How big a file can I analyse?

Comfortably into the hundreds of megabytes on a modern laptop. DuckDB-WASM is the same engine data engineers use server-side — now running in your tab.

Do I need to know SQL?

No. Describe what you want in English; AI writes the query and runs it. SQL is available if you prefer — it's full DuckDB with CTEs, window functions, and joins.

Is it free?

Yes to start — 100 free AI credits on signup, no credit card. Uploads, SQL, charts, and exports are free forever. You only pay when you buy more AI credits, one-time packs, never subscriptions.

Your next CSV shouldn't take your weekend.

Drop the file. AI picks the charts. You ship the answer before the 4pm review.

Analyze your first file