Migrating From v1¶

If you used IndieBI DDA v1, downloading a CSV or Parquet file per table and opening it in Excel, welcome to v2. You get the same data, kept fresh automatically, plus a few new ways to work with it. You choose how hands-on to be: keep working with CSV files exactly as before, click through your data in the browser, or ask Claude in plain language.

What You Need First¶

All three options below share the same one-time setup:

Install Python 3.12+ and uv.
Download and unzip the workspace, then run uv sync in the folder.
Put your USER_ID and SAS_TOKEN in a .env file (Getting Your Credentials shows where both come from).

Getting Started Manually walks through each step. Once that's done, pick whichever option fits you.

Three Ways to Work¶

1. Keep Your CSV Files¶

The closest match to v1. One command exports every table you have access to as a CSV file, one file per table, ready to open in Excel:

Export every table to CSV

uv run export-csv

The files land in an exports/ folder, named after each table. Run it again whenever you want a fresh copy. No queries, no code.

2. Point and Click in the Browser¶

Prefer to browse and filter without leaving a spreadsheet-like view? Launch the interactive UI:

Open the interactive SQL editor

uv run duckdb-ui

It opens in your browser with all your tables loaded. You can sort, filter, write simple queries, and export any result to CSV. See Querying Data for details.

3. Ask Claude¶

Describe what you want in plain language ("show me revenue by platform last quarter", "which countries sold the most units") and Claude writes the query and builds the chart for you. No spreadsheets, no code. See Getting Started With Agents.

Which Should I Pick?¶

Just want your files like before? Use export-csv.
Like clicking through data yourself? Use duckdb-ui.
Want answers and charts without the manual work? Ask Claude.

You can mix and match. Nothing stops you exporting CSVs one day and asking Claude the next.

What Changed From v1¶

Table Names¶

Tables have new, clearer names in v2. The export-csv files use the new names, so here is how the old ones map:

v1 table	v2 table
`dim_portals`	`portals`
`dim_products`	`products`
`dim_sku`	`skus`
`dim_studio`	`organizations`
`fact_sales`	`sales`
`export_fact_sales`	`sales_per_sku`
`fact_visibility`	`visibility`
`fact_wishlist_actions`	`wishlist_actions`
`fact_wishlist_cohorts`	`wishlist_cohorts`
`fact_events`	`events`
`fact_baseline`	`baseline`

Identifier Columns¶

The identifier columns that tie rows together have changed in v2, in particular product_id, unique_sku_id and studio_id. If you relied on these keys to match or join rows across files in v1, expect them to work differently now. Tables & Schema Reference lists the columns available in each v2 table.