Connect data
Connect your data from the hub.
Your website, your Shopify store, the files you already export — and, on plans that include them, your own database and your own data lake. Everything is set up from Connectors in hub.bumbleb.co and lands in your own workspace, so you can ask across all of it. Data on your own computer stays there.
On every plan today
Your website or web app
Google Tag Manager
Add a Custom HTML tag to your GTM container and push events to the data layer — page views, sign-ups, purchases, feature use, errors. You can hash user IDs in the page before anything is sent.
Set up under Connectors → Google Tag Manager.
Your store
Shopify pixel
Add a custom pixel under Settings → Customer events in your Shopify admin. It captures your storefront’s standard events — page and product views, add to cart, checkout started and completed.
Shopify today, built to extend to other e-commerce platforms. Events start from the day you install.
The data you already export
File upload
Upload Excel, CSV or Parquet files — ad-spend exports, Google reviews, order lists, anything with rows and columns — and ask about them on their own or next to your website data.
Upload a fresh export whenever you want newer numbers.
Pro adds — everything above, plus
Your own database, behind your firewall
bb-connector
A container you run inside your own network. It reads your database, applies your rule to every field — drop, hash, round or send, and anything without a rule is dropped — then syncs over outbound HTTPS.
Syncs MongoDB today. You decide what leaves, and in what form.
Data that already lives in your lake
Your own data lake
Bring your own data lake: link its tables with a read-only sign-in and ask about them alongside everything else in your workspace — no export to keep up to date.
Set up from Connectors in the hub.
Data on your computer · every plan
Ask about it without uploading it
A Postgres or MongoDB database on your machine, a folder of CSV or Parquet files: connect them in Acorn for VS Code or BumbleB Crunch Desktop, tick the tables a conversation may read, and ask. Nothing is set up in the hub and nothing is copied to your workspace.
Read where it lives
Your tables are never copied to your BumbleB Crunch workspace or to ours. Crunch sees the names, columns and size of the tables you pick, and reads the rows its queries return.
Reads only
Every query is checked to be one read-only statement, and Postgres is attached read-only. Connect with a database role that sees only what you want analysed.
Every query logged
Each query Crunch runs is listed with its SQL, its row count and how long it took.
Approve each query, if you like
Turn on Approve each query and nothing runs on your computer until you say so.
Secrets stay with you
Connections, passwords and connection strings stay on your computer, and so do the results a conversation keeps — deleted after 30 days unused, or when you delete them.
Conversations you can reopen
Your questions and the answers are saved to your BumbleB Crunch account, so you can pick a conversation up again.
Combine datasets
Two datasets, one question
Because everything lands in one place, you can put datasets side by side and ask across them — website behavior with ad spend, store events with orders, your database with your traffic. BumbleB Crunch™ works out how they relate before it answers.
Ad spend comes in as an export
Download campaign spend from your ad platforms and upload the file next to your website data. No ad-platform connector needed — the export is all it takes to see which spend comes back.
Any dataset
Any dataset
The same reasoning that finds a funnel leak works on an orders table, a listings feed or an operations log. Ask the question you actually have — BumbleB Crunch works out what in the data can answer it.