Stop 13 · Version Control, Git & CollaborationTopics · 3/7

Working in a Team Data Workflow

In short

Solo, you can keep things in your head. In a team of five analysts, you need a shared way of taking requests, writing queries, reviewing and publishing, or numbers start to disagree.

How analysts collaborate without stepping on each other — shared query repositories, code review for important dashboards, and agreed naming/style conventions.

The jaadu in a white cap rides a bicycle carrying a wooden crate full of tiffin boxes.

Think of it like this

Mumbai's dabbawalas

One dabbawala collects the tiffin, another carries it on the train, a third delivers it to the office. Each knows the colour codes and the handover points. Thousands of lunches reach the right desk. Team data work runs on the same kind of agreed system.

Watch it happen

A request's path through the team

Ticket

Branch

Query

Review

Merge

Publish

1/6The request is logged as a ticket with the question, requester and deadline. Nothing lives only in chat.

The big ideas

What you need to know

One place for requests

A board in Jira, Trello or even a shared sheet. Everyone sees what's being worked on and what's waiting.

Shared metric definitions

Agree once on what "active" or "revenue" means and write it down. Then every analyst uses the same one.

Peer review by default

Another analyst checks your joins and filters. It catches mistakes before a VP sees them.

Link everything

Ticket links to branch, branch links to dashboard. Six months later, anyone can trace a number to its query.

Save yourself the pain

Common mistakes

Oops

Each analyst keeping a personal version of the revenue query on their laptop.

Do this instead

Keep one reviewed query in the shared repo and build everything from it.

Oops

Taking requests in private DMs that nobody else can see.

Do this instead

Move every request to the shared board, even if it arrived on chat.

Your mission

Run a mini team workflow

~40 min

Quick check

Two dashboards show different monthly revenue numbers. What team practice would most likely have prevented this?

Go deeper

1Read about a real team's analytics workflow (many companies publish these as blog posts).
2Learn why even lightweight code review catches real errors before they reach a dashboard.
3Draft a simple style guide (naming, formatting) for your own SQL going forward.