Most dashboards start from what the data source offers. Someone connects GA4, adds a few charts for sessions, users and revenue, and shares the link. A month later, people still ask for the same numbers by email, because nobody can tell what the page is meant to help them do.
Start from the opposite end. Name one recurring decision and the person who makes it, then let that decide the sources, charts and filters. Everything else can wait for a separate report.
The tool in this guide is Data Studio. Google renamed Looker Studio to Data Studio in April 2026, so older tutorials may still use the old name. The dashboard below is a hypothetical example. It describes a design and has not been built on real shop data, so check the steps against your own report.
Write the decision before you open the tool
A useful dashboard brief fits in three lines. Here is one for an invented small online shop:
- Who: the e-commerce manager, every Monday morning.
- Decision: if sales fell last week, should this week's effort go to traffic, the checkout, or nothing yet?
- Action: review campaigns with marketing, ask the developer to test the checkout, or wait another week.
Now you can test each candidate chart with one question: which of those three actions would this number change? Total users fails that test here, so it does not go on the page. A device split passes, because a fall on mobile alone points to a different task than a fall everywhere.
If you cannot name an action for a chart, you have two honest options: move it to a reference page or leave it out. "It might be interesting" is a reason to explore, and GA4 itself is a good place for that.
