This might seem simple but there are quite a few ways to access GA4 data:
- You can use the GA4 standard reports under Reports
- You can also modify and filter these and even save them in the library for everyone else to use
- You can use the GA4 Explore section for more advanced analysis
- You can use Looker Studio for the data
- If you’ve set up the BigQuery export (which you generally should), you can do an analysis of raw data
- You can also use 3rd party tools to connect to your GA4 or BigQuery dataset
We’ll cover all of these except 3rd party tools. We also have an interactive form that helps you work out what method is best. Note: this sequence is mainly about GA4 data in GA4, not in BigQuery. If you’re using BigQuery may need to reconstruct some of the dimensions and metrics that are available directly in GA4 itself. However, using this guide can still help you work out what to do there.
Below are best case uses for each method of getting GA4 data:
- GA4 Standard reports are best for very basic questions that involve top performers (eg. top performing pages) and don’t need a lot of precision.
- GA4 edited Standard reports are best for answering such questions for a subset of your data (eg. only Australian traffic, only mobile traffic).
- GA4 Explore is best for more advanced visualisations with multiple filters, custom funnels, cohort analysis and segment overlap.
- Looker Studio is best if you have regular questions that you want answered in a structured way.
- BigQuery is best if you need advanced analysis, full precision and/or are reporting across a large number of items
Finally here’s a summary of features and limitations:
| Method | GA4 Standard reports | GA4 edited Standard reports | GA4 Explore | Looker Studio | BigQuery |
| Long-term reporting? | Yes | No (14 months) | No (14 months) | No (granular reporting will degrade after 14 months) | Yes |
| Demographic, interest/cross-device reports | Yes (if privacy thresholds allow) | Yes (if privacy thresholds allow) | Yes (if privacy thresholds allow) | Yes (if privacy thresholds allow) | No |
| Google’s data-driven attribution | Yes | Yes | Yes | Yes | No |
| Long-tail data (reports all values) | No (only reports top values) | No (only reports top values) | No (only reports top values) | No (only reports top values) | Yes |
| Advanced reporting (eg. cohort analysis, custom funnels, segment overlap) | No | No | Yes | No | Yes (but needs to be set up manually) |
