If you run a small business site or a blog and only ever open the standard reports in Google Analytics 4 (GA4), this guide is for you. Standard reports are good for the basics, but less so for questions like “Where do people drop off before my sign-up is complete?” That is the job of Explorations. Below: what they are, the techniques on offer, two walkthroughs, and the limits worth knowing.
What Explorations are, and how they differ from standard reports
According to Google’s help page “Get started with Explorations”, Explorations are a set of advanced techniques that go beyond standard reports to help you uncover deeper insights about user behavior. Standard reports help you monitor key business metrics, while Explorations give you data and analysis techniques that reports do not, for answering more complex questions. You can run ad hoc queries, switch between techniques, filter and drill into the data, build segments and audiences, share your work with others on the same property, and export results.
You open them from Explore in the left navigation. The screen has a few parts worth learning by name:
- Variables: the left panel holding your dimensions, metrics and segments. You can also change the date range here.
- Tab Settings: where you pick the technique for the current tab and configure it, including options such as Breakdown and Values.
A practical way to think about it: standard reports answer “what happened?” with a fixed layout, while Explorations let you choose the rows, columns and steps yourself. The trade-off is that you must decide what to ask. If your events are not set up yet, start with our guide to setting up GA4 key events for a small business site, because most useful explorations depend on having clean events to work with.
The techniques available
The help page lists seven techniques. Choose one in Tab Settings, or start from a template under Explore.
| Technique | What it does (per Google’s description) |
|---|---|
| Free-form exploration | A crosstab layout that can also show bar, pie, line, scatter and geo map visualizations |
| Cohort exploration | Studies the behavior and performance of groups of users who share a common attribute |
| Funnel exploration | Shows the steps users take to complete a task and where they drop off or do well |
| Segment overlap | Shows how user segments relate to each other, which can help you find new segments |
| User exploration | Lets you examine the users in your segments and drill into individual user activity |
| Path exploration | Visualizes the paths users take through your website or app |
| User lifetime | Analyzes user behavior and value across the customer lifetime |
The two most approachable for beginners are the funnel and path techniques, so the walkthroughs below use those.
Walkthrough 1: a funnel exploration for a newsletter or contact form
Goal: see how many visitors move from the page that hosts your form to a completed submission, and where they fall away. The steps below follow Google’s “Funnel exploration” page; the specific step names are only an example, so substitute your own events and pages.
- Go to Explore and, under “Start a new exploration”, click the Funnel exploration template.
- Click the edit pencil to add steps. Give each step a distinctive name, such as “Viewed sign-up page” and “Submitted form”.
- Add one or more conditions per step. Conditions can be based on events or on dimension values. For step one you might use the page_view event combined with a page dimension that matches your sign-up page. For the final step, use the event your form fires on success, for example the key event you configured earlier.
- Use AND or OR to add clauses, and Add step for more steps. Choose is indirectly followed by (other actions may occur in between) or is directly followed by.
- Optionally click Within to require that a step follows the previous one inside a set time frame, then click APPLY. A Summary card shows how your conditions match your data.
Settings worth adjusting
- Make open funnel: when on, users can enter at any step; when off (closed), they must enter at the first step.
- Breakdown: pick a dimension such as Device category to split each step into series.
- Next action: choose a dimension to fill a “Top 5 Next Actions” card for each step, which hints at where people go instead of completing the form.
- Segment comparisons: you can apply up to 4 segments, for example comparing visitors from different campaigns. If you tag campaigns, our guide to UTM parameters in GA4 explains how to keep those sources tidy.
Funnel limits to remember: you can define up to 10 steps, you cannot define steps based on metrics (only events or dimensions), and each user can enter the funnel only once in the date range, with only their first sequence of steps reported. Click Save if you want a funnel report created from the exploration.
Walkthrough 2: a path exploration after a landing page
Goal: find out what visitors do after arriving on a particular page, such as a popular blog post. This follows Google’s “Path exploration” page.
- In Explore, click the Path exploration template.
- Under STARTING POINT, choose a dimension, or drag one from the NODE TYPE list. The options are event name, page title, page path, screen name or screen class.
- Choose the value, for example the title or path of your landing page.
- Read the result: the starting point is the leftmost column, and STEP +1 shows the top 5 screens or events that followed it.
- Click a node to expand it and add another step; click it again to collapse. Gray nodes sit at the end of a path and cannot be expanded.
Useful controls
- By default each step shows the top 5 nodes. Click + More to show up to 20; anything beyond is grouped into “Others”.
- To trace the route backward, click Start over, then use the ENDING POINT box. A path can have a starting point or an ending point, but not both.
Two notes from the help page affect interpretation. Paths can span multiple sessions, where a new session begins after 30 minutes of inactivity, and node data aggregates across all sessions in the date range. Small user-count differences can also appear between path exploration, funnels, segments and audiences.
Limits that trip people up
Data retention: 2 months or 14 months
This is the most common surprise. Google’s “Get started with Explorations” page says properties retain 2 months of data by default, and admins can change this under Data Retention. The data retention page says standard properties can choose 2 months or 14 months for user-level data and for other event data. Importantly, it states that the setting affects explorations and funnel reports: a date range longer than the retention period shows no data for the extra time. Standard aggregated reports are not affected.
You need the Editor role to change it, it takes effect after 24 hours, and reducing the period deletes affected data at the next monthly deletion run. The two help pages describe the menu path slightly differently (Data collection and modification > Data Retention versus Data Settings > Data Retention), so look for “Data Retention” under Admin. Per the same page, increasing the period applies to data already collected that has not yet been deleted, so checking the setting early matters.
Sampling and thresholds
Per Google, if a query involves more than 10 million events, the exploration may use sampled data. Separately, data thresholds may filter out data that could identify individual users, which applies when demographic information or Google signals data is included. When either applies, the exploration’s status icon turns from green to yellow and a tooltip gives details.
Quantity limits
- Up to 200 explorations per user per property, and up to 500 shared explorations per property.
- Up to 10 segments per exploration, 10 filters per tab and 10 visualizations per exploration.
- Up to 20 dimensions and 20 metrics per exploration.
Sharing is read-only
Explorations are private until you share them, and sharing requires at least the Analyst role on the property. People you share with can view but not edit; to change it they must duplicate or copy it. Exports include Google Sheets, TSV, CSV and PDF. For a recurring dashboard instead, see building a monthly report in Looker Studio.
Key takeaways
- Explorations complement standard reports; they are for deeper, ad hoc questions.
- Start with a funnel exploration for a form and a path exploration for a landing page.
- Check your data retention setting first: the default is 2 months, and explorations show nothing for dates beyond it.
- Shared explorations are view-only; viewers must make a copy to edit.
FAQ
Do Explorations replace standard reports?
No. Google positions standard reports for monitoring key business metrics and Explorations for more complex questions the reports cannot answer.
Why does my exploration show no data for older dates?
Likely your retention setting. For non-aggregated reports like Explorations, a date range longer than the retention period shows no data for the extra time. Standard aggregated reports are not affected.
Can I share an exploration with a client or colleague?
Yes, with others who have access to the same property, and sharing requires at least the Analyst role. They see a read-only version and must copy it to edit.
Sources
- Get started with Explorations (Analytics Help)
- Funnel exploration (Analytics Help)
- Path exploration (Analytics Help)
- Data retention (Analytics Help)
