How To Use the Google Analytics Behavior Flow Report

Understanding user behavior on your website isn’t just about knowing how many people visit. It’s about discovering the actual journeys they take, where they get stuck, and what drives them to convert or leave.

Traffic counts stop short of that. You can watch 2,000 people land on a product page in a month and still not know which page sent them there, or where they went after.

Those two unknowns are what turn a fix into a gamble. Redesign a page that was never the drop-off point, and the problem can simply move somewhere else.

Google Analytics can answer both questions, just not in a report called Behavior Flow. That name belonged to Universal Analytics, and path exploration is what replaced it.

So that’s the focus here: how to build a path exploration report for your site, and how to read it in both directions.

In This Article:

What the Path Exploration Report Shows You

If you came looking for Behavior Flow, the path exploration report is where that analysis lives now — and it gives you more control over what you’re looking at.

The path exploration report visualizes your users’ path through your site. It shows you how people move from one page to the next, and where the experience could be better.

Two words to get straight before you build one, because Google uses both and they don’t mean the same thing. The steps are the columns, running left to right: the starting point, then STEP +1, STEP +2, and so on.

The nodes are what’s listed inside each column — the individual pages or events people landed on at that stage, each one printed with the number of users or events behind it.

In the chart below, the first column holds a single node, session_start, with 80,957 against it. The next column lists the pages those visitors opened first, the busiest being Stationery at 13,021. The pale bands running between the columns are the journeys themselves, connecting a node in one step to the nodes people reached in the next.

It works by letting you highlight different events and pages, so you can follow the flow of traffic across your site.

Using the path exploration report, you can:

• Understand where your users are most likely to go next when they’re on certain pages
• Spot pages people leave and then return to, by seeing an earlier page reappear later in the path
• See how many users continued from one page to the next, and how many left
• Work backwards from a page that matters — a thank-you page, say — to see which pages came before it

For example, say you have a product landing page on your site. You can analyze the exact steps users go through to land on that page and complete the checkout process.

Google’s reference on path exploration in Google Analytics documents the underlying technique. With that picture in mind, the next move is building the report on your own property.

Why Dashboards and Ask Advisor Don’t Answer This Question

Google Analytics has added two new features recently, and if you have been using either one you may be wondering why this guide sends you to Explore instead.

Dashboards arrived in September 2026 and live inside Reports. You pick from score cards, tables, line charts, bar charts, donut charts and funnel charts.

There is no path or tree visualization among them, and segments aren’t supported yet — so the two things this guide leans on most are the two things a dashboard can’t give you. It’s built for watching numbers you already understand, not for tracing a route you haven’t seen.

Ask Advisor, the Gemini-powered assistant behind the icon in the top-right corner, is the other one. Google documents it for questions about your data, for “why did this change” and for how-to help.

Paths and journeys aren’t among the uses it documents, and the closest example on that page returns advice about re-engaging visitors rather than a picture of where they went.

Use them for different questions. Ask Advisor is the quickest way to find out why a number moved. Path exploration is the only one of the three that will show you the order people moved in. And if you have been leaving Explore alone because it looked like the experimental corner of Google Analytics, it’s worth another look — Dashboards, Ask Advisor and the Tasks menu are the recent arrivals here. Explore has been in place for years.

How to Build a Path Exploration Report in Google Analytics

Building the report takes about a minute, and none of it involves code. The whole thing lives in one section of Google Analytics called Explore.

To create the path exploration report, click Explore in the left-hand menu, then choose the Path exploration template:

When you click the template, Google creates the path exploration report for you. It’ll look something like this:

By default, the report shows you a flow built from events. It might look confusing at this point, but don’t worry! We’re going to change it.

Check one setting before you read too much into a sparse-looking chart. Your property’s data retention controls how far back explorations can look, and the standard default is 2 months.

If nobody on your team changed that setting, a path older than two months isn’t there to chart. The report isn’t broken — it’s just looking at a short window. Very high-volume properties are held at two months whether they like it or not, so on a busy site it’s worth checking rather than assuming.

That default event view is the first thing to change — and the same dropdown also decides what the report’s numbers count.

How to Choose the Right Node Type for Each Step

The dropdown at the top of each step decides most of the meaning in this report. Pick the wrong option and the chart still looks right while the numbers answer a different question.

That dropdown sets the node type — what each box in the chart stands for. Click the dropdown in STEP +1 and choose one of the Page title options:

The Event name option makes each node a count of events. The page and screen options count users instead.

So the same path can produce two very different totals depending on which option you pick. It’s worth deciding what you’re counting before you read a single number.

For page-level analysis, Page path and screen class is often the cleaner choice. Page titles repeat across similar pages and get truncated in the chart — you can see it in the screenshot above, where a node reads “Apparel | Google Merc…”.

If you’d rather compare pages than trace paths, our Google Analytics landing pages report guide covers that view instead. Once the node type matches the question you’re asking, the chart is ready to read.

One time-saver if you build the same comparison often: since late 2024, an Editor can save a segment for the whole property rather than rebuilding it inside every exploration.

How to Read a Path Exploration Report

Once you can read it, the shape of this chart carries as much as the numbers do. Two things quietly distort it though — collapsed nodes and unlabeled rows — and neither announces itself.

Click any page to see where those visitors went next. Another step opens up showing the pages they visited afterward.

The chart is drawn to scale, which makes it readable at a glance once you know to trust it. Every node’s bar is sized in proportion to its count, every band between columns is as thick as the traffic flowing along it, and the same scale is used across all the steps — so a node in step three can be compared directly against one in step one.

A hairline branch is a handful of visits. A thick band is most of your traffic.

The counts are printed next to every label as well, and hovering a node repeats the figure, so you can check the shape against the number rather than estimating. That pairing is what makes one detail in the chart above worth a second look: the +15 More bar in the second column is taller than every named page beside it.

Those 15 collapsed pages hold 39,098 of the 80,957 sessions that started — very nearly half the traffic, sitting behind a control you have to click.

Tracing a drop-off means following the widest band forward until it thins out. The step where the width collapses is the page pair worth investigating first.

Looking at the chart as a whole, the proportions come through clearly. Wide bands show where traffic concentrates, thin ones where it scatters, and the occasional row carries no label at all.

By default, each step shows the top five nodes. Clicking the + More control on a step reveals up to 20, and anything past that is grouped into a single Others node.

That’s worth remembering, because a step can look tidier than your site actually is. If a page you expected is missing, expand the step before concluding nobody visited it.

A row labeled (not set) is a placeholder Google Analytics uses when it received no information for that dimension. Read those rows as missing data, not as a real page on your site.

If you want a fixed set of steps rather than an open-ended path, our Google Analytics funnel exploration guide covers that setup. Reading the chart this way gives you something concrete to decide: which page pair to test next.

Using the Start Over Button: Forward vs. Reverse Path Analysis

Path exploration runs in two directions. The second direction is the one that answers “what happened right before the purchase?”

Both directions start from the same control. Start over sits in the top-right corner of your path exploration report, boxed in red below:

What the Start Over Button Does

Clicking Start over lets you begin a fresh path exploration. From there you pick between two kinds of analysis:

• Forward: set a STARTING POINT to see what people did after it
• Reverse: set an ENDING POINT to see what people did before it

Each report takes a starting point or an ending point — not both. That constraint is why the reverse direction is its own exercise rather than an extra column on the forward chart.

Here’s how the reverse direction works in practice. Say you sell hoodies and want to know what leads up to a sale.

Click Start over, choose ENDING POINT instead of a starting point, and set it to your purchase event. The chart then builds outward in the other direction.

The step beside your ending point holds the pages people were on immediately before buying. Each step further out goes one page earlier in the same journey.

You might find that a comparison post sits one step before checkout more often than your category page does. That’s a concrete case for putting a buy link inside the post rather than reworking the category page.

The same approach works for leads. Set the ending point to your form submission event to see which pages produced them.

If that event isn’t recorded on your site yet, our form conversion tracking guide walks through the setup.

Run the report once in each direction and you’ll know two different things: where visitors stall, and which pages earn your conversions.

Four Limits of Path Exploration to Know Before You Trust the Data

The report is precise about what it measures and quiet about what it leaves out. Four limits are worth knowing before you make a decision based on what it shows.

1. One direction per report. You choose a starting point or an ending point, never both. Answering both questions means building the report twice.

2. Only the busiest nodes appear. Each step shows the top five nodes until you expand it, then up to 20. Everything beyond that collapses into a single Others node, so a long tail of small paths can drop out of view. Worth keeping separate from the (other) row you may have seen in tables — that one is a row-limit overflow, and it isn’t what’s happening here.

3. High-cardinality dimensions get grouped. Google’s guidance is that any dimension with more than 500 unique values in a single day should be treated as high-cardinality. Page-level dimensions on a big site can cross that line, and the detail you were after can end up inside a grouped row.

4. Sequence is not causation. The chart shows the order pages were visited in, not the reason. A page that appears before every purchase may be doing the persuading, or it may just sit in your main navigation.

None of that makes the report unreliable. It makes it a source of hypotheses you can then test, which is a healthier way to use any behavior report.

How to Track the Full Purchase Path Inside WordPress With MonsterInsights User Journey

A chart built from thousands of sessions can confirm that traffic reaches your checkout and still leave you stuck. Which pages did the person who placed yesterday’s order actually read, and how long did they spend on each one?

Percentages can’t answer that. The moment those sessions are grouped into nodes, the individual disappears.

MonsterInsights, the WordPress analytics plugin, records that missing detail in a feature called User Journey. In your WordPress dashboard it sits at Insights → Reports → eCommerce → User Journey, and it opens on a list of every order you’ve tracked.

Each row is one order. You get the transaction ID, the purchase date, and the UTM campaign, medium and source that brought that buyer in, alongside the order total and how many steps they took to get there. The filters above the table narrow it by campaign, medium or source, and the search box pulls up a single transaction if you already know its ID.

Steps to Purchase is the column worth scanning first. A buyer who took three steps found what they wanted almost immediately. One who took twelve went looking — and those are the journeys worth opening.

Clicking a transaction ID opens that buyer’s full path: one customer’s page-by-page record, attached to one order, listing the pages they visited on the way to buying in the order they visited them.

A date header groups the steps, and each step gives you three things: a timestamp to the minute, the page title with its URL path, and the seconds spent there. The first row carries no duration, because that’s the entry page, and the homepage shows up as the label (Homepage). Read down the list and you have the shape of one purchase — which page opened it, where they hesitated, and what they were looking at when they decided.

Getting that detail recorded takes a specific setup. Four things need to be in place:

• A MonsterInsights Pro plan or above
• The eCommerce addon active
• The User Journey addon active — that’s how the toggle is labeled on the Addons page
• A supported store plugin holding your order data — WooCommerce, Easy Digital Downloads, MemberPress, Restrict Content Pro, LifterLMS, or GiveWP

With all four in place, every new order records its own journey from then on.

Each journey is keyed to a completed transaction, so this is the report for the question “what route led to this sale?” rather than “where did everyone else give up?” Pre-purchase drop-off lives in the separate Cart Abandonment Report, which is also Pro and above with the eCommerce addon active.

User Journey doesn’t replace path exploration, though. Google Analytics shows you how the crowd moves; User Journey shows you what one buyer did. Both are worth having.

Video Tutorial: Google Analytics 4 Behavior Flow

If you would rather watch the report being built than read about it, this walks through the same steps on screen.

GA4 Path Exploration Report FAQ

What is the path exploration report in GA4?

The path exploration report shows you how your users flow through your site, where they get stuck, and where they most commonly go next after visiting a page. You build it yourself in the Explore section instead of opening it from a list of standard reports.

Is there a user flow report in GA4?

Yes, but not under that name. Universal Analytics had Users Flow and Behavior Flow, and Google Analytics replaced both with the path exploration technique inside Explore. It covers the same ground, with more control over what each step represents.

How do you set up path exploration in GA4?

Click Explore in the left-hand menu, choose the Path exploration template, then change the dropdown in STEP +1 from the default event view to the node type you want. The section on building a path exploration report above walks through each screen.

Does the path exploration report show conversion paths?

Partly. A reverse path exploration that ends on a key event shows the on-site pages that came before a conversion. Google Analytics also has a channel-level version at Advertising → Key events → Key event attribution paths, but that report needs the property linked to a Google advertising product.

Specifically, it needs an advertiser link such as Google Ads, DV360, CM360, or SA360 — a publisher link like AdSense or Ad Manager unlocks the Publisher ads report instead. So without an ads link, reverse path exploration is your route to this question. Our conversion path tracking guide covers the on-site side.

Why does my path exploration show (not set)?

(not set) is a placeholder Google Analytics uses when it received no information for a dimension, and the cause depends on which dimension you’re looking at. Two are worth knowing for path work: landing page reads (not set) when a session has no page view event at all, and values tied to session_start or first_visit can read (not set) when they weren’t present on the session’s first event. Read those rows as missing data rather than as a real page, and try a different node type if they crowd out the paths you’re after.

You may also meet (data not available), which is a different thing. It applies to source and medium, and unlike (not set) it usually isn’t something you can fix on your end.

How far back does GA4 path exploration data go?

Explorations are bounded by your property’s data retention setting, and the standard default is 2 months. If nobody changed it, a path older than that won’t appear. You can raise the setting in your property’s data retention controls, though that won’t bring back data which has already aged out.

Why don’t the numbers match my reports?

If you use consent mode, part of the answer is behavioral modeling. Google applies it to only some of the Explore techniques — path exploration is one of them — so a path chart can include modeled users while a standard report of the same period doesn’t. That’s enough to move the totals apart. Treat a path exploration as a picture of the route, and your reports as the count.

Can I export a path exploration report?

Yes. In the upper right of the exploration, click Export data and pick a format: Google Sheets, TSV, CSV, PDF, or PDF (all tabs). You can also share the exploration itself with anyone who has access to the property.

The format changes what you get. Sheets, TSV, and CSV give you all the data in the visualization you selected, which can be more than the graph shows on screen. A PDF captures only what’s currently displayed.

We hope this article helped you build a path exploration report you can actually trust — and read it in both directions. If you liked this article, check out the following beginner-friendly guides:

• How to Add Google Analytics to WordPress
• eCommerce Customer Journey Analytics: WordPress Guide
• A Complete Guide to GA4 Conversion Tracking for WordPress
• How to Create Google Analytics 4 Custom Reports
• Top Website User Engagement Metrics in Google Analytics
• User Behavior Analytics: Tools, Key Metrics & More

Follow us on Twitter, Facebook, and YouTube for more Google Analytics tips.