Knowing precisely how people are using with your site is massively valuable. If you’re struggling when it comes to conversions or you’re worried about users leaving halfway through the user journey/funnel, getting a better idea of how users are engaging with key areas of the website can tell you the issues that users are having, and which changes could potentially be made.
Even if you’re just curious and you want to see how people are interacting with content – no goal or conversion in mind – it’s interesting to note precisely where users are going and at which points they’re leaving.
In order to get a better idea of this, we can look at the User Flow report available in Google Analytics. It’s one of the lesser known parts of GA, but it’s so, so useful. Here, we’ll have a quick look at precisely what the user flow report includes, how it can be used, and precisely why it’s so useful.
What is a User Flow Report?
The User Flow report is available in Google Analytics, and it offers information on how users are interacting with your site by detailing the journeys that they take throughout their time on the site. It does this by offering a graphical representation of where users started off, the pages that they visited during their visit and where they left your site.
By looking at a User Flow graph, you can get a much deeper understanding of how people are using your site, which can help enhance your technical SEO strategy.
Now, looking at a user flow graph for the first time can be a bit confusing as there’s a fair bit going on. Here’s an example of one:
Green blocks – represent a nodes, which is a page or a group of closely related pages. You’ll likely see/representing the homepage, whereas /blog can include all blog posts on the site.
Grey paths – are connections, representing segments of traffic going from one page/group to another.
Red arrows – indicate the drop-off rate: users leaving the intended user flow, or ditching the site entirely. Pages with a high drop-off rate are to be monitored, especially if they’re key pages that may also have high bounce rates.
Going further into the user flow graph, here are the key features that are available in the interface:
In the User Flow report, you can change the type of traffic you’re looking at by selecting custom segments.
Above, we can see that you can view different traffic sources, allowing you to contrast and compare different types of visitors, looking into their differences in behaviour. With the segments feature in Google Analytics, you can of course venture into a whole array of different territories, seeing as they’re massively customisable.
For example, you could set up segments for desktop users and mobile users, allowing you to see the differences in behaviour between users on different devices.
Another way for you to chop and change the subset of data you’re looking at would be to use custom dimensions, available on the left-hand side of the user flow interface.
Here, you’ll be able to look at users based on dimensions such as the country they’re from, particular on-site behaviours, their ad group, etc. A particularly helpful one here would be Landing Page.
This allows you to choose particular pages on the site where users landed, offering a more in-depth look at particular pages and how they’ve performed. It’s particularly handy for blog posts, and seeing where people have ventured to upon landing on one of them.
Highlight Traffic/View Traffic
You can highlight particular segments of the user journey by clicking on a node and selecting “Highlight traffic through here”
This offers a really clear look as to how people are interacting with the site via certain paths, with it being very clear to see where users are going.
Level of Detail
You can select the level of detail shown within these reports, offering either a more detailed or a more simplified view of how users travel throughout the site.
Fewer connections offer the main user journeys, and are generally helpful for more basic overviews of the site and can make things a bit simpler if you’re looking to showcase this to a client. More connections allow you to have a more in-depth look at the connections throughout the site.
The initial graph that you’re shown only offers information on the first few steps in the user journey. On the right-hand side of the graph, you’ll find a + Step button, which will add an extra set of pages that users have visited. This is handy for much deeper analysis, identifying the user journeys of people who have ventured very far into the site, or for sites with a rather lengthy conversion process.
Conducting User Flow Analysis
Now that we know precisely what we’re looking at and understand the interface, we can properly look into why the user flow graph is so useful, and what information can be found through it.
Looking at the general overview of your site is a real key asset here. Just from an initial first-hand look, you can see where people started off, and where they went to next.
This can offer insight as to whether certain calls to action are working as intended, whether users are interacting with key navigational points and internal links, as well as whether the pages are engaging or not.
For example, a site that cropped up for us recently:
Looking at this – without changing any of the dimensions or segments – we can see that the homepage has a substantial drop-off rate. On top of that, people aren’t going through to the pages that have been explicitly linked on the homepage, they’re going through to more informational pages, bouncing around between them without viewing any key landing pages.
Sounds like an issue, doesn’t it? The homepage isn’t retaining users well enough with its massive drop-off rate (also coinciding with a large bounce rate), with the calls to action on the homepage seemingly not doing their job either, suggesting that changes should be made here.
Moving further into custom segments and dimensions, we can compare two different types of users. One example mentioned previously would be to compare mobile and desktop users, looking into their behavioural differences.
For example, for another site, here are the first few steps of the user flow graph for the desktop site:
Also, here’s the same chart but for mobile users instead:
From the homepage, they tend to go through to different parts of the site. Mobile users are more driven towards a single brands pages, whereas desktop users tend to go through to more standard category pages, and then onto product pages. Looking further into this particular path, we can see that these users tend to back to the homepage before dropping off/leaving, indicating that they’re not really getting what they’re after.
The internal linking and call to action structure of the homepage is different on mobile when compared to desktop in this instance; something which needs to be assessed going forward.
You can also use the user flow graph to assess booking/checkout processes, identifying any potential issues that users are having.
The node on the left represents the initial landing page, the middle node represents users who went through to the first stage in the booking process for the site, while the nodes on the right are the next URL the user went to.
Depending on your own booking/checkout process, you can look into areas such as:
- Steps in the process which have massive drop-off rates: are users leaving the process at a certain step? Which issues are users facing here?
- Where users were before they were before they went to the first stage of the booking/checkout process
- Where users went throughout their journey. Did they bounce around between various stages before converting or dropping off? Again, what were the issues that users were having here? Use the aforementioned Next Step feature, as well as Highlight Traffic Through Here, in order to get a better understanding of longer journeys.
So, that’s a rather simplified look at the User Flow graphs in Google Analytics and why I personally find them so useful. You can really understand how people are using the site, and can pinpoint what is or isn’t working, as well as particular issues that users are having.