The key web analytics metrics fall into three groups: acquisition (how people find you), engagement (what they do once they arrive) and conversion (whether they complete an action that matters to your business). Start with visitors, sessions, traffic sources, engagement or bounce rate, pages per session, exit pages, conversions and conversion rate. Then choose the handful that tie directly to a business goal, because no website needs every number a dashboard can produce.
This guide explains what each metric measures, how to read it without being misled, and how to pick the right ones for your own site.
A long list of metrics is hard to act on. A simpler way in is to sort them by the question they answer. One widely used analytics guide organises web analytics into acquisition, engagement and conversion: how visitors find you, which pages and actions they interact with, and whether they complete an important outcome.
Keep those three questions in mind as you read the rest of this article:
If you want a broader primer before going deeper, our own web analytics basics for SMEs covers the groundwork.
These are the figures most people see first, and they are also the easiest to confuse.
Why the difference matters: a rise in pageviews with flat visitors usually means the same people are looking at more pages, or refreshing. A rise in visitors with flat pageviews suggests new people are arriving but not exploring. Neither is good or bad on its own. Each is a prompt to look further.
In Google Analytics 4, the acquisition overview report shows users and new users, plus new users and sessions grouped by channel. Treat these as your baseline: record them monthly so that you can see direction rather than reacting to a single day.
Traffic sources show where visitors come from. The usual groupings are organic search, paid search, direct, referral and social. Used well, they tell you which channels bring people in and which are underperforming.
Google makes a useful distinction between two reports. User acquisition is scoped to new users, while traffic acquisition is scoped to sessions. In plain terms:
GA4 also defines new users by channel as people who have never visited your site or app before, grouped by default channel and sorted by most new users. That makes it a quick way to see which channel is introducing your brand to fresh audiences.
Compare channels by what visitors do afterwards, not by volume alone. A channel that sends 200 visitors who enquire is worth more than one that sends 2,000 who leave at once. If organic search is thin, our guide to proven strategies to get more website traffic is a practical next step.
The old definition of bounce rate is the percentage of visitors who leave after viewing only one page. GA4 changed the thinking. It measures engagement rate, the percentage of sessions that were engaged. An engaged session meets any one of three criteria: it lasts longer than 10 seconds, it has one or more key events, or it has two or more page or screen views. Bounce rate in GA4 is simply the sessions that were not engaged.
That matters because a visitor who reads one page for four minutes and phones you is no longer counted as a failure.
Even so, bounce rate on its own can mislead. A high rate on a landing page might mean the traffic was irrelevant to the page, or that the page answered the visitor’s question, and the number alone cannot tell you which. A contact page with a high bounce rate and plenty of phone calls is doing its job.
Pages per session (also called views per visit) is the average number of pages viewed in a session. It is calculated as total pageviews divided by total sessions, and a value close to 1.0 means most visitors view only a single page.
This is most useful for content-led and ecommerce sites, where browsing is the intended behaviour. For a single-purpose landing page, one page per session may be perfectly healthy.
Session duration measures the time visitors spend in a visit. Be careful with how your tool calculates it. Some tools count the engaged time the page was actually visible and focused, while others work from the gap between the first and last event in the session. Because of that, a visitor who reads the final page and then leaves may not add any measured time at all.
So read duration alongside other signals. One analytics reference notes that high duration combined with high bounces can point to navigation problems, and that duration should be interpreted with bounce rate, page depth and conversion context.
If engagement looks weak, the cause is often the page itself. Our guide on how to create a website that converts covers layout, clarity and calls to action.
Entry (landing) pages show where visits began, and exit pages show where they ended. Together they sketch the route people take.
The terms are easy to mix up:
An exit is not automatically a problem. Everyone leaves a site eventually, and the best exit is a confirmation page after a purchase or enquiry. But a high exit rate on a pricing or checkout page deserves a look. As one funnel guide explains, it is a signal to investigate, though it does not show where in the funnel people got stuck.
Conversions are the completed actions you have defined as important: a form submission, a booking, a download, a sign-up or a purchase. Conversion rate is the percentage of visitors who complete a goal, or conversions divided by visitors, multiplied by 100 where goals are configured, as one analytics tool documents it.
A quick worked example, using invented numbers for illustration: if 4,000 people visit in a month and 80 submit your enquiry form, your conversion rate is 2 per cent. If a campaign lifts visitors to 5,000 but enquiries stay at 80, the rate falls to 1.6 per cent. Traffic grew, results did not, and that is exactly what the metric is there to show.
Conversions only appear if they are tracked. In Google Analytics, an event measures a specific interaction, and a key event measures an action particularly important to business success. Event tracking is also what lets you see actions inside a page rather than just page loads, for example button clicks, downloads or video plays.
For most UK small businesses, the events worth setting up first are:
One measurement specialist advises tracking events that represent meaningful steps toward becoming a customer, and avoiding collecting random granular events without a hypothesis. That is sound advice: ten well-chosen events beat a hundred nobody reviews.
To go further on lifting the rate itself, see our pieces on conversion rate optimisation for SMEs and 10 essential conversion rate optimisation tips.
Where your site sells directly, revenue tied to conversions is the metric closest to the bottom line. Pair it with average order value and, over time, with how much a customer is worth across repeat purchases (customer lifetime value).
Lifetime value is useful as a planning figure, but it is an estimate built on assumptions about future behaviour. Treat it as a guide to which channels attract your better customers rather than as a precise number, and recalculate it as real purchase history builds up.
The most common mistake is starting from the dashboard. Nielsen Norman Group recommends the opposite: define the goal first, identify the behavioural signals that indicate it, then choose the metrics or methods to measure them. Analytics data is the means of gaining insight, not the end.
It also helps to structure your measurement in layers. One approach separates business outcomes, performance indicators and diagnostic signals:
Another analytics vendor makes the same point: choose KPIs that reflect your business goals and connect the stages of the customer journey, and there is no need to measure everything.
If you are not sure what your goals should be, our guide to setting marketing goals with SMART strategies is a good place to begin.
Choosing the right tool matters too. Our comparison of the best website analytics tools sets out the options, and our article on website performance monitoring and ROI shows how to keep the numbers working for you.
Most businesses we meet are not short of data. They are short of a decision. A dashboard with thirty metrics feels thorough, yet nobody can say what it should change on the website next week. Our approach is to work backwards. We start with what the business needs, whether that is more enquiries, more online sales or a stronger brand presence. Then we agree the two or three behaviours that signal it, and only then set up the tracking. This keeps reporting short, honest and useful. We also believe analytics works best when design, development, SEO and marketing are joined up. A conversion rate is the result of a site that loads quickly, says the right thing and makes the next step obvious. When one team handles those parts together, the numbers tell a clearer story and fixes happen faster. The best first step is rarely a new tool. It is a clear list of what a visit is worth to you. Rob
Most businesses we meet are not short of data. They are short of a decision. A dashboard with thirty metrics feels thorough, yet nobody can say what it should change on the website next week.
Our approach is to work backwards. We start with what the business needs, whether that is more enquiries, more online sales or a stronger brand presence. Then we agree the two or three behaviours that signal it, and only then set up the tracking. This keeps reporting short, honest and useful.
We also believe analytics works best when design, development, SEO and marketing are joined up. A conversion rate is the result of a site that loads quickly, says the right thing and makes the next step obvious. When one team handles those parts together, the numbers tell a clearer story and fixes happen faster.
The best first step is rarely a new tool. It is a clear list of what a visit is worth to you.
Rob
Once you know which numbers matter, the work is acting on them: improving the pages that attract traffic but fail to convert, strengthening the channels that bring in the best visitors, and making sure your tracking records real enquiries and sales.
That is where we can help. We design and build websites, including WordPress and ecommerce, and support them with SEO, PPC, social media and video marketing. If search visibility is your starting point, our Cheap SEO Packages begin at £255 as a one-off price, and you can see everything we offer on our services and packages page.
If you would rather talk it through, have a look at what we do or get in touch through our website for a free consultation.
Start with visitors, traffic sources, engagement or bounce rate, conversions and conversion rate. Visitors and sources show how people find you, engagement shows whether the page held their attention, and conversions show whether the visit achieved anything. Add revenue if you sell online.
No. A high bounce rate can mean the traffic was a poor match for the page, but it can also mean the page answered the question straight away. Bounce rate alone cannot tell the two apart, so check conversions and traffic source before changing anything.
A bounce is an exit that happens on the landing page. Exit rate counts visits ending on a given page at any point in the journey, as explained in this journey analysis guide. Every bounce is an exit, but most exits are not bounces.
Fewer than you think. Pick the goal first, then the behaviours that signal it, then the metrics that measure those behaviours. For many small sites, five to eight well-understood numbers reviewed monthly will outperform a long dashboard nobody opens.
Any completed action you have decided matters to the business: a purchase, a form submission, a booking, a download or a newsletter sign-up. It only shows in your reports if you set it up as an event or key event first.
Book a Demo
Forgotten Password
Get your free SEO guide
Thank you, please check your email
Sign into Brainiac Media
Please sign-in using your email address and password.
Forget your Password?
no worries, click here to reset your password.