| |

How to Use Google Analytics to Build a Useful Website Performance Report

Part 2 of a series on SEO reporting best practices

(ICYMI: Part 1)

Traffic data, found in Google Analytics, helps you quantify your site’s engagement level. Using this data, you can understand more than just how many users visit  your site. Traffic data can reveal what information users find useful and help you prioritize optimizations to emphasize the right content for your audience. 

The notion of “traffic” makes intuitive sense to people. You have visitors to your website; go ahead and count them. But the reality is that there are multiple ways to quantify traffic. Understanding the nuances behind what differentiates each type of traffic insight can help you ask better questions and unearth better answers.

Note – the metrics discussed in this article are available in Universal Analytics (UA), the data that Google will soon be sunsetting in favor of GA4, a newer model. We will be discussing the subtle but impactful differences between these two platforms in a later post. 

Pageviews vs. entrances
Sessions and users
Engagement metrics
Understanding behavior with event tagging
Measuring conversions using goal completions
Your arsenal of e-commerce metrics
Sync with your CRMs
Caveat regarding traffic insights
Getting the most out of Google Analytics
Preparing for GA4

Pageviews vs. entrances

A pageview is, quite literally, a visit by a user to a specific URL on your site. Comparing the number of pageviews across your site tells you the number of visits to each separate URL. 

Pageviews are a great measurement of existing performance and provide a data point that anyone can understand. However, pageviews only become actionable when combined with other metrics that provide context around the numbers. 

Pageviews help understand which pages are consistently attracting users and helping them in their decision journey, regardless of how or why.

Entrances are a subset of pageviews. An entrance is the first pageview a user makes in a particular instance of visiting a site.

The utility of assessing pageviews and entrances can vary widely, and heavily hinges on understanding what pages you are looking at. Most businesses wouldn’t consider their “About Us” page to be a money page, but they may receive high numbers of pageviews from users trying to learn about your brand and form a decision.

data table comparing entrances and pageviews by page
An analytics exercise for you to try out: sort your top pages by Entrances to identify your most popular landing pages. Then, sort your top pages by Pageviews to identify your most popular content—period. You can then check that the top pages by Entrances have obvious, accessible links to the top pages by Pageviews.

 

Sessions and users

A user is a distinct individual to your site. A session is an instance of a user visiting your site. Both numbers are used to measure total site traffic. 

It’s important to remember that a user can (and frequently does) have multiple sessions over time. While these metrics tend to be closely correlated, the number of users can speak to the size of your audience, while sessions can shed light on re-engagement patterns.

We frequently look for trends in sessions and users to identify whether a site’s audience is growing, shrinking, or stable over time. 

It is important to consider both metrics alongside each other. Sessions dropping could be tied to a decline in the number of users, or a decline in repeat visits from an existing set of users. These two instances of session drops point to two very different problems.

chart tracking growth of users, sessions, and sessions per user
Traffic growth is always a positive sign and a general measure of a site’s success, measured both by Users and Sessions. For an editorial client like the one above, Sessions Per User growth also provides a positive signal that users are beginning to recognize the site as a trusted source of information, resulting in several return visits.

If sessions drop while the number of users holds steady, you will see the number of sessions per user decline. This may indicate that something on your site is deterring users from making return visits or trusting your site. 

Combining this with other data points can help you evaluate whether the issue lies with your messaging, branding, UX or how your product/service measures up with your competitors. 

On the other hand, a decline in the number of users could point to an issue with your marketing campaigns, SEO, or PPC, preventing customers from visiting your site in the first place. If possible, the reason for this drop should be attributed to a specific channel, page, geographic region, etc. so that you can take the most strategic approach to optimize —and accurately quantify its effects.

 

Engagement metrics

Two key data points for user behavior are average site duration and pages per session. Both data points showcase how deeply users engage with a website. 

You can discover whether users gravitate towards a quick engagement with the site, whether they move quickly on pages and only engage with a few pages, or if they spend more time on more pages.

Usually, these behavior metrics will measure higher in e-commerce or highly technical industries, where visitors are trying to learn about products or complex concepts.

Higher is not always better, though. Suppose users spend excessive time on the site or click through tons of pages to find what they want. This could be the result of poor user experience or content mismatched by what the metadata promised. 

Again, you must consider these behavior metrics together with other values, like sessions and users, to really understand what’s happening with your visitors.

data table comparing average session duration and pages per session
In the example above from a pre-school, we see higher pages per session coming from the “Our Schools” and “Programs” sections, where parents can evaluate a specific location or program. This is where parents can determine if it’s right for their child, and are likely closer to making a conversion than those on the home page, where these metrics are much lower. This lets us know that the secondary navigation and CTA’s on these pages are critical in guiding users to relevant information.

 

Understanding behavior with event tagging

Event tagging helps you better understand  how users interact with your pages, getting a bit more granular than just “what pages are they visiting.” An event occurs whenever a user interacts with any element of your site, including visiting a particular URL, loading an image, downloading an app, making a purchase, and more. You can track any number of these events on individual pages by tagging them in Google Tag Manager, and you can report on the number of events that occur in a particular amount of time.

The events you choose to tag drive the level of insight you’ll gain. For example, tagging form submissions, button clicks, and scroll depth can reveal detailed engagement insights that help identify pain points in the conversion process. One of the most common uses of event tagging is to A/B test calls to action, so you can ultimately rely on the one that consistently converts.

You also get to set the data parameters for each event you track. Say you’re examining e-commerce conversion; for every purchase, you may choose to track things like the item value and coupon codes used in conjunction with a particular event.

If high-traffic pages have very few clicks on whatever item you need to convert–calls-to-action, form submissions, etc.–you may have an issue with effectively driving users on a clear path to complete these actions. This would likely trigger your re-evaluation of the page’s layout and/or UX features. 

Do you need improved heading text, more images, better navigation, or a change to another page feature? You may need iterative rounds of A/B testing to successfully optimize this page, but with each round, you learn more details about your target audience.

When you consider which events to tag, get granular. Events are a fantastic solution to answer discrete questions about user behavior.

web page screenshot with event tagging opportunities highlighted, including calls-to-action and hyperlinks
In the example above, Event Tagging helps the client understand how effective their various CTA placements are in the conversion process & make adjustments accordingly. Tagging items like “Add to Cart” is also essential for an e-commerce business to understand which items drive interest even though they may not lead to sales, and which items have a larger dropoff rate from people who Add to Cart to those who make a transaction.
Table of events data in Google Analytics with clearly labeled categories
When a site has clear, organized event tagging, data will funnel into Google Analytics to help you understand the metrics behind each of your CTA’s or other page elements. Labeling these actions is equally as important as tagging them in the first place, since you’ll need to be able to make use of the data based on naming conventions, as seen below. You won’t always see which visual element each Event represents, so clear naming helps you down the road.

 

Measuring conversions using goal completions

Goal completions help you track and optimize your conversion rate. Unlike events, goal completions are configured to measure activity that your company considers a “conversion”, such as a lead- or revenue-generating action. 

A goal completion might be a specific button click, form submission, phone call, or even a “Thank You” page indicating that a user completed a form or made a purchase. You can measure the frequency of these goal completions on your existing web page, or as you deploy any new site campaigns or changes, including content, messaging, UX features, or calls to action.

You can use these metrics to assess the incremental value of your marketing efforts including your PPC, SEO, and more. You may evaluate goal completions and website changes side-by-side to understand how many conversions and leads were driven by these digital marketing strategies.

A helpful way to understand the difference between goal completions and events is that it typically takes several events to lead to a goal completion.

 

Your arsenal of e-commerce metrics

Anyone with an e-commerce site can speak to the dream of data that describes how much revenue is generated at each step of the customer journey, while reporting detailed variables around products, pages, demographics, and other audience characteristics. The deeper this reporting, the more sophisticated and accurate your insights—and the more value you bring to stakeholders.

Of course, the two fundamental e-commerce metrics are transactions and revenue; these likely bear no explanation to anyone who’s set foot in e-commerce. However, several more detailed metrics lend themselves to more actionable insights.

  • Conversion rate & new visitor conversion rate – Not only can you measure conversion rate, but you can filter our specific channels, audiences, and demographics for comparison. 

Maybe you don’t need to optimize your website; perhaps it’s your email marketing that isn’t working. You can tell from this data, and then decide how you want to address that (remember, you can use site components—even something as simple as a landing page) to help optimize other channels.

  • Average order value – For real-time monitoring of purchase trends broken down by channel, audience, or demographics, rely on Google Analytics to provide you with data on average order value. 

Based on the trends you observe, you might decide to market your offering a bit differently by channel. If that is the cause, you’ll likely set up tests—using events and goal completions.

  • Cost per acquisition – This metric is determined by dividing the total spend by the total number of leads generated, telling how much it is costing you to generate a lead in a particular campaign. CPA data helps evaluate the efficacy of PPC advertising, and determine whether certain keywords or ad campaigns are more profitable.
  • Exit pages/bounce rates – When a user exits or bounces from your website, they likely lost interest, did not find what they needed or sought to quit shopping with you and look elsewhere. 

Understanding where users are bouncing can help you narrow down the problem to a specific product, product characteristic (e.g., pricing), or maybe more evergreen content that turns users away. 

  • Site search keywords – What are users directly searching for in your site’s search bar? This can be a bit more qualitative of a measure, depending on scale, since smaller sample sizes may not illuminate trends.

Consider why users are searching for these keywords, and what challenges they are facing. Clearly, they are not finding what they need from your site; is this a navigation issue, a content issue, or some other problem? 

You can also leverage site search to ideate around new content or keywords to rank for, since site search is a cue around interest.

As you set up your data tracking, avoid falling into the trap of believing that there is one cause for any issues. More likely, you are seeing the convergence of several smaller issues that each need to be addressed separately.

 

Sync with your CRMs

Many businesses have steps in their sales process that cannot be automatically measured in Google Analytics. In fact, unless you run an e-commerce website, some of the key actions towards making a purchase happen off-site. Think of a university where students can apply online (measurable in GA), but must have their application reviewed, work out their financial aid, etc. before actually enrolling (not measurable in GA). The same goes for many B2B or B2C services that generate leads via their website (measurable in GA) but provide a consultation and finalize their sale in person (not measurable in GA).  

Although these offline actions are not measurable in Google Analytics, they are still measurable. Businesses in this scenario should have details on their leads, and how many of those leads convert into customers, identifying which leads come through their website or other sources. 

Many CRMs like Salesforce and Hubspot offer integrations with Google Analytics. You can incorporate leads or sales data from your CRM into your reporting to help you better understand your bottom line. From plug-ins to API solutions, options are available to help connect your CRM to Google Analytics and attribute leads, and revenue to specific pages or campaigns. Such connections are worthwhile for better understanding the impact of your website. 

Graphic illustrating that qualified lead data is often missing in Google Analytics
As discussed above, some of the most important steps in your sales process happen after the client leaves the website and becomes a lead. Qualifying the lead and making the sale happen later, but it’s key to tie these later steps back to your website activity to understand how much value the site is driving.

Find ways to incorporate leads or sales data from your CRM into your reporting to understand the bottom line. 

Caveat regarding traffic insights

It’s also important to remember that more traffic does not always translate into more revenue. You can “do everything right” with your SEO and PPC work, and see increases in impressions, clicks, traffic, sessions, events, and even some goal completions, yet still not see increased e-commerce revenue. 

Perhaps the product is not meeting the wants and needs of your audience, or the user experience is not ideal, or you are attracting the wrong traffic (people who are not your target audience). 

If this is the case, the report will show that conversion rate optimization (CRO) is the issue, rather than a lack of traffic. Events and related metrics like cart abandonment may reveal where attrition occurs.

Always consider all possible variables that may influence trends observed in your data, and then set up tests to measure each one.

 

Getting the most out of Google Analytics

There is no single perfect metric. Every data type listed here illuminates a different aspect of website visibility and engagement, and must be considered holistically and over an appropriate amount of time. 

SEO reporting is not just about assessing the current state of your site but also about understanding the next critical steps to optimize your pages and their content. Data is your guide in  your quest for maximum utility and visibility in front of your target audience. Carefully think through the purpose behind each data point you collect and how it connects to your target audience’s journey, and you can set yourself up for maximum success.

 

Preparing for GA4

As you may have heard, Google Analytics will soon update from Universal Analytics, their previous data model, to Google Analytics 4, a newer method of tracking and reporting that focuses more on engagement metrics. Many of the principles above will still apply, but there will be subtle changes in the naming of metrics and which metrics/dimensions are available. Check back for updates related to this change. 

 

If you need help setting up your website for GA4 or are looking to learn more about your own site through SEO reporting, contact us today.

Similar Posts