This guest article graciously contributed by Alan K’necht, Founder of Knechtology, Inc.
One of the earliest KPI (Key Performance Indicators) to emerge back in the 1990s was pageviews. Reporting on how many page views there were in total or per visitor, was about all that was available to be reported on at the dawn of the web. Yet this KPI has lingered on as a KPI 30 years later. As marketers, we need to truly understand its meaning and determine if there is any value in reporting on it for measuring marketing efforts or site engagement.
If you can, think back to what a website might have looked like back in mid-to-late 1990s. There were no screen overlays, no iframes, just basic HTML pages. Each page of content was an actual HTML page and the view of that page was easily captured in log files and reported on.
Now think about your site today! If someone completed a form on the site, is there a thank you page? About half the sites now (could even be more) don’t have thank you pages, but merely pop-up an overlay saying thank and confirming a successful form submission. In the old days, accessing the form and getting a thank you page would generate 2 page views. Modern websites only generate 1 page view for that form.
You are no longer comparing apples to apples when measuring pageviews between an old site and a new site. The same issue arises if your site has a database lookup feature (for example show all inventory) and where the user can refine their search criteria or result search order (price lowest to highest vs. Highest to lowest). Frequently these refinements don’t generate pageviews but merely redisplay the same data on the same webpage.
Over the nearly 30 years of website marketing, management has come to perceive pageviews as a KPI. How many page views did a specific marketing campaign generate or which ads generated the most pageviews per ad click. For most marketing efforts as described, these types of KPI are meaningless and hold no value for measuring success.
Do pageviews as a KPI hold any value on modern websites?
There should only be one type of website where measuring pageviews as KPI or a factor with KPI holds any value. That is for sites that generate ad revenue based on display ads (e.g. revenue per hundred displays). In this case, where a site generates revenue based on how many ads are displayed, yes there is value in evaluating how successful different efforts to drive pageviews are. Just keep in mind to only measure pages where these ads appear, and to account for any pages where ads may get refreshed but no page view is generated (e.g. a list reordering)
It is still more beneficial to measure and report on ads displayed as a result of a marketing effort, but it may take extra effort versus the ease of reporting page views.
Alternatives to Reporting Page Views!
Once you’ve wrapped your head around the fact that page views (for most websites) are not a good indicator of true user engagement, you need to define what is. Each website should have its own unique KPI for user engagement. Here are a few ideas to bonder and some issues you need to address with your analytics manager:
Time on Site
This works great, but remember most analytic tools only know the start time (when a user arrived on a page) and when they clicked something that is recorded (i.e. link to another page). So that last page of a visit has a start time but no end time and will be treated as a ZERO. The solution is to set-up a timer event (for example every 10 seconds), that is triggered as a user spends time on any page. At least on that last page of a visit, you’ll have a good approximation of how long they spent. Another issue is what happens if a user simply leaves that page open in a browser tab for an extended period of time? The solution is to set a maximum number of timer events (i.e. 10 events at 10 seconds = 100 maximum time on that last page).
Page Depth
Do you have long pages? Does your site employ an infinite scroll concept? If yes, contemplate measuring the percentage of users who reach the 50%, 75% or the end of those long posts. The only issue is, you may have some short pages, which when loaded are 100%. You’ll need to define which pages you need the scroll tracking enabled on. This is typically something you may want for your blog posts.
Clicks on Links to External Sites
If your site has links to 3rd party sites or to your social media accounts, track when people click on those links. If these events are recorded, then that time stamp will mark the end of engagement on your site. Links can be categorized. The added benefit of tracking these types of events is you’ll learn if your users are actually clicking on those links. If they’re not, then perhaps you can use that screen real estate differently and move the links to somewhere more effective.
Track the Obvious
This includes tracking all lead generation forms, user registrations and of course orders.
Lastly, within your analytics report define what constitutes an Engaged User vs. Unengaged. Then create a comparison report of Site visitors who engage vs. Don’t Engage and breakdown by marketing channel, campaign, and perhaps geography. Now you will clearly see which marketing efforts are driving engaged users and not merely users who like to click around your website.
Alan K’necht started his involvement with the digital world back in 1996 building websites and participating in the precursors to what would evolve into Social Media. Over 20 plus years he’s provided consulting services ranging from web development, analytics and social media marketing to fortune 500 companies, government departments, military and small & medium size businesses. At present he operates his consulting company K’nechtology Inc.
Interested in being a guest contributor to the Digital Summit content library? We’re looking for articles, short-form video, and more. Get more info here.