At the April 26, 2021, Newtown Technology and Communications Committee (T&CC) Zoom meeting, several people complained that it was difficult to navigate the township website (www.newtownpa.gov) and find the information they were seeking (listen to the discussion).
Considering that the website is the main way that the township communicates with residents, it is crucial that its usability be improved. As Peter Drucker, who was arguably the world's greatest management consultant, famously said, “If you can’t measure it, you can’t improve it.”
It just so happens that Google Analytics is one of the best tools available to “measure” the usability of websites – and it’s freely available. All that is required is a Google account and a special code to be inserted in every page, both of which are currently in use by the township.
Should the township periodically run Google Analytics on its website to determine how it is used and where improvements may need to be made? Below, I review the analytics for my website to get an idea of what data is most useful to look at.
Case Study: JohnMackNewtown Website Analytics
I run Google Analytics on my website (www.johnmacknewtown.info) every month. It’s actually done automatically by Google. To access the data, I merely login at which point I’m at the home page. The home page automatically includes charts and tables for different time periods – usually the last 7 or 30 days. To get more specific data, all I need to do is click on the various reports available and set the time period. Easy peasy!
Let’s look at some of these data to get an idea of whether or not they would be useful to have for the township’s website as well. [Spoiler alert! Yes, it would be useful to have at least some of this data for the township website.]
The Audience Overview is probably the first thing you would look at when running analytics. The following chart shows the Audience View for my website for the month of April 2021.
There is much useful information here, including the number of users, sessions, pageviews, and average session duration. About 77% of those sessions resulted in the user navigating away from the site after viewing just one page. This is referred to as the “bounce rate.” I need to improve that number, but it’s nice to know that 14% of visitors were new.
I notice that there was an unusual spike in users on April 28. What was that due to? For the answer see the section “How Do I Acquire Visitors to My Website?” below.
How Many Pages Were Viewed in April?
Probably the most important parameter you want to keep track of is pageviews. A pageview is an instance of an Internet user visiting a particular page on a site. A pageview is recorded whenever a full page of the website is viewed or refreshed. In April 2021, my website recorded 2,092 pageviews.
Of these 1,721 (82%) were “unique” pageviews i.e., pageviews that combine the pageviews from the same person (a user in Google Analytics), on the same page, in the same session. So, if you reload a page or visit a page two or more times before leaving the site, it is counted as one unique pageview.
The average monthly number of pageviews this year (Jan through Apr) is 2,048 (1,334 in Jan, 2,140 in Feb, 2,626 in Mar). Thus, there was a total of 8,148 pageviews in that period compared to 5,174 for the same period in 2020 – a 57% increase! I must be doing something right!
What Were the TOP Pages Viewed?
Google Analytics can show you the Top 10, 25, 100, or whatever pages viewed. The following is a representative table of pageview data.
From this I see that 58% of visits to the homepage were “bounces,” which means the visitor had no interaction with the page; i.e., did not click on a link in the page to other pages on the site. For 2021 to date the bounce rate for the homepage was 52%. Obviously, I’d like to improve upon that.
One of the most important pieces of information to gauge a website’s usefulness is the origin of visitors; i.e., referrers such as websites, search engines, etc. The following chart shows the referrers to my website during the month of April, 2021.
You can see that on April 21, patch.com and scoop.it were the major referrers accounting for 88% of referrers, whereas on April 28, surveymonkey.com was a major referrer accounting for 75% of the known referral sites.
What was going on those days? What I know is that on April 21, I posted the article “The Newtown Planning Commission Versus Wawa” on Newtown Patch. That article includes multiple links to the website.
On April 28, my survey “What Should Newtown Spend Federal COVID-19 Relief Funds On?” had a spike in responses (123). That survey also has links to the website. Looking at the Surveymonkey analytics, I see a spike in responses via the Newtown Patch on April 28 (see chart below). These responses most likely came from my Patch article I posted on April 25. But why the spike on April 28? I’m not sure. Perhaps Patch promoted it that day. I bet that if I were more adept at using Google Analytics I could find the answer to that question. I’m just happy to know that Patch helps me get visitors to my website and surveys.
Google groups traffic sources to websites into 4 “channels”:
- Referral (see above)
- Organic Search
Google Analytics defines direct traffic as website visits that arrived at a website either by typing the website URL into a browser or through browser bookmarks.
NOTE: Visitors also are directed to my website via links in my email newsletter, which is sent to 650 opt-in subscribers. In April, I sent 5,752 emails, which resulted in 360 clicks – mostly to my website. I’m not sure these are counted among the 663 sessions via the direct channel.
Another interesting piece of information refers to the devices used to access my website. In April, 2021, 55% of sessions were from desktop computers, 44% from mobile devices (i.e., phones) and 2% from tablets. Interestingly, of the 869 total users in April, a majority (58%) accessed my website via their mobile phones. That means I need to pay close attention to making sure pages are readable via small screens.
When Do Users Visit My Site?
One more piece of data is the number of users by time of day, shown in the chart below. Note that Wednesdays (including April 28) had the highest concentration of users, particularly in the hours between 12 pm and 3 pm. Consequently, I surmise that somehow the Patch article was highlighted at 12 pm that day.
How Does This Apply to Newtown’s Website?
Not all the types of data presented here may help improve the usability of the Newtown Township website. IMHO, the most important questions we need to ask include:
- how many users visit the site?
- what are the top 10 or 20 pages viewed?
- are important pages being viewed?
- what are the major referral sites?
My hope is that the TC&C can make recommendations to the township regarding what analytics to collect and how often to collect the data. The data must also be analyzed and put into reports that will help the township improve the usability of the website.
Posted on 03 May 2021, 01:48 - Category: Communication