Home Blog How Looking Deeper into Analytics Changes Impressions and Strategy


How Looking Deeper into Analytics Changes Impressions and Strategy

Posted by drew on March 17, 2016, 2:22 p.m.

Comparing recent analytics with past numbers is a prime way to understand how your SEO, social media, and paid search efforts are affecting your traffic. It is useful to see how traffic before a site redesign stacks up against traffic after the redesign. This will give you a good sense of whether you managed to hold on to SEO ranking through the redesign. However, it isn’t sufficient to look only at total traffic numbers to make a diagnosis on the state of your site. Often, digging deeper will provide additional insight that will change the way you move forward with your site.

As an example, the year-over-year data for client X shows a 26% decrease in sessions, and a 5% decrease in users (for the given time period). Here are the changing impressions I had as I continued to search through the data.

Screen Shot 2016-03-15 at 1.41.10 PM.png

Impression #1

This would initially indicate that marketing efforts have worsened, that the site’s new design or SEO were suffering, that search traffic had decreased dramatically, or some combination of these. However, the acquisition channel of the traffic shows an increase in organic search traffic and direct traffic. The biggest decline was actually from referral traffic, which dropped 81.44%.

Impression #2

With this new information, a webmaster or SEO might look to link building. Likely, they would check how their backlinking profile has changed over the last year, looking for large sources of links that had been lost. But a huge loss in linking authority would likely have an impact on ranking and organic search traffic as well, so the fact that organic traffic still improved in this time calls this theory into question.

In the coordinating period of the previous year, there were 1,552 sessions from referral sources, and in the current year, there were 288. That leaves 1264 sessions lost. Looking at the actual referrers, I see 84 sessions in the previous year from semalt.semalt.com and 1183 sessions from livereport.in. Both of these are spam referrers, and neither one contributed a session in the more recent year. Those two spam sources count for 1267 fewer sessions year over year, which more than accounts for the decline in referral traffic.

Impression #3

A little digging shows that the decrease in referral traffic is negligible, because most of the sessions were lost from spammy referral sources, and not from actual users. In fact, when you filter out traffic from those two referral sources, there was actually a 6.8% increase in overall traffic.

This kind of referral spam can be filtered out in Google Analytics. The goal is to have a more realistic sense of actual users visiting your site, so you can make meaningful changes to your marketing efforts and web presence.