Analytics: Pageviews v. Impressions

Perhaps the title of this article should be Google Analytics versus PHP… you be the judge. I have heard often about the disparity between other site analytics packages and Google Analytics, but this has me wondering a lot. Read all of the facts below, before you make any conclusions.

Just so that you know that I am not just tossing things out there, or making things up, let me give you some insight on my understanding of the difference between impressions and page views.

Example #1

Let’s say a web site has a banner ad placed above its header on all pages. An Internet user / visitor opens up the home page, and therefore they see the ad.

The count so far: the web site has received one page view, and the ad has made / received one impression.

Let’s then pretend that the user clicks through to another page, and that same ad is above the header.

Now, the blog has received two pageviews, and the ad has also received two impressions.

Can we agree on those statistics? 2 & 2.

Example #2

In this example, let’s say a web site has a banner ad placed above its header on JUST THE HOMEPAGE.

An Internet user / visitor opens up the home page, and therefore they see the ad.

The count so far: the web site has received one page view, and the ad has made / received one impression.

Let’s then pretend that the user clicks through to another page, but that same ad is not present at all, on that page.

Now, the blog has received two pageviews, but the ad has one impression.

Can we agree on those statistics? 2 & 1.

Analysis:

Because there are so many different advertising opportunities across the web, it is NOT accurate to say that pageviews and impressions are exactly the same. In fact, the impressions (of an ad) are often noticeably lower than the pageviews. For good reason – often, advertisers (for a host of reasons) do not have an ad on every single page throughout a web site.

Allow me to explain my conundrum:

I queried Google Analytics for my pageview statistics from September 18, 2016, to September 18, 2017, and here are the stats. (I know – yuck. It’s like I completely abandoned the site and moved to Antarctica!):

Figure 1: Google Analytics Screenshot. Click to enlarge.
Figure 1: Google Analytics Screenshot. Click to enlarge.

Here is where I start questioning things. My banner / campaign ad tracker shows ad campaign ID 1, which presently shows, and has ALWAYS shown, on all web pages throughout the web site, since I have been using this campaign and ad management tool. Everywhere. In this particular regard, truly, the number of impressions should pretty much dead-on equal the number of page views in Google Analytics.

Figure 2: My campaign ad tracking summary page. Click to enlarge.
Figure 2: My campaign ad tracking summary page. Click to enlarge.

But as you can see – the number of page views is grossly different than the number of impressions. The way I am reading these statistics:

Pageviews = 295,419, or straight average: 24,618 per month.

Impressions = 15,775,950, or straight average: 1,314,662 per month.

Now, I am thinking – there must be something wrong with the campaign / ad tracking mechanism, right?

No, there isn’t. When a single web page is loaded from my web site… the campaign tracker creates a very distinct, singular row in the campaign tracking table. Now, another note to make… in Figure 2, the campaign management tracking summary shows the ENTIRE lifetime impression number (in this case, lifetime = how long I have used this management software/plugin). To capture specific data from September 18, 2016, to September 18, 2017, I had to do a custom query to arrive at those specific numbers:

Figure 3: My custom SQL for pulling data from September 18, 2016, to September 18, 2017. Click to enlarge.

So, what to do? Nothing, really. But I created this post, to contest what a potential advertising client might offer, when they see Google Analytics data versus my own tracking data.

We're happy to share this resource that we found. The content displayed on this page is property of it's original author and/or their organization.

Leave a Reply