One of the concerns I had about switching from seven-day-a-week posting to just five was the possible impact on my web analytics — page views and visitors.
I check web analytics fairly often to try to keep an eye on how I’m doing. I watch for things like page views, unique visitors and bounce rate and compare prior counts.
Presumably, higher numbers now mean you’re doing better than you had previously.
Back in July, I did something I had been thinking about for a while. For me, I just took a while to make the decision. But I decided to drop the number of days I posted here from seven to five, weekdays only.
I posted seven days a week for more than six years, after all, and that’s a lot of pressure for any blogger. But when you consider my real job involves writing for the web and managing news and social media sites, the pressure was a lot higher. I feared the quality of the posts might be suffering so I finally took action.
If you’re like me, you want some signal that the decisions you make are right.
I turned to analytics seeking that answer. But things have gotten a bit more complicated than they used to be.
Yes, web analytics are now a different animal.
You may remember that in May of last year, a European law that affected most of the planet took effect. Known as the General Data Protection Regulation — GDPR for short — it forced changes on data sites tracked.
In doing so, it also was expected to affect the web analytics site owners would see.
You probably noticed many websites, including this one, asking you to opt-in to limited tracking. If you opt-in, I get numbers; if you don’t, I get far fewer numbers to work with.
For the record, I don’t really care where you browse or who specifically is visiting. I’m more interested in where visitors — whoever they are — go once they get here. I want to know which pieces of content they go to most often. Some might surmise that this is a good indication of the kind of content I should be producing more of.
We knew to expect this. Websites run by people with a better grasp on such data warned us it would happen. They said we’d see drops in traffic, page views, and more.
Worse, they said we’d lose the first pageview for visitors — meaning we’d see their second and later pageviews after they opted in. We’d see shorter sessions (the time viewers spend on your site). The bounce rate, a measure of the percentage of visitors who stay on a single page and then leave, might be shorter than it really is.
In a nutshell, the numbers I see in 2019 are mostly lower than they were after May of 2018. Prior to that, I posted some gains, modest, but gains nonetheless, year to year.
That’s definitely not the case in 2019. Part of it may be a loss of interest and a genuine loss of visitors. But it’s at least possible that GDPR may have at least some impact.
That means you have to be careful with web analytics.
You’re better off looking at web numbers year-to-year when looking at performance. Each month is different, after all. As seasons change, as vacation times comes along, as school ends or begins, people’s habits change. Their web habits may be part of that.
But as I said, with GDPR now in effect, comparing year to year is brutal.
I made the change from 7-day to 5-day posting during the first week of July. So when I compare July 1-August 31, 2019, to the same dates in 2018, I see a 41% drop in page views and unique visitors. GDPR took effect in May 2018, so on one hand, one should expect the dropoff to have already taken place.
But not everyone visits a blog every day. And with a blog that pulls in as much organic search traffic and as many new visitors as mine, GDPR still packs a punch.
What if I compared July 1-August 31 of this year to the two months before that?
Well, even that’s a bit complicated, it turns out, because I had a spike from a few search engines I’d never heard of in the middle of May. I’m still trying to figure out exactly why the spike happened. The number one post at that time was one from years earlier, a grammar post about the idiom, “the spitting image.” Why would that post, written in 2015, have a spike in May of 2019?
You have to love the web!
With that spike, comparing July-August to May-June, I show a 12% drop in page views and a 10% drop in unique visitors. Since that spike seems like an anomaly, if I remove the first third of the first month, and compare July 21-August 31 to May 21-June 30, I have a 4% increase in page views and a 7% increase in unique visitors.
So what should I conclude from that?
I’m going to take it as a win. The data seems to suggest that I haven’t had a major loss in visitors since switching to what feels like a more manageable posting schedule. When I took out the numbers that included a bizarre spike, I had a slight improvement. Had the spike not occurred at all, the numbers may have been closer to flat.
I’d still take that considering the fact that I’m generating less content.
So if you’d read this far, thank you for visiting this blog. And I will thank you for visiting again and again over the years.
I really hope you will be back!