April was a great opportunity to test a few things I’ve been meaning to for months now and to see how our audience reacts to changes.
In this dispatch I’m going to look at a few big changes we’ve made including:
- New newsletter designs (does short form content convert?)
- Updating our publishing schedule (can we get the same response and engagement from less posts?)
- Can we change the way people read?
There’s lots to cover, so let’s get rolling with the metrics:
Page views: 205,246 (-18.42% from March)
Unique page views: 190,011 (-18.04% from March)
Users: 72,495 (-21.14% from March)
Sessions: 86,922 (-18.81% from March)
Bounce rate: 3.05% (+55.55% from March)
Average time on page: 38 seconds (+4 seconds from March)
Pages per session: 2.36 (+0.48% from March)
Total time gained/lost with readers: 765.225 (-11.9% from March)
New newsletter subscribers: 361
Facebook mobile (1,094)
Top 5 posts:
Why I killed my standing desk (18,908)
Time gained lost & the problem with organic traffic:
Before I get into some of our experiments from last month let’s take a look at some the metrics and try to make sense of them.
As I mentioned last month, time gained or total time reading (to borrow from our good friends at Medium) is one of the main metrics that I want to track to see how we’re progressing as a blog. Page views are great for measuring exposure, but what really matters is that people stick around and read.
Which isn’t an easy task. Do any sort of research into reading habits online and it’s a pretty dismal scene.
Since 1997, studies have shown that web users will only read, on average, 28% of an article. And that’s even only if they stick around.
In 2013, Slate writer Farhad Manjoo asked Chartbeat data scientist Josh Schwartz to analyze how readers were interacting with articles on their website. What he found was that 38% of people bounce right away while 50% are gone by the time they get one third of the way through an article. And by the end of the post? In Farhad’s words: “basically, only my mom is still reading this”.
Now, I don’t blame you online readers (I’m one of you, after all!). There’s just so much content out there vying for your attention that it’s impossible to commit to reading anything of length. That’s why it’s so important to build a core audience that trusts you (one of our main goals).
However, trust is something that needs to be built. It doesn’t come right away.
To illustrate this, let’s have a look at where our traffic came from over the last two months.
In April, we received 57.4% of our traffic from organic search. Here’s the big picture of where our readers came from:
Now, here’s a look at March (where our readers spent over 100 more hours with us):
Compared to last month there’s less organic traffic, more referrals, more traffic from social, and… more time spent reading!
To me, this shows the importance of context in reading online and in building trust with your readers.
Compared to coming in blindly from organic search, coming to a blog post from social media gives at least some context of what the brand’s voice is about, while being referred from another site transfers your feelings about the original page and adds clout.
I know we’ve tried to steer away from obvious and in-your-face design elements on the blog (there are no pop-ups, no big this-is-us-now-sign-up calls to action), but maybe this is to our own detriment? As I’ve talked about in the past, it’s all about creating a safe and comfortable reading environment, but there are many factors that influence that, not the least of which is knowing what a site is about as soon as you get there.
As I wrote in a recent piece about the psychology of simplicity, readers give your site 1/20th – 1/50th of a second before they judge it.
This isn’t to say that we shouldn’t be focusing on that direct traffic. We’re lucky to rank highly in SEO for a number of terms and get a lot of traffic to the blog through search. What it does mean, is that we need to be aware of the types of readers we’re getting and how to cater to them. We’re still trying to grow our core audience and the more of these one-time readers we can turn into return readers the better.
New newsletter format
Alright, now that’s out of the way, let’s look at some of the fun experiments we tried last month.
First (and foremost!) was the launch of our new newsletter template.
We started the project in Crew and were matched with the talented Ed Williams who worked with us for a few weeks, designing and refining the new newsletter templates.
For the past year or so, we’ve been sending out full articles in email campaigns, which was fine when we published once or twice a month. But recently I’ve received more and more emails from readers telling me they just don’t have time to read a lengthy article 2-3 times a week, especially in their email.
We get it. Your inbox is a busy place and it’s rare that you’ll want to sit down and read 1500-2000 words when there are more urgent tasks continually popping up.
So we went in a different direction (here’s what they look like in case you’re not signed up for our newsletter):
Pretty snazzy, eh?
While I don’t have much information yet on how the new design has effected traffic (we only launched it mid-April), here are a few initial observations:
Click-through rate went up: This one was pretty much expected. Instead of reading the full article in our inbox, you can now click through to read it on our blog (much nicer!).
Open rates stayed pretty much the same: This is a bit of a Catch-22. We couldn’t expect you to open a campaign because of a new design you didn’t see (because you didn’t open it…). While we still have a pretty decent open rate on our emails, this is something that is more based around content and will be a continual effort to improve.
What’s interesting is seeing how much the click and open rates have evened out since launching the new templates. Instead of spikes on the weekend and low opens during the week, we’re settling into a nice pattern and upwards trend. I’m excited to see how this continues.
Short form vs. Long form:
While the response has been mostly positive about the new newsletters, not everyone has been happy. One concern is that these short-form messages come across more like marketing emails, which is not what we want at all.
I decided to take a look at some stats from similarly themed posts, but which were sent out with the different templates.
New (short form) newsletter style: Creativity conditioning
Open rate: 29.6% (list average: 33.8%)
Click rate: 4.2% (list average: 3.2%)
Clicks per unique open: 14.1%
Old newsletter style: The hidden power of ‘I don’t know’
Open rate: 29.9%
Click rate: 2.3%
Clicks per unique open: 7.7%
This gives some surface information, but the next step will be to track the links from our campaigns and see how those readers are acting compared to others.
Getting more from less: Why I changed our publishing schedule
This month I also decided to experiment with publishing frequency to see what sort of effect would come from dropping from 3 posts a week to 2.
Since mid-October I’ve been posting on the blog on Mondays, Wednesdays, and Fridays, while pushing Backstage posts as they come up. I decided to drop this down to 2 times a week (Mondays and Thursdays) for a few reasons:
- I wasn’t convinced I was seeing enough of a return on 3 articles a week compared to the amount of work it took to write, edit, curate, and send out campaigns.
- Traffic was still going down at the end of the week no matter what we posted on Fridays.
- I consistently felt behind on my publishing schedule due to a self-imposed deadline, which meant that from time to time I’d run articles I wasn’t 100% confident in.
While I think it’s still a bit early to tell, I think the results are interesting to say the least.
While our overall page views were down MoM from March, they were still up a significant amount from February and we’re continuing to see steady growth. The drop in page views could just as easily be attributed to the lack of a runaway story like last month’s How to put an end to workload paralysis by Belle as it could to the change in frequency. Only time will tell which has more of an effect.
I also changed up our publishing schedule to see if I could influence when people read. Like most online publications, we saw a sharp decline in audience at the end of each week. I wanted to see if sending out new articles on Mondays and Thursdays would counteract that at all.
The short answer? Sort of.
After switching up the schedule we started to see more traffic on Thursdays when new articles went out, but continued to see a drop off on Friday and Saturdays. Is this just human nature? Maybe. But it might also have to do with the type of content we produce. It’s hard to convince someone to invest time in a thoroughly researched post when all they can think about is the weekend.
There’s still a lot more experimenting and tracking to do to see if these changes are having a negative or postiive impact on the blog, but I think the important part is that we’re doing them.
Sitting idly by and being happy with how things are is no way to see continual growth so we’ll continue to mess around and let you know how it goes.
If you have any suggestions or comments, let me know!