Science blogging: The essential guide - Christie Wilcox, Bethany Brookshire, Jason G. Goldman 2016

Metrics

Measuring the success of your blog

Matt Shipman

Most people start communicating about science on the Internet because they want to reach a wide audience. But now that comment threads have moved to Twitter and Facebook, it’s becoming harder to tell who is seeing your work and what kind of influence you have. Matt Shipman is a public information officer at North Carolina State University and blogger at Nature Publishing Group’s scilogs.com. He will explain how to tell if there’s anyone out there, and if they are listening.

So you started a blog to give yourself a place to communicate with people about science. And hopefully you started the blog with specific goals in mind. You may want to inspire a new generation of astronauts, raise awareness about the importance of the microbial life that surrounds us, or simply share an online diary of your field expeditions in the frozen Arctic or the Arizona desert. Whatever your intentions, you need to have some way of determining whether you are meeting those goals, whatever they are. In short, you need metrics.

Metrics are simply tools that you can use to measure, well, anything. But you will want to focus on those metrics that actually measure progress toward your goals. It’s easy to get overwhelmed by numbers. There are a lot of different things you can track—from how long people spend reading your work to how many people are sharing it—and it’s important to figure out which numbers are important to you and how you can use that information to share your work more effectively.

In this chapter I will explain the nuts and bolts of collecting data, what those data can tell you, and how you can use that information. I will also highlight the importance of creativity and critical thinking when it comes to measuring success. It may be easy to track numbers, but numbers alone can be misleading. You probably didn’t start writing or making videos with the ultimate goal of getting a thousand viewers or of having people spend more than ninety seconds on your landing page.

Most people get into science communication because they love science. They want to change the way people think about scientists, educate them about scientific subjects, encourage them to become scientists, or get them involved in citizen science projects. Numbers alone won’t tell you if you’re achieving those sorts of goals. So while it’s good to keep track of the numbers, it’s important to come up with additional metrics that will help you determine whether you are accomplishing what you set out to do in the first place.

Conventional metrics

If you’re managing a blog, you probably have some fundamental questions: How many people are visiting my blog? Which posts are most popular? Where did these people find my blog in the first place? The best way to collect this sort of information is by using analytic tools that are provided by your blogging service or that you can add to the blog yourself.

Analytic tools give you a variety of metrics about how many people are visiting your blog, where they came from online, and which of your pages those visitors are viewing. While the terminology may differ slightly from program to program, I’m going to give a basic overview of a few of these metrics.

When looking at analytics for your overall blog, most programs will tell you how many “visitors” and “unique visitors” you have. “Visitors” refers to the total number of times anyone has visited your blog, while “unique visitors” refers to how many specific individuals have visited your blog. For example, if one person visits your blog every day for a week, that will show up as seven visitors but only one unique visitor.

Similarly, most analytic programs will also give you information on “page views” and “unique page views.” “Page views” refers to the number of times that anyone has visited a specific page, or post, on your blog, while “unique page views” refers to how many specific individuals have visited that page. So if one person opens your website and clicks the refresh button a thousand times, you’ll have a thousand page views, but only one unique page view.

These metrics are useful for tracking overall traffic to your site, and you can use them to make editorial decisions regarding what you write about and how to write about it (or, for multimedia folks, what sorts of art or videos resonate with your audience). For example, if the metrics tell you that 50 percent of all the visits to your site were for a post you wrote about bioluminescence, you might want to write about bioluminescence again in the future. By the same token, if posts written in conversational language draw more visitors than posts that incorporate technical jargon, you may want to increase your efforts to write in casual language.

Two other metrics that can influence your editorial decision-making are “bounce rate” and “average time on page.” Bounce rate tells you what percentage of visitors clicked a link to go directly to a specific page on your site and then exited the site without viewing any of your other pages. Knowing this can be incredibly useful. For example, if there’s a page that stands out as having a low bounce rate, it’s worth taking a closer look. What distinguishes that page from your other posts? Did you include hyperlinks to other pages in the body of the post? Did you do a good job of highlighting related posts in a sidebar or “related stories” box? Figure out what you did that kept people on your site, and try to incorporate those techniques into future posts.

“Average time on page” tells you how much time, on average, your visitors are spending on a specific post. If people are spending minutes on the page, that means they’re probably reading it. But if people are spending only fifteen seconds on the page, you should investigate, because it means that something is scaring them off. As with bounce rate, see if you can figure out what is different about that page and adjust your overall strategy accordingly. For example, if posts that use technical jargon in the first sentence are scaring off readers in less than twenty seconds, it’s a clue that you need to refine your writing technique. After all, you (presumably) want people to read what you wrote—and you can always introduce the technical language further down in the post.

Analytic programs can also help you determine how you are reaching readers. This information goes by different names in different analytic programs, from “source” to “transitions” data. But all of it tells you how people arrived at a specific page on your blog. For example, it might indicate that 10 percent of the people coming to the page came from a Reddit post, 15 percent from Facebook, and so on. This sort of “traffic source” data is useful in two ways. First, it can tell you about activity related to your page that you didn’t know about. For example, I usually don’t find out that reporters have linked to my blog posts until I see it in my traffic source data. Seeing the stories that are related to my blog post can help me to evaluate the subject from a different angle—and so ultimately inform or inspire my future blog posts.

Second, and more importantly, traffic source data let me know which avenues are most effective for disseminating my blog posts. For example, if I find that the bulk of my readers are coming to my blog from Twitter, that tells me I need to maintain my presence on Twitter. Conversely, if I’m spending a lot of time and effort trying to push out my posts on Twitter, but getting very little traffic, I need either to change my approach or to focus my time and energy elsewhere.

Social media metrics

To determine whether you are making progress toward meeting your goals for the blog, you need to know whether you’re reaching your target audience. And as useful as blog-specific analytic tools are, they don’t tell you who your readers are. For clues to that, turn to social media.

I think Twitter is the most effective social media platform for disseminating blog posts. In part, this is because Twitter is home to a wide variety of discipline-specific networks in which people interested in a specific field share links to relevant material. This is particularly true within the sciences. While these networks are usually informal, they are very effective ways of reaching a lot of people with shared interests.

To begin tracking whom you reach on Twitter, simply click on the “notifications” tab. This will show you every tweet that mentions your Twitter handle, as well as any “retweets” (or shares) and “favorites.” But the notifications tab won’t tell you anything about tweets that link to your blog if those tweets don’t include your Twitter handle. Luckily there are other tools to help you do that. I’m particularly fond of services like Topsy.com that show you everyone who has shared a link to a given page. If you can see who is disseminating your work, you can get a good idea of whether you’re reaching your target audience.

Facebook is more problematic. It is undeniably useful for sharing information with friends and family. If you create a page for your blog, however, it is no longer reliably useful as a tool for sharing information with your followers, thanks to the vagaries of Facebook’s news feed algorithms. In particular, most Facebook pages are able to reach only a tiny fraction of their followers with any given post. Unless you have the budget to “promote” your Facebook posts, then, I recommend that you focus your social media efforts elsewhere.

Unconventional metrics

There are a lot of social media platforms out there, and dozens (if not hundreds) of social media metrics tools. In addition, both the platforms and the metrics tools are changing all the time. You have to stay current and be willing to experiment with these tools to find what works best for you. It also pays to remember that although the universe of metrics tools is evolving quickly, there are two tools that will always work: creativity and critical thinking.

Let me explain. As I mentioned earlier, to have a successful blog you first need to have a clear idea of what you’re trying to accomplish. Once you’ve set clear goals, you can bring critical thinking to bear and figure out what you need to measure to determine whether you’re making progress toward those goals. And once you know what you need to measure, you can use critical thinking (again) to figure out creative ways to measure it.

The conventional metrics I described earlier—analytics and social media—are useful. But developing your own unconventional metrics may be even more useful for determining whether you’ve made progress toward your specific goals.

I’ll give you an example. I work as a science writer at a university, and I often write for the university’s research blog. The primary goal of the blog is to raise the profile of the university in a positive way by highlighting the work done by faculty and students. A great way of doing that is to write stories for our blog that we think can be picked up by news media. We want people to visit our blog and read what we’ve written, but we know that stories that run in external news outlets will reach a much broader audience. As a result, while we do track visitors to our blog, an unconventional metric that we use to measure success is the number of external news stories that a blog post generates. We also track social media mentions of our posts, because they can serve as indicators of interest in the work we write about (because, presumably, people won’t share links to a post they don’t find interesting).

Another objective of the university’s research blog is to help our faculty and students achieve their own goals. Secondary goals like these often require their own unconventional metrics. A case in point would be a post I wrote in 2013 about a research assistant at the university who had recently earned his undergraduate degree.¹ The assistant had begun working on a product in his senior year and, after graduating, had teamed up with another inventor to bring the product to market. He and his partner had set up a Kickstarter page to fund an initial production run, and I was contacted about the work he was doing.

I wanted to write about the research assistant because it would be a good story for the university, highlighting student ingenuity and entrepreneurship. The research assistant wanted us to write about it because he wanted to draw attention to his fledgling company and generate interest in his company’s Kickstarter campaign.

I wrote the story, posted it on social media sites, and brought it to the attention of a few journalists whom I thought might be interested. And then I used two sets of metrics to determine whether the post was a “success.”

The post led to more than a dozen stories in external news outlets, including the largest newspaper in our metropolitan area. It was also shared more than a thousand times on social media. That means the university successfully highlighted student entrepreneurship to a large audience, and that the audience was extremely interested. In other words, the post accomplished the primary goal we had set for the blog.

Did we also achieve our secondary goal? Yes. And one of the unconventional metrics we used to determine whether we were successful was the amount of traffic we drove to the research assistant’s site. The assistant reported that there had been a huge jump in visitors, which was good—we had stirred up interest in his project. The second unconventional metric was activity on his project’s Kickstarter page. Donations jumped by more than $200,000 after our blog post (and the ensuing media coverage).

In short, the post did exactly what we hoped it would do: it raised the university’s profile in a positive way and helped a fellow employee or student achieve his or her goals. We had the unconventional metrics to prove it. Significantly, if we had relied only on conventional metrics, the post would have looked like a failure. The post received only 3,800 unique visitors and had a bounce rate of more than 93 percent.

The moral of the story is that conventional metrics are great, but don’t rely on them to tell you whether you’re reaching the goals you’ve set for yourself. It is worth taking the time to come up with metrics that give you specific, meaningful insights into whether you’re doing what you set out to do.

Matt Shipman is a science writer and public information officer at North Carolina State University. He writes the Communication Breakdown blog for SciLogs, and is the author of The Handbook for Science Public Information Officers (University of Chicago Press, 2015).

Matt is based in Raleigh, N.C. Find him at his blog, http://www.scilogs.com/communication_breakdown/author/shipman, or follow him on Twitter, @shiplives.

Note

1. Matt Shipman, “Science You Can Use: Engineer Designs Mug to Keep Coffee Temperature Just Right,” The Abstract (blog), North Carolina State University, December 11, 2013, https://news.ncsu.edu/2013/12/wms-hot-coffee.