More Data Analytics - a Deep Dive into The "How-To" Gun Girl Traffic and Engagement

More Data Analytics - a Deep Dive into The "How-To" Gun Girl Traffic and Engagement

I am into data.

Now that we’ve got that covered, I want to share what I learned since my last analysis of The “How-To” Gun Girl.

But first, let me explain why I did this. Over a year ago I started this blog with the goal of providing information that I personally found difficult to dig up online. For example, what is a hunting tag and what does it look like? How do I set up a compound bow? Or if I am interested in buying a gun, what do I need to do? 

Beyond providing tips and info for beginners, I also wanted a place to connect and share more technical information with experienced shooters and hunters. There aren't many sites out there that provide content about how to shoot Olympic smallbore and air rifle or getting into biathlon. With 20+ years of competitive rifle experience and 10+ years as a coach, I can provide that kind of content.

So I started this blog. And along the way, as I learned a little about websites, I developed another goal - to build a personal brand. For both personal and professional networking purposes, having my own website is really helpful. It serves both as a resume and a portfolio of my work.

With these three goals - 1. Info for beginners, 2. Deep dives into more advanced shooting and hunting topics, and 3. A home on the web for networking purposes - The “How-To” Gun Girl developed into what it is now. While I like where I am, there is always room for improvement. And in my engineering experience the best way to improve is to analyze current state and best practices, and then figure out how to get from where I am to where I want to be.

In this video, I share a brief overview of my results. I made this presentation for the final project in my Data Visualization course as part of Duke University's Masters in Engineering Management.

In this video I analyze blog content and traffic at my website, www.howtogungirl.com using Gephi, Tableau, Voyant, and Excel. Data comes from my blog as well as Google Analytics and Facebook via NetVizz. This was a presentation for the Data Visualization course in Duke University's Master of Engineering Management program.

Blogging Research

I read several blog posts and listened to a lot of podcasts as I was trying to determine best practices for managing my website and social media. If this is something you are interested in, I really enjoyed these podcasts:

The executive summary of what I learned is this: 

Facebook and social media in general doesn't matter nearly as much as I thought it did.

Facebook Analysis

My goal with social media usage was to connect with my audience and share with them not just my blog content but a lot of other photos and videos as well. According to the experts, however, my sharing of that content on Facebook rather than on my own site means that I am giving my best content and effort to a platform that I don't own, and that might not use it well. The moral of the story is “don't build a house on rented land.” Someone else’s platform (Facebook) is not mine. They make the rules, they share the posts. This graph illustrates the limitations of Facebook.

As the number of businesses using Facebook has increased, "organic" (non-paid) post reach has diminished. To keep your posts in front of your followers, Facebook encourages pages to pay to "boost" their posts, and most pages are doing so. ("How-To" Gun Girl image)

As the number of businesses using Facebook has increased, "organic" (non-paid) post reach has diminished. To keep your posts in front of your followers, Facebook encourages pages to pay to "boost" their posts, and most pages are doing so. ("How-To" Gun Girl image)

As more users and more businesses sign on to use Facebook, Facebook shares each page’s posts with fewer and fewer people. I can understand this - it is their product and they want to keep content fresh for their users. But as I learned in my previous analysis of my Facebook posts from the calendar year 2017, less than half of my Facebook posts had any engagement (likes, shares, comments). None at all. That result is directly opposed to the goal of using my platform to share knowledge and to connect with others. And it is a waste of valuable time to post something that doesn't engage people! So after I finished my Facebook page analysis using Gephi (an open sourced network analysis software), I focused the rest of the analysis solely on The "How-To" Gun Girl. 

Blogging Analysis - Keywords

My hypothesis is that a best practice for blogs is to have good traffic and engaging posts. According to Michael Hyatt's "Killer Blog Post" podcast, the best posts are something that "people want to read and share." So I directed the first part of my analysis toward keywords in order to figure out what people wanted to read. I downloaded the content of this site (by exporting it on SquareSpace as an XML, then converting that to a text file that I then removed the HTML code from). I took that text file and uploaded it to Voyant, a (very cool) text analysis tool. From there, I downloaded the ranked list of words used most often on my site, a file that looked something like this:

The Voyant Tools word rank results from The "How-To" Gun Girl site export shows the count and rank of the most commonly used words on this blog. (The "How-To" Gun Girl image)

The Voyant Tools word rank results from The "How-To" Gun Girl site export shows the count and rank of the most commonly used words on this blog. (The "How-To" Gun Girl image)

Voyant lets you download the results into a csv format, and I took that and filtered it some more to remove common words like "like" and "now". Then I uploaded the results into Tableau and set up a word cloud to get a better visualization on my most-used keywords. The bigger the size of the word, the more often it is used. Side note - isn't it neat how looking at the data this way gives you a different perspective than looking at a ranked list of words? I used a word cloud to come up with the title of this blog, too.

I created this word cloud using Tableau. It shows by size the frequency of terms used on my blog. Terms like "shooting, archery, rifle, Olympic, and hunting show up pretty frequently. (The "How-To" Gun Girl image)

I created this word cloud using Tableau. It shows by size the frequency of terms used on my blog. Terms like "shooting, archery, rifle, Olympic, and hunting show up pretty frequently. (The "How-To" Gun Girl image)

Using Google AdWords' Keyword Tool, I searched for each of the top 15 words used on the blog to get an idea of how frequently those terms are searched each month. One word of warning - since signing up for AdWords to use the Keyword Tool, I've gotten an email about every two days asking me to complete my first ad. I'm not interested in advertising, but it is still a helpful tool for figuring out if the terms I'm using are things that people care about. 

Now back to the analysis. The Keyword Tool showed me that of my top 15 words, the most popular Google searches are "Olympic" or "Olympics" at over 3 million searches a month. Most of the terms in my top 15 words are over 1 million searches a month. The popularity of these terms might on the surface mean that I should have a lot of people finding my page from Google. But, I don't think that's the case. I will show my traffic analysis in a bit, but I am definitely not getting close to a fraction of the people who search for my top-15 terms to click through to my site. One theory for that is that my keywords are not specific enough to attract the niche audience I write for, so my posts are getting lost amidst the thousands of other website writing about these same terms. 

To visualize the AdWords Keywords Tool results versus my word use results, I created a tree map in Tableau. In this visualization, the red-orange color spectrum is used less frequently on this site. The blue spectrum are the most frequently used words on this site. The larger box sizes are the most-searched for terms on Google. Note that I do not have a match between the terms I use most frequently and the terms that are most popular on Google. Once I figure out some more narrow, specific niche keywords to use, I hope to get more of a match between my blog and Google searches.

This treemap is sized by Google's Keyword Tool (number of monthly searches for that term) and colored by the frequency of use on www.howtogungirl.com. Note that the most popular terms on Google (largest box size) are not the terms used most frequently on this site (blue color). (The "How-To" Gun Girl image)

This treemap is sized by Google's Keyword Tool (number of monthly searches for that term) and colored by the frequency of use on www.howtogungirl.com. Note that the most popular terms on Google (largest box size) are not the terms used most frequently on this site (blue color). (The "How-To" Gun Girl image)

Blogging Analysis - Traffic

The second metric to keep track of after keywords is traffic. Keywords matter for getting the right people to the site. I want to know how they found this site and what they do when they are here (click around, read something for a long time, etc.).

The best measures for traffic appear to be users and bounce rate. Users refers to the number of individuals visiting my site. Bounce rate is the percent of those users' sessions that ended after just one page was visited. An ideal state would be a high number of users and a low bounce rate, because that would mean a lot of people are viewing my site and they are finding the content engaging enough to click through to other pages.

This data from Google Analytics shows the number of unique users visiting my page from January through mid-April 2017. (The "How-To" Gun Girl image)

This data from Google Analytics shows the number of unique users visiting my page from January through mid-April 2017. (The "How-To" Gun Girl image)

Google Analytics provides me with the data for traffic analysis. It anonymously tracks how many people (users) visit www.howtogungirl.com. Google Analytics is a pretty powerful tool with a lot of information to dig through. For example, I know that when I pulled the data there were 819 individual users who visited, spending an average of about 1 minute on the site. Of those users, only about 18% are repeat visitors. That is something I'd like to work on, because it seems that a more engaged reader is one who comes back for updated content. 

Another way to look at the data was to break down the sessions into how they found my site and then look at what they did once they got here. The graph above shows four channels and a count of how many sessions came from each of those channels. 

Using Google Analytics I was able to look at the channels directing users to www.howtogungirl.com and how those channels differed in bounce rate. (The "How-To" Gun Girl image)

Using Google Analytics I was able to look at the channels directing users to www.howtogungirl.com and how those channels differed in bounce rate. (The "How-To" Gun Girl image)

Organic Search refers to Google or other search engines directing a user to my site. Social includes Facebook, Instagram, and Pinterest. Direct refers to people who type in my website to go directly to it, and Referral is the category for everyone who gets to my site by clicking a link on another site that isn’t included in social.

The data are color-filtered by bounce rate. Remember that bounce rate refers to the percentage of users who visit only one page on my site and leave. Red is 62% and dark blue is 84%. This graph illustrates the power of a referrals and direct visits to my site – about a third of those sessions click to another page on my site. While I get more new users through social networks and organic searches, about 80% of those sessions are limited to only one page on the blog. Bounce rate is kind of like a measure of engagement. The longer someone stays on the site, and the more pages they click through, the more engaged they are in the content.

Additionally, I learned from the Digital Marketing Podcast and Marketing School that I need to do a better job directing people once they get on my site. If I can help readers find the next topic they are interested in reading about, they will be less likely to "bounce" and leave. Google Analytics provides some insights with this topic, too. The image below shows that lots of users exit after my home page, so that is something that I need to correct.

This image shows some of the terminal pages on my site, as well as where users flow after reading a specific page. (The "How-To" Gun Girl image)

This image shows some of the terminal pages on my site, as well as where users flow after reading a specific page. (The "How-To" Gun Girl image)

So What's Next?

So, what’s next? I’ve already got some next steps in place. For one, I redesigned my home page to include some interesting images for each of the categories I write about – target shooting, rifles, archery, and hunting. I also tried to direct users into three categories and include a lot of links for them to click through, depending on their interests. Hopefully this will address the problem of users terminating their visit on the home page. As of mid-April 2017, the home page looks like the image below.

The new look for the home page of The "How-To" Gun Girl as of mid-April 2017. (The "How-To" Gun Girl image)

The new look for the home page of The "How-To" Gun Girl as of mid-April 2017. (The "How-To" Gun Girl image)

I am also in the process of changing my keywords that I use for titling and organizing my blogs. I want to make sure I am hitting on popular terms for the audience I want to write for. One example is the term “hunting gear”. Google AdWords told me that Hunting gear and a few variations account for about 352,000 searches monthly. I consolidated my gear reviews under one keyword – “hunting gear”. This will be a long and iterative process to try to capture the audience I want, but I’ve got to start somewhere. 

One final thing that I will make an attempt to do is to post at a regular interval, so that readers know when to expect a post. This makes total sense - it would be good to be predictable so that users build up a habit of visiting my page at a certain time during their week. 

If you have any advice or feedback on the new site, please get in touch. I appreciate each of you that read this and reach out.

4 Types of Goals to Help You Shoot Better

4 Types of Goals to Help You Shoot Better

How I Got Two Free Turkey Calls

How I Got Two Free Turkey Calls