Part VI: ANALYTICS
The thrust of analytics isn't data; it's improvement. Improvement just starts with data because you can't improve what you don't measure.
But measurement alone isn't going to lead to improvement—or any action at all. The road to business hell is paved with good data.
The key is marrying sound measurement methodologies with thorough analysis and tying all of that glorious data and those keen insights back to an organization's business objectives.
So Part VI focuses on both prongs of that fork: setting up reporting systems that will collect the data for analysis at a later juncture, and honing in on the truly actionable insights.
Having started off college as a biochemistry major, I still have a strong inclination toward scientific processes. So what we're going to do in this intro is look at a handful of marketing conundrums currently being mulled over by entirely fictional people and businesses. Hey, it's the closest I could come to forming a hypothesis to test in a Petri dish! We'll dissect the process of how to separate the signal from the noise in data sets accessible to marketers.
Ecommerce Budget Allocation Decision
Anita has an ecommerce shoe store and a handful of trusted drop-shipping companies that keep her supplied with the latest fashions. Her cousin Monica took a social media strategies class in college, and at Monica's urging, she signed the store up for Facebook, Twitter, and Pinterest and has hired a couple of college students to work on building up a following for the store. She's experienced moderate success and has just over 3,000 followers on Twitter, just under 700 Facebook likes, and 88 Pinterest followers (but has only been pinning a few months). Anita also has a well-vetted email list that she uses to announce her weekly specials.
Monica says that email is dying, and that Anita needs to invest more time and effort into social media or her shoe store will eventually die. Her argument is that no one under 40 uses email for shopping. She's a judgmental little juggernaut of persuasion, huh?
Burning Question
Anita wants to know if she's focusing too much on email and if she could make more money by shifting her priorities from email production and distribution to social media.
Data Needs
Anita hawks over the metrics she gets from MailChimp like it's her job (because, well, it kinda is). She also has lists segmented and stays on top of keeping them scrubbed. Her problem is she doesn't know what those visitors do when they get to the site, although she's tried to figure that out using Google Analytics (GA). But, from what she can tell, social media doesn't seem to be generating much revenue for her.
Analysis
Well, for starters, Anita doesn't have the data she needs to make any decisions at this time about social media or email. She would need to start tagging her emails with campaign parameters (covered in this Part of the book). But then she also needs to understand that GA, by default, uses last-click attribution in crediting conversions, so the ecommerce reports she's been downloading from GA are only giving her part of the story. It's kind of like writing a movie review for The Sixth Sense without getting to the end. Sort of. Okay, not really. But wow, what a movie!
Furthermore, upon further investigation, we realize that Anita is only including referrals from twitter.com, facebook.com, and pinterest.com in the social media report she created and has emailed to herself every week. We need to inform her that Twitter traffic especially comes from many sources, not just twitter.com, and introduce her to GA's robust social reports.
Finally, because she's not tagging links in her emails that point back to her site, much of her traffic from email is getting dumped into the direct bucket. She doesn't know that desktop email apps, like Outlook and Mac Mail, don't pass referral data. As long as we're talking about depressing GA limitations, webmail providers that use a secure (https) server by default, like Gmail, don't pass referral data either.
Go ahead. Kick something, Anita. You'll feel better.
That said, Anita definitely needs to read Part VI before making decisions that could hurt her profit margins.
Paul's Grief Over His Lack of Blogging Success
Paul runs a pet loss blog. He has a counseling degree and enjoys helping people who have experienced the loss of a beloved pet. In conjunction with his blogging, he's very active on Twitter, with 12,400 followers. He follows everyone except for bots back because he wants people to feel comfortable DMing him with questions.
Paul uses bitly to shorten his URLs and he monitors clicks. About twice a week, he has tweets that go viral, but he hasn't been able to figure out the secret sauce of what makes one tweet more retweetable and reply-worthy the others.
But Paul's real pain point is that he feels like a failure as a blogger. He's read that bounce rate is the golden metric (hat tip to Avinash Kaushik for that one!). Like a good Avinash disciple, he monitors his bounce rate very closely, along with visits. But his bounce rate is consistently north of 60 percent. He's tried everything to get it under control: videos, a couple of infographics, a “related post” WordPress plugin . . . all to no avail.
Convinced that he just can't cut muster as a blogger, he's considering selling the pet loss blog and starting a blog about his other passion: kiteboarding.
Burning Question
Paul wants to know what more he could do to make his blog “stickier” so people will stick around longer. Secondarily, he wants to know if he's being strategic enough in his use of Twitter.
Data Needs
Paul wants to know if there are GA reports he can use to diagnose his fatal unattraction. He also is at a loss for how to quantify his own Twitter trends to figure out what captures the attention and curiosity of his followers.
Analysis
Oh boy . . . There's just so much wrong with this picture. First of all, Paul is making a mistake that's very common among bloggers: obsessing over a single metric to measure his success. It just so happens that for his case, he couldn't have chosen a worse one to obsess over.
Bounce rate may be the golden metric in many cases—because it helps you identify pages on your site that visitors just don't grok on— but it's a terrible metric to measure the success of a blog. There are a couple of reasons for that:
• Blogs are notorious for high bounce rates. People typically land on a blog post to read that post. If your site offers more than just a blog, you can craft a strategic call-to-action on your blog posts that may woo readers deeper into the site. But Paul's site is just a blog—a place for people to read insights about dealing with the loss of a pet and share their stories. Paul needs to understand this phenomenon about blogs and not hawk over this metric.
• Even if a visitor comes to your blog and reads for 20 minutes, when s/he leaves it will be considered a bounce. And, to add insult to injury, that visit won't add to the average time on site because a visitor has to click on a second page to register how long the time on page (and on site) is. You could get fancy with event tracking and trigger events based on time on page—and even time on a page in the front tab of a browser using the Visibility.js library—but that's way beyond the scope of this book. In Paul's reports, every bounce is going to show up as 0 seconds time on page/site and skew those metrics as well.
Applying the “Non-Bounce Visits” advanced segment that comes baked into GA will help Paul cut down some of the weeds, but there are some other metrics he should be focusing on to round out his perspective anyway. For example, Paul should monitor the number of comments he gets per post to measure engagement. He could track this in GA by attaching an onSubmit event to the form in his comments.php file since an enterprise tracking system like Radian6 is out of range for him. An analytics aficionado would be able to help him set this up.
Another suggestion I'd make to Paul is that he should measure “microconversions” like newsletter signups and downloads of his eBook using event tracking and goal conversion tracking.
And since he gets a decent number of shares and likes and because GA can now track social interactions, I'd recommend that he use either the AddThis or ShareThis plugin on his site.
Finally, I'd recommend that Paul take a look at his Twitter use and try to ascertain what makes some of his tweets go more viral than others. Is it the length? Subject matter? Resources shared in the tweets (he shares a lot with his followers)? Rand Fishkin shares some great techniques for how you can use the Twitter and bitly APIs to get this data in Part VI.
You'll learn a lot about how to set some of these things up in this section. Half the battle is nailing down the proper metrics and key performance indicators (KPIs) to monitor.
Naoki's Obsession with Being #1
Naoki runs a popular sushi bar in Philadelphia. He's hired an SEO consultant to help him get to the first position in Google for sushi in local searches but to no avail. He thinks his consultant is a snake oil salesman because it's been nine months (and eight days), and all he's been able to accomplish is getting Naoki listed in the seven pack of local results. But Naoki has to scroll to even see the local results when he searches for sushi. And freaking Wikipedia is at the top. Because people want to know the history of sushi when they search, I'm sure. Pfft.
Burning Question
Can he justify paying his SEO consultant when he can't get him to the top of Google Page 1?
Data Needs
Naoki needs perspective. And new KPIs. At minimum, he should create a custom report to track visits from Philadelphia (and some of its surrounding metro areas, if applicable) that's filtered by ‘Medium' Matches exactly ‘organic'.
Analysis
First of all, Dr. Pete did an interesting eye tracking study (included in this section) that shows that visitors tended to gravitate toward the local one box, even when it was farther down the page. If someone's searching for sushi because he's hungry and totally jonesing for a California roll, he's not going to click on the results at the top of the page that lean more informational in intent. He's also not going to care about images of other people enjoying sushi. He's going to make a beeline for the local box and focus most on the result at the top of that listing and/or the restaurant that's closest to his growling tummy.
I'd encourage Naoki to set up some custom alerts in GA for when organic visits increase or decrease by more than 10 percent (or whatever threshold he wants to set).
Furthermore, Naoki really needs to connect his Google Webmaster Tools account with his GA account to monitor impressions for the main keywords he's tracking (covered in this section). He can also use his Google Places Dashboard to monitor the queries his site is showing up for.
And before jumping to conclusions at the first sign of a dip, I'd encourage Naoki to check Google Insights for Search (now Google Trends . . . again) to see if the ebb and flow he sees in impressions correlates with the natural cyclical pattern of searches for sushi-related terms. He'll learn interesting things, like people just aren't as into sushi in May and November. Who knew?
At the end of the day, if Naoki's consultant took him from invisible to prominently placed in a local box, he should be thrilled. He may not be able to track conversions without some call tracking in place (which is an expense most small businesses don't assume), but he can certainly take these insights and reassess if he's getting enough value for the money he's spending on SEO.
To Shoot or Not to Shoot; That Is the Question
Amy has a wine glass painting business that has really taken off. She started teaching classes locally, in her San Francisco shop, on Sunday evenings after it closes. Most of her revenue comes from online orders, so she figured she'd connect with her community by teaching them something she enjoys doing anyway. But she's been encouraged by some of her Facebook fans to create videos that she can post on her site. Being an artist and a perfectionist, she can't bring herself to just have her husband record her painting wine glasses on his phone. She wants professional-quality videos on her site.
Burning Question
Can the cost of creating videos be justified?
Data Needs
Amy needs to see data that will give her some assurance that creating and hosting videos on her site will bring her increased traffic and conversions.
Analysis
This is a tough one, because without creating and hosting at least one video, she has no data to suggest whether adding these videos to her site will increase her traffic and conversions. However, there is anecdotal evidence that Amy could use to help her inch her way closer to a decision.
For starters, in Dr. Pete's eye-tracking study (see Chapter 24), he showed that video-rich snippets are hotspots. That should be a focal point for Amy's cost-benefit analysis. How can she increase her chances of getting the highly coveted rich snippet for the queries she decides to target, if she were to move forward with this?
There are keywords she can optimize for that may increase her chances of getting a rich snippet like how to, tutorial, and video. She could also use a third-party video hosting service, like Wistia, for even higher chances of getting the snippet. Google shows preferential treatment toward YouTube pages over pages that merely have a YouTube video embedded in them.
She would also want to create dedicated landing pages for each of her videos, and track them in GA to see what kind of love they get over time. If she put all of these pages in a specific directory, creating a segment for these pages would be a piece of cake.
She would also benefit from multi-channel funnels to see if visits from her Facebook fans are resulting in conversions at some point. The cookie only lives for 30 days, so these reports are hamstrung by that limitation. But she could gain some valuable insights about the impact of her Facebook sharing. Multi-channel funnels are covered more (you guessed it) in Part VI.
Hopefully, as you read this Part—and all the great ideas and insights contained herein—you'll think about how real people with real businesses and projects can apply these strategies to garner real results!