Tracking content marketing
Hello content marketer.
I need you to ask yourself some difficult questions today. I know you think that because your site has tracking installed that you're data-driven and make insightful guesses based on interaction information.
But do you? How much do you really know about how your content is being consumed? You might have services like Google Analytics, Mixpanel, or Kissmetrics doing a lot of grunt work for you. But they work in another department. They report to sales, not marketing. They're measuring funnel data for the boss. Content marketing friends, you're invited to the party but you gotta bring your own booze.
Those tools are incredible. They come out the box, batteries included, designed to help you collect tons of data and tear it apart. They're designed to help you filter through it and segment it and visualize it. Their plug-and-play nature is perfect for this type of collection.
But the problem, and yes you do have a problem, is that these platforms aren't really collecting content consumption data. They're collecting the behavioural data - the actions based on your content. Popular analytics software tracks clicks and scrolls but it doesn't tell you anything about how specific sentences and paragraphs are consumed. They're tracking data at a level of abstraction away from where they need to be collecting the data.
That's a pretty broad accusation. I'll concede that video companies like Wistia can show you exactly which parts of your video each viewer has seen. That's so beautiful. Isn't it strange that this feature isn't standard for tracking text?
Mixpanel does have a feature called autotrack that comes very close. The platform has this cool autotrack editor that lets you add event tracking without any code. No code means widespread adoption which is excellent. Once you set it up, the tracking for your new event will even provide you with older data prior to your event's definition. Magical historical data.
Don't take my word for it
Josh Schwarz said it best. “Traffic measures are useful for understanding the scale of a site, but they are not meaningful for the evaluation of content”. Preach, brother.
Measuring your content's engagement rates means you can infer whether it’s providing any value. Value is determined from a range of signals and the pageviews metric isn't one of these.
Scalar values aren't able to provide you the data depth you need to build insights. Your Google Analytics account can tell you average session times and page read times for your site, but Google Analytics metrics aren't the right metrics for tracking content engagement. Not without custom events anyway. GA metrics track time on page and time on site by calculating the differences in timestamps of page loads. It has no concept of whether someone is consuming your content or whether it’s sitting in inactive tab number 34 on your browser.
Inflated numbers lead to poor decision making and can have actual monetary implications. Implementing tracking that measures actual engagement time will mean accurate data and a better feedback loop for your content strategists.
In case that’s all a bit confusing, look at the 2 counters below. They both started counting the moment you loaded this guide. The left one will count until you close this page. The right one will pause if this page becomes inactive. Try opening a new tab and coming back a few seconds later.
Now that you've tabbed back to this page, notice the difference? The right counter is more accurate about your time spent on this article. I get it, you might think that a few seconds in time difference might not mean much. But if 5,000 visitors read your content while browsing around dozens of tabs, that difference can significantly warp true read times.
Below are some more counters. The first one will only tick over while its orange container is visible to you. If it's off-screen, there's no point in tracking your engagement with an element you can't access. This is significantly more accurate than either of the previous two timers.
The final timer above started counting the first time you visited this page. It ticks only when you're on this page.When you close your tab or browser window and come back later, it'll resume from where it left off instead of starting from scratch.
What if you decide to leave this page open while grabbing a snack from the kitchen mid-article? You might come back 5 minutes later, resume reading and finish a bit later than you would have otherwise.
Google Analytics will report yet another inflated time on page figure. That's because it wasn't designed for granular engagement tracking. But with the right kind of thinking, you're able to cover cases like the kitchen snack conundrum, without too much effort.
The examples above use a level of granularity detailed enough to warrant being called content marketing engagement metrics. Knowing how much attention people give to certain parts of your writing can guide you writing more interesting content. Knowing who's reading what can help you segment people and understand exactly what works for different parts of your funnel.
That's content marketing tracking. As in tracking the content. As in not blindly deciding what to publish next. As in you've got real world numbers to corroborate or disprove the demand of topic X for power users or guide about Y for first-time users.
I'm completely certain that there are many people out there that share my views. I'm also certain that a portion of you folks have written and deployed your own custom code to track this kind of stuff. Hit me up on Twitter and tell me about your implementation. I'm hungry to learn what neat tracking tricks you've rolled out?
49% of the B2B crowd think measuring content effectiveness is their biggest content marketing challenge. Maybe more of the internet's meta-marketing needs to delve deeper into measurement strategies. You can read a bit more about content marketing analytics in my guest post on the Mention blog.