Move Over Content, Data Is Now King!

This may be somewhat controversial, but I’m fast coming to the conclusion that content is no longer king. Some of my peers and other industry professionals may not agree with me, but it’s a hypothesis I’ve gradually been warming to over the last few months.

Before delving too much deeper, I should clarify that I’m not saying content is dead, SEO is dead, or any other ever-present marketing discipline is finished. Content is the lifeblood of what marketers do everyday. On social media, on websites, on blogs, even offline. If you’re not creating regular, engaging content, it’s likely your organisation will find the next few years increasingly difficult - particularly online.

However, there’s one issue around this. I’m completely sold on the idea of content marketing, but what truly drives me insane is that I can’t remember the last marketing conference where one of the speakers didn’t said something like;

“create great content to stand out”

That’s all well and good, but it’s not particularly useful, particularly to the one-man business operation or novice marketer. Why? Because it doesn’t actually tell you how to create great content. The reason they don’t tell you is, in truth, because nobody really knows - there’s no single, right answer. Great, engaging, shareable content is often created by larger organisations with huge budgets - for example, Coca-Cola’s Share a Coke campaign. Other times smaller organisations have a strategy that consistently produces sharable content - but even they can’t guarantee a slam dunk every single time (incidentally, the link to the above Movoto blog only works when the parameters above are added - seems they’ve removed it from their site!).

The Movoto article is a very, very good read. The reason for this is because they go into really granular detail about how they go about their business. What worked, what didn’t, how they could improve. It’s all in there. They made a relatively “boring” subject like Real Estate work for them. But one of the key things that drove their success was their analysis of the information and data available.

We have for some time now been living in the “information age”. In this information age, we have a wealth of data available to us, some of this information is so vast that the applications that used to process this information are unable to keep up - and thus a new term - “Big Data” - was coined.

Infact, in Movoto’s article they put in capital letters;

RELY ON DATA, NOT WHAT SEEMS POPULAR

However, it’s important to appreciate that whilst you should rely on data to inform your decisions, you should consider where this data comes from, and what sources you should use to collect, and analyse, this information.

What data should I use?

So, now we have a (relatively) compelling case for why data is more important than the content, let’s start looking at some stats that are easy to get hold of - your own website.

Google Analytics

Google Analytics has a few useful reports that show you what content appears to be working well on your site. Behaviour reports in Google Analytics will help you find optimisation opportunities. You’ll find Behaviour reports on the left hand side menu;

If you navigate to All Pages (as above), by default you’ll get a list of your most popular content on your website listed by page views;

This is a simple way to see what type of content is working well. You’ll typically see your homepage at the top, so it’s worth using the filtering certain sections so you have a better understanding of what content is working well on your blog or news section;  

You can also choose which metrics to sort the list of pages by - this is useful to see how long people are on average spending on your pages, and the bounce rate (the percentage of users who visit one page of your site and then leave (“bounce” away) of your content.

The Content Drilldown section will break your website content down by subfolders - helping you establish which categories of content are the most popular, and this is especially useful for deciding which topics to focus on. Again, you can choose a number of filtering or sorting options to further dissect your data. You could also do this using Content Grouping - but that’s best left to another tutorial.

Other useful reports in Google Analytics include the Landing Pages report, and the Exit Pages report. Both of these are relatively self explanatory, but show the top entry and exit pages for the site - enabling you to either work out the best and worst performing pages. You can use secondary dimensions or segments to break this data down by channel. Why is this useful? Well now you can see whether certain articles, or content types, work better in driving traffic from social media, from referral sites, organic search, or from another channel.

Advanced Google Analytics Implementations

There is more information available in Google Analytics, but to access it requires further implementation than just inserting the tracking code. Things like scroll tracking and adjusted bounce rate are good examples of methods that help you determine how engaging a piece of content is. For example, a page view tells you someone has viewed the page, but what if they don’t read the content? Scroll tracking helps determine how far a user has scrolled down the page - letting you determine whether or not your content is valuable - beyond just the standard page view.

It’s clear here that just the content alone isn’t enough to tell you what’s worked - you need that insight - preferably with as much detail as possible - to tell you what’s working and what hasn’t. That granularity of data is vital to improving, or starting, your content generation process.

In addition to Google Analytics there are a few other tools that are worth exploring.

Crazy Egg

Crazy Egg lets you see where people are clicking on your site. You can do this with Google Analytics (and other web analytics platforms) but it does require custom implementation. With Crazy Egg, you just install the code create a “snapshot” of your page, set the time frame you’d like to see data for, and then wait for the information to come in. The data you get is really interesting. I’ll give an example from the old Klood software website.

The site had three buttons underneath the main hero banner and sign up form. When user would visit, they would see the below screen;

They could then hover over one of the other buttons, and on-click see some information about the scheduler, or reporting function;

We implemented Crazy Egg and got some interesting results;

Nobody was clicking on the scheduler or reports links! This meant our carefully crafted content was being wasted, and the user was getting a frustrating experience because they were clicking on something, expecting something would happen, but nothing occurred. Without a tool like Crazy Egg, we could never visualise this, and look to rectify the problem. This is another good example of how important data is when informing decisions around content - in this case, nobody was finding out content because of a (in all honesty) poor user experience.

Buzzsumo

This tool is one of my favourite tools at the moment - it’s been around a while now, and has been picking up mentions from well known speakers like Rand Fishkin and Paddy Moogan. What is Buzzsumo? Well it’s best to get it from the horse’s mouth;

“Analyze what content performs best for any topic or competitor”

Buzzsumo allows you to input either a topic, or a domain name, and will then pull in articles, and list them by the total shares across social media platforms.

Imagine I’m a real estate agent in the UK, and I’ve heard about Movoto’s success online and want to replicate their content marketing machine this side of the pond. All I need to do is throw their domain name into Buzzsumo to see what content is working well on their site;

Instantly I can see that of all the content on Movoto’s website, four of the top five are “listicles” (articles that are presented in a numbered format). It’s clear then that my content strategy should incorporate a large number of articles that are similar in this regard. On the free version of Buzzsumo, you only get the top five results, hence why I’ve only shown five, but if I show you the next five, there’s a pretty clear pattern emerging that confirms the above hypothesis.

If I go even further, out of the top 20 articles on Movoto, 18 of them are listicles! Now I would never of known this information without digging deeper. I could of guessed at the type of content that would work well, but tools like Buzzsumo really help pull out data that can confirm my assumptions.

Wistia

Wistia is a video hosting platform but whilst YouTube focusses on consumers, Wistia is very much targeted at professional organisations. Wistia essentially allows you to embed videos onto your website much like YouTube, but it has many advantages such as calls to action, email capture and is more SEO friendly than YouTube - but I won’t delve into here. The important thing to note is that the data you can pull is exceedingly useful.

Wista provides top level statistics like average amount of video watched, total plays and the percentage of people who watched the video - YouTube provides these statistics too, but Wista has an distinct advantage from an SEO perspective if you’re looking to have videos on your site. This data can help you decide whether the content in the video is engaging enough, or whether the video should be moved further up or down the page (or replicated) to increase engagement with the video itself.

It also provides heat maps that show a single session of each video. An example can be found below;

Image courtesy of http://wistia.com/doc/embedded-video-analytics

This heatmap gives a variety of information, that can be integral in spotting trends, as well as optimising the content itself. For example, if you see multiple users watching the same section of the video repeatedly, does that section need rewording, or re-explaining so that it’s easier to digest?

Furthermore, it has integrations with marketing automation platforms like Hubspot, that will show you whether or not a particular person has viewed a video or not.

image courtesy of http://wistia.com/product/hubspot

An example of how you could use that data is that if the same person has watched a video multiple times, or watched several videos, are they right to reach out as a potential customer?

All this kind of information is invaluable to marketers - and is arguably more important that the content in the videos themselves!

Huballin

Editor's note: Huballin unfortunately shut down in April 2016. They list a few alternatives on their website.

If Buzzsumo is one of my favourite tools, then Huballin has got to be up there too. Huballin is a great place for you to go if you’re struggling for inspiration on what blog content you could be creating. Unfortunately, it’s still in beta, but the insights you can get from the tool are still pretty impressive.

Now you may be asking how this links into data? Well it’s relatively simple. Content marketing is brilliant, and undoubtedly works - but what’s the point in creating all that content if no one cares about it? That’s where Huballin comes in. The premise is simple - enter a topic, Huballin then suggests a few more topics that are associated with your original entry. You can then choose to select a few more, or keep your original query. Huballin then throws back a list of potential article titles. Not only that, but it gives you the average number of search queries per month for the given keyword as well as the average CPC from Google AdWords. Let’s take the example “coffee”.

I type in the word “coffee” and hit return;

Huballin suggests a bunch of related topics (or you can add your own). You then select the ones you think are applicable and click continue;

Huballin then gives a list of related queries, with their search volume information and their CPC too;

You can then sort the information by search volume or CPC. As you can see from the above result, what you’ll get back won’t be perfect, but it’s a very good start.

It’s important to note here that what Huballin is trying to do is bridge the gap between blog title generation tools (like Hubspot’s or Portent’s), and the data that will help you determine whether the blog title has enough interest behind it to potentially be a success. It’s also important to remember that Huballin is still in beta - so it won’t be perfect. But what they’re trying to do is exactly what Google is trying to do - understand the intent of a user’s query, not just serve a list of results that relate to the phrase entered.

Why data is so important

I’ve really just begun to touch the surface of the subject of data. I’m not trying to say that it’s content vs. data - my point is that if your content generation session ends with the sole hypothesis of “I think that this will work”, you’ll be incredibly lucky if it does. Most people are not Coca-Cola or Pepsi. We don’t have the time or resource to sit there and digest whether each article has worked, commence a full debrief for every campaign, and then go again next time implementing a refined strategy that involves spending thousands on content promotion. We need to use the limited resources we have available. Tools like Buzzsumo and Huballin are excellent at doing that. They give us insights very quickly, that allow us to make informed decisions, and by doing that, you really are mitigating the risk of your next content marketing campaign being a failure.