Creating a Data-Driven Content Strategy

As more and more brands become publishers, it becomes more and more salient to become data-driven. Let’s talk about what that means.

By now, most people are sick of hearing the term data being thrown around like a buzzword. For many years, it’s been all the rage to ask any content marketing team “how do YOU use data to drive your content strategy?” but at the end of the day, most people don’t really understand what it means.

What does data-driven mean, then? First, let’s look at some of the dimensions of data.

Structured vs. Unstructured Data

Data isn’t just one thing it’s many things. There is the structured data that most people understand. This includes visitors on a piece of content, time on page, bounce rate. It also includes social engagements such as number of shares, likes, and comments. Some of the more traditional media measurements like impressions or reach are also examples of structured data. Even sentiment, which is a more inexact version, is structured. Structured data refers to raw numbers and measurements that are mostly unbiased. It’s the what and sometimes how of the data world.

The tougher stuff to read is the unstructured data. Unstructured data asks the why of the engagement. There are many tools for gathering, but aren’t many automated tools for analyzing unstructured data and those that exist are missing the point. To be unstructured means that there are no pre-defined data models. Some examples include: transcripts from interviews, written or spoken customer feedback and reviews, email messages, videos, voicemails, the content of comments and social media posts, the conversations between members of a focus group, and so forth.

[image credit: IBM]

According to Gartner, IDC and IBM, “as much as 80 percent of data is unstructured.” To date, unstructured data is seen as the holy grail of insights, though Apache’s Hadoop, an open-source software framework that stores and processes large swaths of unstructured data from multiple sources, is allowing for advancement in this area.

Still, I have a soft spot for pouring over reams of unstructured data to really get a unique sense of the why behind the data.

Media Monitoring vs. Social Intelligence

The next dimension of data is about the perspective. The idea of media monitoring has been around since 1852 in the form of ‘press clipping agencies’. Vuelio, which is an evolution (many acquisitions and iterations later) of these early agencies, is the behemoth in this area, but since the advent of social media, many others have emerged. In the most basic form, media monitors scour mentions of brand, competitors, or general industry news to create alerts and reports for clients. The data you pull from media monitoring can be used to determine things like share of voice and sentiment. Though they are listening to the audience, they really only listen to a specific slice of the data.

["Stories in Press about Nasscom" by Mark Hillary on Flickr]

Social intelligence data, on the other hand, is all about getting a deeper understanding of what the audience cares about, what they are interested in, and how those interests are changing over time. This could include mentions of the brand, but it doesn’t have to. But why would you care about what your audience is talking about other than your brand and competitors? You care because gathering social intelligence data is incredibly helpful for informing content ideas, understanding emerging trends and culture, informing product development, and knowing who else is influencing your customers. It will lead to better content, better products, and better partnerships.

[image from McKinsey Quarterly: How 'social intelligence' can guide decisions]

Moreover, while most data is gathered to get feedback on content or products, social intelligence data, can be gathered in advance of creating content or products. By pulling data from proxy audiences, you can discover insights that will help you form a better strategy.

The Importance of Data to Your Content Strategy

Data can be very useful to overall strategy, but it is absolutely essential to content strategy. The old adage was “half the money I spend on advertising is wasted; the trouble is I don't know which half.”

"Half the money I spend on advertising is wasted; the trouble is I don't know which half." 
John Wanamaker

This is no longer the case. Saying that half works is overestimating the impact (most advertising has an ROI of <1%) and we DO know what works.

 Since John Wanamaker’s time, the number of marketing messages aimed at getting customers’ attention has multiplied. But it’s not just marketing messages that are begging for our attention. Our friends, family, celebrities, celebrity pets, viral videos, and more messages are waiting to be consumed. If you want to break through the noise and get people’s attention, your content needs to be relevant, entertaining, and useful to your audience.

But how do you know what is relevant, entertaining, and useful to a particular audience? The short answer? Data.

Data is how you figure this out. It may not guarantee that you’ll break through the noise, but it will increase your chances. Being data-driven means that you are investing in all forms of data - structured AND unstructured, media monitoring AND social intelligence – and then feeding the intelligence and insights back into your creative team to help improve your content. Data-driven means you can measure your progress – “knowing which half works” - and course correct as needed.

Data will help you understand your audience and measure your progress, and to be effective with it, you need to make it a core part of your strategic process:

For data to work for your organization, you need to have the data team be part of your creative team and be willing to invest in the data-gathering tools as well as the analytical time it takes to deliver useful insights.

Setting Up Your Data-Driven Content Strategy

Contrary to popular opinion, data isn’t cheap. It’s definitely more affordable than it used to be, but it requires investment into various tools, training, and time. The average monitoring tool with access to the firehose of social data will cost around $1000/month. On top of that, you need to account for the hours of training and monitoring. 

Most tools are simple to set up and use, so they will only require a few days to get rolling, however, setting up the processes for gathering, analyzing, and reporting are more complex. When it comes to social intelligence, the best approach is to break your audience out into segments and have individuals assigned to one or a few segments – or beats – so that they can track the trends over a period of time.

These beat analysts ideally work alongside their creative counterparts to share the information that will spark content ideas. To be incredibly effective, the beat analysts join the community and become part of the conversation. With their eyes and ears to the ground, they will be able to really understand the needs of that audience segment, getting ahead of the competitors.

For example, the following segmentation analysis from Affinio shows that the crossover audience for HBO and Netflix includes a segment of ‘Nerds who read a lot’:

Within that segment, you can figure out who they follow (Robert Downey Jr, Tom Hiddleston, Ian McKellen, etc), what kinds of articles they like and retweet (they are excited about an X-Files revival), the topics they like (movies, geek, music, gaming, food/travel), and the hashtags they participate in (#tbt, #thewalkingdead, #supernatural). From there, you can start to fan out to finding where they are hanging out, what they are talking about, what else they are interested in. 

The further you dive into the culture of ‘Nerds who read a lot’, the more you will understand what they respond to. Alongside your brand’s editorial voice, you can find the perfect content mix. Whichever tools you use, the most important part of the set up is that you fully integrate the data team into the creative team.

In Summary

The reason people got sick of hearing the term data being thrown around is because those throwing it around weren’t following the advice with why it’s important or how to use it effectively. Data is incredibly powerful and should be central to any content strategy.

It’s about understanding that there are several types of data that work together to really inform a content strategy, as well as understanding how to set up your processes for success in your organization.

If you still have doubts or questions, you should reach out to me personally. I love talking data and have made it my personal mission to demystify how it works.