How to Get Started With Data Storytelling: 6 Actionable Steps


Create engaging content for free

This is the 2nd of 3 posts I’ve put together to help you improve your team’s data literacy. If you haven’t already, start with “Visual Data Literacy: 3 Fundamentals for Content Marketers” to put the amen in your fundamentals. When you’re done with this one, check out the 3rd piece on “How to Bring Your Data Storytelling to Life” to learn how to get your data insights into a strong narrative structure and ready for design. Feel free to reach out to me anytime at jason at visage dot co to explore ways you can bring your data to life in your content marketing. 

You know this: creating original content is hard work.

Your audience has a lot of really good options to choose from when considering which content to consume.

It’s your responsibility, as a content marketer, to create something that’s actually worth consuming.

That’s because you’re competing with the world’s best publications and entertainment outlets when you create content.

If you do what everyone else is doing, it hurts your brand because you’re not standing out—you’re not saying anything original.

There are many ways to be original. 

For instance, you can look to your organizational culture for ideas.

As you look to express more facets of your brand, bringing your internal data to life is one of the most powerful ways to differentiate your organization from the herd. It’s not the only form of content required along the way for marketing success in most cases, but it can certainly be your Silver Arrow (if you’ve never played Legend of Zelda, go do that for a few weeks and then come back).

121113-zelda05.gif

The challenge lies not in recognizing that data storytelling is valuable but in knowing where to start.

So where do you start? How do you get going—without feeling overwhelmed?

Here are six steps to help you get started in creating original stories using data. 

Step 1: Create a list of questions your audience wants answered

When you start with a list of questions that your audience wants answers to, you’re starting from the right place.

That’s because creating good content is all about providing value to your audience.

This initial stage is about getting your mentality right; you’re simply exploring ideas.

As part of that, you want to ignore any preconceived notions you have about the process being hard. You also want to ignore whether or not somebody is going to be available to help you dig into the data and find these answers. Just get the questions down for now, no matter how lofty they seem, as we’ll refine the list in the next steps.

Some answers will be subjective, such as what type of content to produce; some answers will be objective and can be answered or supported by your internal data.

Hilary-quote.png

        

Step 2: Filter your list of questions

Now it’s time to filter your list of questions and identify which ones are qualitative and which are quantitative.

Once you have your list, you want to start determining what each one requires:

  • What conditions are required for us to get these answers?
  • What needs to be true in order for us to be able to get this data?
  • What resources are required?
  • What limitations already exist (eg pressing deadline)?

Here’s an example.

Let’s say you have a question: “What percentage of marketers struggle with creating visual content?”

You assess what will be required to get to a solid answer, and learn that you will need to:

  • Access the opinions of 500 marketers
  • Write a survey
  • Collect the survey data
  • Visualize the data

Your next question to help your audience answer might be: “What type of content should I create?”

And in that situation, your conditions might be that you need to know:

  • What type of content your audience is already creating
  • How all that content is performing
  • How that content aligns with your audience’s culture, business and marketing goals

You may start looking at this particular question and realize, wow, this question is getting too complicated and subjective compared to the survey idea, which is something you can quantify and put together into a pretty well-structured story—without delving into research projects on each of the people in your audience who might need to answer that particular question.

This isn’t to say that one idea is inherently better, but your choice of which question to answer with your data will be informed by your available time and resources.

Step 3: Scoring your questions

Now it’s time to score your questions.

Score the questions on a scale of 1 to 10 for:

  • How quickly can you obtain this information (10 being the fastest)?
  • How confident are you in the reliability of this data (10 being completely confident)?
  • How valuable will this answer be for your audience (10 being most valuable)?

Total up the scores from each of these three categories, and then prioritize the high scores. This method of scoring ideas was inspired by Sean Ellis’s great product for scoring marketing experiments, Canvas.

Step 4: Where do you look for your answers?

Simply put, there are two places to look for your answers: within your organization, and around the world in a plethora of external sources. Here are a few of our favorites.

Internal sources

Your data person (or team)

If you’re a part of an organization that takes data seriously, you likely have data analysts or data scientists. They are the people who are asked 10-50 times per day if they have any interesting stats on ______. If you’re on a small team, it might be your director of engineering or a technical co-founder.

There are other organizations where this role is filled by a highly technical marketing manager or analyst. Whoever you have access to and who understand data and analytics are solid choices.

As a content creator, take the initiative to educate yourself on what it takes to improve your collaboration with your talented data team. I’ve talked to many, many forward-thinking data analysts and with very few exceptions, they truly want to see their organizations get better at publishing original content based on the gold mine of data insights they are sifting through daily.

surveys

Customer surveys

The other source to consider—not a bottomless well—is to periodically engage your customers; you’ll receive a diverse range of perspectives. A survey is typically the best way to gather a lot of information from a lot of people at one time. We use (and love) Typeform for collecting survey responses. You can keep these short and sweet, or go big with something like an annual census such as the one Burning Man conducts each year. 

External sources

white-papers

White papers

A good external source of original data would be white papers or a brand that has published white papers but hasn’t gone to the trouble of pulling out interesting visuals or highlights from the white papers.

You could be almost certain that no one’s read it and covered it, so you can dig into a 50-page or 100-page white paper and you’ll be rewarded for your excavation process by finding something that very few people have read or created a visual piece of content from.

case-studies

Case studies

Take a look at case studies that other brands have done that show unique insights or results that are relevant to your audience.

public-data

Public data

A good way to get started and get some confidence going is to look at public data, such as authoritative industry studies.

The advantage of using public data from solid sources is that the data is often already well structured, and it’s already collected for you.

The other advantage is that a reliable third party is trusted; it has conducted professional-caliber market research using well-structured, already analyzed and clean data—which saves you a lot of time.

The downside is that someone else in your industry has probably covered that same study or has written his or her own take on it. But you still have the opportunity to find an original angle in somebody else’s data for a piece of original content.

The goal is to at least make an increasing portion of your content truly original. That might mean that you take that public data set and uncover some kind of analysis or correlation in the data that no one else has seen.

Step 5: Get a quick win

The goal here is to get started and get a quick win under your belt.

Now that you’re familiar with the places to look for your data, select the questions that look like they have the most accessible information.

Of course, you want to publish reliable, trusted data that will be valuable to your audience. If the most readily obtained information for a particular question scores low in either or both of those categories, just archive the question for now.

Select the questions that would likely have the most accessible information. You may even have a throwaway project here, where you create a quick, single-chart graphic and practice and warm up on it.

Remember: Getting your first experiment completed goes a long way to boosting your confidence and keeping the ball rolling. You can move however you want into the other questions you evaluated in Step 3 – by overall score or by prioritizing another quick win.

Step 6: Stay unbiased (or whatever)

223a376e621666fa5b92f92715ff51ab557f4830cd21f8b6008914fff4e59590 (1).jpg

As hard as it is, be aware of your own biases and look at each question through fresh eyes. Be as objective as you can be. But also share your own lens and process, and consider making this journey and your choices of what to display into part of your content – this adds your own personality and originality. Share the raw data, and turn your audience, as Moritz Stefaner says, from consumers into fellow travelers who enjoy exploring the data with you. In describing the editorial angle of a data visualization author, Stefaner writes, “Let’s make no mistake — even a very data-heavy, ‘sober’ representation of data has an author who made clear decisions on what to include or not, what to combine, or not and what to prioritize.”

It can feel counterintuitive to share data that contradicts previous content you’ve created. But finding those nuggets of truth and unique answers in your data can open up many different doors for future content creation, and potentially represents a powerful facet of your brand. Rather than bending truths until they are convenient, you can stand in awe of what the data actually shows and share your methodology and open yourself to critique in the process, which helps you get better. We can all improve.

If you’re finding data that contradicts what most people think (or what most people are espousing, and you have good reason to rely on that data), then you could have a really original voice in the conversation. These surprises in your data can often be a prompt to dig deeper to uncover valuable insights.

Ready to get started?

In the next piece in this series, I’ll show you how to start presenting your data.

For some quick inspiration, you can see how we drink our own Kool-Aid by checking out this recent survey we conducted that was picked up by Forbes. Our Visage team is also here to help answer any of your questions in our platform if you sign up (or sign in), and check the bottom right hand corner for the question mark icon to shoot us a note.

Learn more about creating original content through data storytelling by downloading our free e-book, Content Marketer’s Guide to Data Storytelling.

After reading this e-book, you’ll discover:

  • Why it is essential to learn a shared language of collaboration with the data experts on your team
  • How to get more interesting data by asking better questions
  • How to create visual data storytelling to more easily connect with your audience