Data Storytelling: The Art and Science of Social Media Metrics

Understanding how your social campaign is performing requires a fine mix of data and interpretation, as many marketers know. But it’s not always as simple as that. A lot of brands are left asking, “What does all of this data mean to my social channels?”

Social media firm 5Loom, which is owned by Council member IMRE, created a new concept to turn a pile of analytics into contextual opportunities for brands.

To find out more, the Firm Voice dug a little deeper in a Q&A with Crystalyn Stuart, president of 5Loom powered by IMRE, about data storytelling and how it can help brands tell compelling stories.

What is data storytelling?

CS: In two parts, it’s (1) how we use data visualization to help us see and read the story social data tells, and (2) how we as social media experts package that story and make adjustments to campaigns.

We approach social media like a 24-hour-a-day stream of data that should have an impact on how we’re building and managing social communities, programs, promotions and campaigns.

It should, but unless we can find the answer to the question “so what?” all that data just seems time-consuming. That’s why we practice data storytelling. It’s the act of data visualization before, during and after mining/analyzing data.

Which pieces of social media data are most relevant?

CS: There are infinite social media data inputs, some are half-baked (in channels such as Pinterest), and some are over-baked (in listening/monitoring tools). Either way, which data matters? Our motto: All data matters, capture all, analyze less.

To be more specific, every program should map to two key reporting elements:

  1. Program success: Measured by Key Performance Indicators (KPIs) that map to unique goals.
  2. Program adjustments: A changing set of metrics that let us evaluate all aspects of a program and steer the ongoing strategy. We modify these in real-time and monthly.

As we load data points into various visual formats, we begin to see and read the story that the data shares. Here’s a simple example, I may ask my team to cut content data in three ways: (1) show me volume of posts against a certain category, (2) show me the reach of those posts by content type and (3) show me the engagement by content type.

By using some simple data visualization, my group can easily read the story, and quickly identify the important headline and theme. From there, they can begin to further analyze the trends.

What do you mean by context?

CS: In the social media realm, context has become more important than ever. Evaluating campaigns, data points, results and overall channels not just against themselves month-over-month, but against competitors, aspirational brands and industries (niche and broad) is a critical layer that helps brands gauge the question, “are we doing this well?”

But context is not just about whose Facebook page has the best virality. It’s about taking wider trends and answering the question, “what does this mean to me?” For 5Loom, social context is about blending together outside and inside contexts to help uncover meaningful opportunities.

Here’s a 101 example. I may discover that one category of content I’ve been generating on Facebook isn’t getting as much engagement as a similar topic shared by a competitor. In evaluating the trend, I can uncover meaningful examples of a call-to-action or the use of a subject-matter expert that has helped my competitor make this topic more engaging.

From this context, I can begin isolating variables and testing various ways to package and deploy content around this subject. Constraining my tests, benchmarking carefully and applying creativity to the opportunity typically results in top ROI.

Why do you emphasize data packaging?

CS: I can’t tell you how many clients have asked me “so what does this mean to me?” Every time we share a report, we focus on how the report tells a compelling story–not just for the audience–but for the story itself.

Often audiences don’t know what data they should be looking at, or how that data should impact their program. It’s our job to tell them and it’s the data’s job to inform us.

A few key tenets of our model include:

  • Less data, more story
  • Less linear, more contextual
  • Less data showing-off and more zeroing in on opportunity
  • Less general to-do’s and more rich strategic guidance
  • Less month-over-month indices and more comparative insights

Do you think you have data storytelling all figured out?

CS: We’re not perfect, and measurement/reporting in the social realm is definitely a moving target. But we are relentlessly optimizing, changing, testing, altering our approach based on our understanding of what’s needed to manage a social strategy and what our clients want and need.

We’ve launched a new series of reporting programs that follow this process, and by adding a social media graphic artist to our team, we’re able to tell richer, more compelling stories than ever before.

2 Responses to “Data Storytelling: The Art and Science of Social Media Metrics”

  1. henrik dresboll said on May 19, 2012 at 4:42 am

    Dear Crystalyn,

    Great post. I will be looking forward to follow you on this subject. Actual I am working on a project of mapping KPI in immersive storyworlds by talking with people from LA, NYC and San Francisco. Maybe I can meet you next time in NYC? I will be there in october?

    Looking forward to your respons,

    All the best from Denmark. Henrik

  2. Find Meaning in the Metrics | idaciti said on August 5, 2014 at 1:10 pm

    […] post by Crystalyn Stuart with a nice infographic from 5Loom on uncovering the story from the data of […]

Leave a Reply