How to Apply Data and Machine Learning to Boost Your Marketing Strategies



For the team at Socialbakers, data isn’t the whole story; it’s merely a tool that allows us to tell stories with heart and connection.

We are excited to announce the first round of leaders who will bring our 2020 theme HUMAN.X to life at the Broad stage this June (17-18).

Socialbakers’ Imre de Daranyi started his SMWLDN talk with a statistic that marketers should find startling: 32% of chief marketing officers consider marketing analytics to be a vital support to their company’s strategy. “It should be 90%! It should be 100%!” the Senior Director for Global Client Solutions lamented. And a pair of additional statistics undergirded the core of his presentation: while 61% of marketers say they’re doing a good job providing clients, customers, and followers with quality content via their social channels, just over half (51%) feel like they’re seeing too much irrelevant or “noisy” content.

Join SMW Insider to watch this #SMWLDN session


How can the wide gulf between how much loud content exists, and how much good content exists, be closed? How can the wide gulf between how effective we believe we are, and how effective we actually are, be closed? And what role can data play in those goals? Using the outlet de Daranyi drew from for his session – Instagram Stories – we can explore what data can do to create a better customer-creator match.

Here are the primary insights and takeaways:

  • Wait and See” Can Work…When Done Well
  • Use Your “Dwell Time” Wisely
  • Study Your Direct Competitors to Help Define Goals

“Wait and See” Can Work…When Done Well!

Socialbakers operates with a “graded” content model, that allows the system to grade content based on how it performs in its initial deployment. Within the first two hours, de Daranyi said, posts, blogs, videos, and the like can be graded from A+ to D. And to the surprise of few in the room, he reported that just over a third of content (33.5%) is graded D.

The highest graded content takes advantage of the unique attributes of the platforms where they’re posted, and helps the reader or user solve a problem or learn something new. For Instagram Stories, this means using features like polling, format-customized video, and opportunities to let customers or users meaningfully participate. Following these cues from your best performing content, rather than just creating without looking at how these grades are being earned, is your best opportunity to learn from the data that is constantly being presented.

Use Your “Dwell Time” Wisely

When de Daranyi was at AOL, he frequently monitored a stat called “dwell time,” or time on site. Now, at Socialbakers, he thinks about “completion rates,” or the time that users spend in a Story before clicking out and moving on. Many of the same stats that were once monitored for web traffic, also matter deeply to traffic and time spent on social. “Social data and web data are not mutually exclusive,” he declared, pointing out that each can effectively inform the other.

The data that’s pulled from web engagement has the potential to inform the creation and maintenance of updated customer personas. If these personas are frequently updated, content can be designed, redesigned, and “re-graded” with the goal of preserving that reach in Stories that many are lamenting the loss of as the feature’s use increases.

Study Your Direct Competitors to Help Define Goals

Anytime the notion of being graded comes up, thoughts of studying can come up as well. In the case of Socialbakers, this means comparing your brand’s rates of engagement or exit against direct competitors, others in your vertical, or even across verticals. These rates can serve as a crude rubric of sorts, helping you define a goal to aim for.

Data on attention and engagement can be helpful in this process, but so can other qualitative measures like relationship to story. In this way, the focus isn’t wholly on numbers, but also on the feelings that drive people to stick with you and what you create. “We know AI isn’t flawless; there are countless instances where it could go wrong,” de Daranyi was sure to point out; while AI can help with the data part, the creative part—the human part—is up to us. And by ensuring that both are in play at all stages of the marketing experience, we can ensure that our strategies are efficient and productive, but also heartfelt and humanized.

Join 100,000+ fellow marketers who advance their skills and knowledge by subscribing to our weekly newsletter.

Newsletter Subscription

Get the latest insights, trends and best practices from today's leading industry voices.

Learn More

Write for Us

Interested in sharing your ideas and insights with the world? Become a SMW News contributor and reach 300k readers each month.

Apply Here