Synthesizing Social Data

For the last several years, marketers have been told that data will be the key to success. That’s not entirely true.

In the social media marketing world, there’s so much data available, it can be overwhelming. Despite the hype, marketers have struggled to find real value from “big data”, so much so that many now think it’s a bad word. What is useful and what isn’t? Too often, marketers don’t know the answer to that question. They just collect data for data’s sake, put it in massive spreadsheets, and hope that something might just hold the golden insight.

That’s not the path to success. Marketers need to be intentional about what data they’re collecting, making sure that it’s telling them something useful and actionable. A good way to start is to understand the different types of data.

First-party data is specific to each business and is something marketers can usually get from web analytics. It’s tracking who is coming to the website, who is signing up for a newsletter subscription, who is buying what — that kind of thing. At the other end of the spectrum, third-party data (public, aggregated) has plenty of scale, but lacks the depth of information unique to the brand. It’s great for industry norms and benchmarks, best practices, and supplemental data. And while technically, social platform data is just the social networks’ first-party data (making it the brand’s second-party data), the brands themselves have a unique social presence and interact with their fans on these social platforms. To some degree this data is public (what people ‘like’ on Facebook, for example), but it is powerful in that it is also unique to the brand (e.g. 30 people commented on a post about Product A).

Each type of data offers a different perspective on the customer, but collecting data from the various sources isn’t the complicated part. The complicated part is layering and piecing various parts together to make sense of it. It’s taking information from your first-party data and seeing that a piece of content stands out from what your third-party data says is typical, and then correlating it with what you did on a particular social platform. Or even deeper and more granular: it’s seeing that when Customer A shared Product 1, 200 of her friends saw it on Facebook, ten of them went on to buy something from the brand, and five more shared another piece of content from the website…and then using your first-party, third-party, and platform data to identify and target similarly powerful influencers to recreate that network effect with more customers.

HOW?

So now that we’ve defined the disparate data and discussed the potential benefits of integrating the data sources, how do you do it? How do you make the data work together to create the full customer picture and uncover new customer insights?

Step 1) Consolidate
While it’s nice that Facebook, Twitter, and most of the other social platforms offer pretty good native analytics, having each set of analytics separated in silos (Facebook only measures what happens on Facebook, Twitter only measures what happens on Twitter…) creates a difficult barrier for marketers. To do a decent job comparing across platforms, marketers need to take the native data and export it to Excel, where they can then compare data. Third-party social analytics often do a better job of consolidating data, making it easily digestible, and therefore, more actionable.

Step 2) Be Aware of Inconsistent Metrics/Use Consistent Metrics
But even after consolidating native social data into a single view (or using analytics that aggregate it), marketers are still stuck comparing apples to oranges. From basic measurements to how many things they measure, native social analytics don’t play nicely with each other. Even for something as table stakes as impressions, Facebook measures impressions differently from Twitter, and neither are equivalent to an impression on Vine (defined by “loops”).

To that extent, engagement metrics are generally subjective, and are especially hard to measure across different sources of data.

Attribution and acquisition data, too, differ from source to source. Most analytics still use last click attribution, even though it’s widely recognized as not being representative, as well as referral data, where dark social creates a major hole.

Marketers must be aware of these inconsistencies and be careful to not make decisions based on something that could be explained by different measurement methods. Consolidation plays a part in solving this problem, as does just making a decision on what the “truth” is and making sure the data is consistent in representing that.

Step 3) Focus
Identify the metrics that matter to your business and stick to those. Even with unified metrics (step 2) on the same analytics platform (step 1), it’s important for marketers to focus on actionable data and to be thorough about collecting it. While a shotgun approach to data collection and analysis may capture a wider range of data, insights are more likely to slip through the cracks. It’s a generalist versus a specialist trade off. When it comes to defining and capturing the nuances of the customer journey, depth is more important than breadth, and narrowing in on key metrics plays a big part in that.


Organizations are increasingly making holistic and personalized customer experiences a priority. If you want reliable customer insights that not only shape your marketing, but your entire organization, make sure your analytics can layer and synthesize your various sources of data.