Tips on how best to use available metrics to help prove the return on social.
For brands, proving the return on social is becoming increasingly significant as clients are faced with justifying growing budgets. To date, the Facebook Insights offering has been limited in helping to answer this question. Though the most recent update provides a tremendous amount of potentially useful metrics, it also provides a large amount of noise.
So, how do you navigate through the myriad of metrics? Here are five tips to help!
1. Read the definitions closely
Perhaps the most notable update to the new Insights is that you now have increased access to data which provoke the conversation, “What does success look like to my client?” KPI’s are seeing increasing customization based on the new metrics available. However, the breadth of new metrics can be daunting to face down. Remember, not all the metrics will be suited towards your client. Think through your metrics plan causally; the metrics that matter are the ones that summarize relevant reactions or responses to your brand goals and communication activations. For example, as much as the “People Talking About This” metric is heralded as an overarching success indicator, if you had a campaign that was meant to drive clicks to a site, “People Are Talking About This” isn’t useful since stories aren’t generated by link clicks.
2. Learn through iteration
Facebook Insights is based on snapshots. What makes this great? You get a broad view of how your community is performing at a moment in time. Why isn’t this so great? There’s little opportunity to understand how things change over time. In the social world, we’re always looking at the movement of messages. Snapshots don’t give us that. When reporting, remember to keep and collect the data consistently over time. Create a reporting schedule and master template to help you manage this and reap the benefits of iteration. Whether done by hand or automated using the Open Graph, tracking iterative results is key to assessing the impact of social content.
3. Maintain healthy incredulity
Facebook Insights is full of inconsistencies. While 1 + 1 +1 = 3, in Facebook insights, paid + organic + viral impressions does not equal total impressions. When building your metrics plan, you’ll often have to choose between “total” counts and component value counts and will have trouble combining them. Don’t assume the definitions mean what you think they do. Make sure to get into the Insights exports and read that definition. Many of the metrics “look” the same but there are subtle differences that can make all the difference. Step back from the numbers and ask yourself “are these metrics comparable?” Assuming face value validity can be dangerous. We’re still trying to get confirmation on the definition of paid/organic/viral in clear specific terms. So, until we know, approach usage of these detailed metrics with caution.
4. Be consistent
The plethora of metric options can be extremely daunting considering the thousands of data points in the exports, which tend to resemble “number soup.” This is because key metrics are now viewed from multiple perspectives: page vs. post; lifetime, daily, weekly, 28 day; unique vs. total. Focus on a few key metrics that allow you to evaluate success and when pulling the export, delete unnecessary items. What’s important is that you’re consistent in your choice and remember, “fewer metrics, more long term analysis.” Also, if you’re not sure what something means, it’s best to not use it at all.
5. Facebook Insights is not a one stop shop
With some time to mature in the market, Facebook Insights has demonstrated itself to be the “end of the beginning” for evolved social insights rather than simply “the end.” In many ways, it provides the most basic set of metrics one can get. Take stock of what you can get from your client’s platforms and Facebook and then combine your data points. Empower yourself with the Open Graph (graph.facebook.com) and API tools to get data in a much more organized and efficient way. Don’t just rely on analytics; spend more time on analysis. When in doubt, speak to your local Digital Insights / Measurement practitioner to get some ideas on how to design an Insights framework that can move from fanning and engagement to deeper consumer insight and merge those ideas with the overall brand goals from account teams and strategists. In many ways, information in the information age is as valuable as stone was in the Stone Age. The most valuable insights come from the contextualization of data, not the data content itself.
By Richard Ng and Erica Buckley
Continue reading here.
Originally published on Edelman Digital, republished with kind permission.