Co-produced with Chuck Hemann on Edelman Digital
What’s the next big thing after “listening”? It has something to do with data—but more importantly the decisions a business makes after sifting through the data and understanding what impact it can have. The social-digital sphere as it turns out is full of data. Some of it is conversational data, some of it is search related, but all of it leaves a digital trail of clues for us to follow, only it’s up to us to decipher those clues. (hint, Facebook loves your data.)
As social media has made its way into almost every communications program, we have also seen the need for more advanced approaches to monitoring intensify. In fact, monitoring as a discipline is being replaced by listening. What’s the difference? The term monitoring implies a passive action. If we’re monitoring, we watch for mentions of our brand for the purposes of acting in case our reputation is threatened. If we’re listening, we’re using that data for more aggressive action either in terms of real-time content development or the eventual use in the communications planning process.
Lets explore both of those models for a second…
Listening for Program Planning
The most common model for listening is using the data for program planning. It’s a very linear process, and typically takes several weeks to fully execute. With this model we gather data on our brand, competitors and category, and develop insights that inform a strategy and tactics. It’s important to keep in mind that insights derived from social conversations have application well beyond social. Those key learnings can be applied to hybrid media, owned media and traditional media properties. After we’ve used the data for program planning we execute and then measure.
Listening for Real-Time Content Development
With this model, our goal is to use information we’re gathering from social media conversations about our brand, competitors and category to inform the development of content in real-time. The additional layer with this model is that we’re listening to social conversations with the intention of spreading the insights we’ve gleaned to other parts of the organization. It’s our belief that social conversation data has application across several different parts of the organization, including product planning, CRM, strategic planning and human resources. After we’ve determined where the information should be sent within the organization, we determine where and how we communicate. When that piece of communication has been issued, a new signal has also been created for us to analyze and measure. This cycle is much more condensed than the previous model – typically spanning days instead of weeks to months.
Which of these models is best for you? The answer is you probably need a combination of both, especially if you’re engaging your community in real-time. The key to both is insights that help the entire organization. If we’re mining consumer conversation data, we need to be thinking about the organization more holistically. If we see a conversation that our product planning team could benefit from, we need to be looping them into the data mining process.
While we are focusing on looking at and deciphering data (signals) in the communications space, It’s our view that listening is going to continue to evolve and that we’re already starting to see the beginnings of true social intelligence systems within larger companies. That means we’re listening to conversations in real-time, and making business decisions based on the insights we can glean from the data. Walmart Labs for example is experimenting with analyzing conversations and trends on Twitter to impact how they stock their shelves. This experiment gives us a glimpse into the future of integrating social data into business intelligence.
Are you leveraging either of the above models? Do you think social data has a role both within and outside of the communications realm? We’d love to hear your thoughts.
*Walmart is an Edelman client.
Original Post: http://darmano.typepad.com/logic_emotion/2011/11/data.html
