Network Science: A New Way To Analyze & Predict The Impact of Social Media
On October 15, 2017, Alyssa Milano sent a simple tweet asking all women that had experienced sexual harassment to tweet #METOO. Millions of Tweets, Facebook Posts, Snapchats, Instagram Stories followed her request. The sheer enormity of the situation, it seems, started to resonate with the world.
What started as a simple tweet became a global phenomenon. Time Magazine even named “The Silence Breakers” as the person of the year featuring a series of high-profile women involved in the movement on their cover.
This is a great story of how social media influences people to take action. How one tweet sparked millions about a topic never previously discussed. But how does this apply to businesses?
Recently a colleague sent over an article on social media’s impact, both positive and negative, with different brands specifically around Trump. The New York Times article looked at the impact of tweets from the Tweeter in Chief when specifically targeting brands, Nordstrom, Lockheed Martin, etc. and evaluate if there is a negative or positive impact amongst brands.
If this is a possibility, the President can send a targeted tweet and impact an entire business in either direction, how does a brand anticipate or analyze the impact of social media?
Most brands will start by analyzing the sentiment of the conversation as key indicators.
There are lots of tools out in the “technosphere” to analyze social media sentiment so you can take a large spike in mentions for a brand and begin to understand why the spike may have occurred. This takes a bit of detective work. After an influential tweet is released, the reaction towards that brand can range from “Let’s head out to shop!” to “I’ll never, ever go there again and you shouldn’t either.” Understanding the percentages here can be helpful for a brand.
However, they don’t tell the entire story.
Can you really trust the percentages of positive versus negative sentiment as being representative of the population? How often will I tweet about the excellent sandwich I received at a chain restaurant? Most often not, but if I get a food-borne illness from a said sandwich, personally my likelihood to tweet a complaint has raised significantly. I’ll be honest I don’t have scientific evidence of this beyond conversations with friends and colleagues. But it makes me question a singular reliance on these tools.
At what point do we see the floodgates open, where everyone responds to a business just because they want to be part of the conversation? Or do people start the conversation more often?
Current widespread social media listening tools are backward facing; how do we look forward?
What if the President tweets angrily about my brand and my impressions skyrocket from a thousand to a million?
Could United Airlines have predicted the impact of outrage following an incident in Chicago, that went viral in part to social media, to their sales forecast? What about Pepsi and the Kendall Jenner incident?
The answer is yes.
Network science has opened the door over the past few decades to forecast and predict the impact of a social media spike.
Network science is defined by Wikipedia as:
Network science is an academic field which studies complex networks such as telecommunication networks, computer networks, biological networks, cognitive and semantic networks, and social networks, considering distinct elements or actors represented by nodes (or vertices) and the connections between the elements or actors as links (or edges).
Network science isn’t new, but its use in forecasting and prediction is starting to become more prevalent as it reaches a level of maturity.
Remember the “Six Degrees of Kevin Bacon” game? The goal is to link any actor to Kevin Bacon by no more than 6 connections. There is still a website you can visit to find out an actor’s Bacon Number. In its simplicity, this is an example of network science.
It is understanding people rarely act alone and are all connected to a society or a community. Network science, when applied to a model, can be the key to determine how people will react to a social media event.
As we are setting goals and strategy for ourselves, we internally at Concentric are looking at what is the impact of this blog and others written by my colleagues? What is the impact of 280 characters? Do we listen to each other?
There is a beauty in the simplicity of how we answer these questions. We just run our own model.
Learn more about how we use Network Science within our model here.