Social Media for Forecasting

Author: 8th and Walton

bigdatasBig data can include both structured data, like sales reports, and unstructured data like conversations in social media. Social listening is an obvious source of information on consumer response to products, but it can also provide information you can use to fine-tune your forecasting — or even to get it started.

Annibal Sodero of the Walton College of Business spoke with Saturday Morning Meeting about just this subject.

“Social media can provide suppliers cues signaling whether a campaign has the potential to be a hit or a miss in specific markets,” Sodero said. “We have consistently found that positive, neutral, and negative sentiments conveyed by social media conversations can predict sales.”

A case study from DataRank, a social listening service, provides an example. In this case, a new beauty product wasn’t doing well on the shelf at Walmart. Social listening data allowed the supplier to identify the right target market and see that the product had greater potential that it had shown, as long as it was marketed to the right consumer segment.

“Suppliers can retro–feed social media data into their supply chain management systems to better plan for producing and distributing inventory to specific regions and placing it at the right place at the right time,” Sodero points out.

The first question for any supplier planning to incorporate social media data into planning and forecasting: “What is your strategy?” The data set for your brand, your products, and your category is probably far too big for you to be able to glance at a spreadsheet and see the actionable data points.

Instead, you have to identify the questions you need to answer and the data sources which will best provide the answers you need.

Sodero offered four key points suppliers must consider when they plan to use Big Data, including social listening, as a source of data for the supply chain:

  • Volume: Big Data is… well… big. What you see when you heck your Twitter feed on Sunday morning is not a sufficient sample size to give you actionable information.
  • Variety: Retail Link is essential, but supplementing it with social listening and other information, and harmonizing the data helps you to get the full picture.
  • Velocity: The closer you can get to real time, the better.
  • Veracity: It’s essential to use reliable data. Don’t include sponsored tweets, for example, if you’re trying to get a clear image of consumer response to your brand.



This entry was posted in Focus on Suppliers, Supplier tips. Bookmark the permalink.