Alternative Data, Social Sentiment Data

What is Social Sentiment Data?

Social sentiment data is processed text from social media, news, company management communications and other publicly available sources. Sentiment data includes public reactions to the news, product ads, scandals involving public companies, etc. Data can come in the form of textual posts, digital images, or videos. And, includes any form of online interaction between people on social media sites like Twitter, Facebook, LinkedIn.

Put simply, this data is all about attitude and feelings people have about a brand on social media.

This data is relevant for all companies but may be more influential when looking at smaller companies than larger, more established ones. Bottom line is that this data may be an early predictor for price movement in a public company stock. It provides a clear read into current trends and brand virality, especially when tagged properly and used in combination with other data sources to complete the picture.

Why is this Data Important to a Brand?

  • Provides Consumer Insight – from a brand management perspective, brands want to collect as much information possible from their consumer audience. Understanding the reaction of the brand’s consumer base helps that company plan strategically for future campaigns, content, and products.
    • If there is a lot of negative attention or negative remarks from consumers post launch of a new product, a brand will take action. This may include pulling the product from stores, making a public apology, and attempting to clean up the messaging if it happens to be insensitive.
  • Customer Service – for a brand to be successful, the customer service aspect of their business must be buttoned up. If consumers express dissatisfaction and the negative sentiment spreads, this heavily impacts public-facing businesses that rely on positive reviews to thrive.
  • Business Crisis Management – negative news spreads quickly on social media, especially when negative sentiment has a hashtag linked to it. Depending on how severe the original issue was, negative sentiment could lead to a crisis.

How Can this Data be Used?

A common example of social sentiment data is Twitter. Twitter sentiment analysis is a popular way of gauging public reactions to a product release, rumors, event or public announcements. But, because this data is largely unstructured and requires some processing. Namely, the data consumer need to find: 1) a way to collect the text; 2) a way to classify which company the text is about; and 3) the tools to evaluate sentiment – whether it is positive or negative and what that means.

Use Cases:

  • Trade support – Assist in trading or initiating positions based on breaking news expected to create near-term market movements
  • Market awareness – Provide situational awareness and context into sectors, topics, and companies of interest, and contextualize asset price movements
  • Thesis generation – Surface trends to support development of general investment theses, through historical and analytical tools

Social Sentiment Analysis Data Providers and Tools

  • Dataminr – Social sentiment and news analysis based on Twitter data. The exclusive relationship with Twitter allows for access to all publicly available Tweets published in real time.
  • TipRanks – Sentiment and ratings data on sell-side and investor stock opinions.
  • Prattle – Sentiment analysis based on various public news sources and social media.
  • Parsely – Parse.ly’s network data is comprised of the biggest media and content driven sites and what topics (companies, products) over 1 billion people are paying attention to online. Sentiment analysis based on various public news sources and social media.
  • Estimize – Consensus estimates on public companies provided by both buy-side and sell-side analysts. By sourcing estimates from a diverse community of individuals, Estimize provides a more accurate, more timely, and more representative view of expectations compared to sell side only data sets which suffer from several severe biases.
    • Use cases: The most representative consensus of market expectations, post earnings drift, earnings revisions, history surprise, earnings risk mitigation, earnings yield factor, earnings growth factor, estimate dispersion factor
  • Hootsuite Insights – Automatically analyzes all your social media platforms, news sites, forums, and blogs to reveal insights that include influencers, stories, trends, and sentiment.
  • Social Mention – Automatically analyzes all your social media platforms, news sites, forums, and blogs to reveal insights that include influencers, stories, trends, and sentiment.

Bottom Line

Sentiment analysis helps brands and investment managers with the following:

  • Identifying negative mentions about a business, a service, a company, a marketing campaign, an event in social media and on the Web
  • Spotting angry customers on the verge of starting a social media crisis Analyzing how those customers react to product changes and company announcements
  • Spotting happy users who, for example, are more likely to become brand ambassadors and influence other consumers to purchase from a certain brand at a higher frequency
Share this:

Leave a Reply