Back to insights
Article • 25/06/2024

Sentiment Analysis Measured Your Way: Introducing Configurable Sentiment

For Corporate Communications leaders, sentiment analysis has been the key to understanding public perception. But sentiment can be subjective—from company to company, what your internal team decides is ‘positive,’ ‘negative,’ and ‘neutral’ sentiment can be a subjective reporting decision.

Now, Signal AI’s new Configurable Sentiment Analysis feature gives communicators the power to choose how to measure positive, negative, or neutral sentiment, with five new ways of aggregating sentiment. 

In this article, we’ll explore the most advanced sentiment analysis for media monitoring on the market, details on the new feature, and how it works.

First, what is Sentiment Analysis?

Sentiment is a measurement term that can help you decipher whether coverage is positive, neutral, or negative. It’s a critical tool for determining whether a product message or announcement has been well received.

Sentiment Analysis is the process of understanding the meaning behind the words spoken online and in the media. It’s often used in PR and Communications measurement, reporting, and media monitoring to determine how messages and corporate communications resonate with the desired audience. 

For more on sentiment analysis, read our complete guide here.

Why is it important to measure sentiment?

Analyzing media coverage without sentiment analysis is like a map without a compass. 

You need the context beyond how many people mention your business to how they feel about you, your products, or a campaign. Knowing only your total number of mentions or volume of coverage tells a limited story of how your messages land and the public’s perception. 

Sentiment analysis can inform communications campaigns and corporate messaging, map an approach to reputational risks and your competitor landscape, and increase confidence in your business decision-making.

Using AI tools like Signal AI with sentiment analysis can paint a more holistic picture of your current reputational landscape and offer valuable data to make decisions.

Signal AI’s Entity-Based Sentiment Analysis Model

Most sentiment models on the market are sentence-based, meaning they typically take an average of all positive and harmful words and apply it to all entities within a sentence, which can lead to inaccurate and unreliable results.

Our AI-powered search terms, which we call entities, allow you to quickly capture mentions of an organization, a person, or a location. We can then map the connection between these entities and other AI-trained topics in our platform. 

Signal AI’s sentiment model uses transformer-based AI to understand the context of a sentence and, therefore, recognize when multiple entities should be assigned different sentiments. 

Sentence-based Sentiment model vs. Signal AI's Entity-Based Sentiment Model

Signal AI’s AI-driven Sentiment Analysis model has given Communications leaders the utmost confidence in their reputation data.

Now, Signal AI has cracked the code on Sentiment Analysis. We’re introducing a proprietary configurable aggregation methodology—a new customizable way of measuring to give you more accurate insights.

Measurement Your Way: Introducing Configurable Sentiment Analysis 

As data-driven communications leaders, it would make sense to feel all reporting is entirely based on hard data and quantifiable metrics and, therefore, is an objective practice. But with measuring sentiment analysis, there’s more to the story. 

Is one bad mention enough to label an entire article as negative? Does a headline mention matter more than a mention in the body text? With Configurable Sentiment, you can customize how the Signal platform aggregates sentiment on the article level.

How Configurable Sentiment Analysis Works

Configurable Sentiment matrix based on sensitivity to negativity and mention position within the article

Every time an entity is mentioned in an article, sentiment is determined – negative, neutral, or positive. When an entity (company, person, or location) is mentioned multiple times in an article, sentiment is determined for each mention. Then, all mentions are aggregated to create an overall sentiment label for the entity in the article.

The five methodologies of Configurable Sentiment are categorized based on two factors: 

  • Sensitivity to negativity
  • Mention position within the article

With Configurable Sentiment, you can take control of your data and choose from 5 options based on your own sentiment appetite:

  • Negative Bias: “I don’t want to miss a single unfavorable mention.”
  • Negative Lean: “I need to emphasize negative sentiment to measure reputational risk.”
  • Balanced: “I need a proportional approach to measuring sentiment.”
  • Positional Balance: “Mentions in the headline and first paragraph of an article are more important to me when measuring sentiment.”
  • Positional Negative Lean: “Mentions in the headline and first paragraph of an article are more important to me when measuring sentiment, and I need to emphasize negative sentiment to measure reputational risk.”
5 Options for Signal AI's Configurable Sentiment Analysis feature

This empowers organizations to change their sentiment aggregation to best fit their needs.

Benefits of Configurable Sentiment

This innovative, proprietary feature is the first of its kind available in the media monitoring market. Along with flexibility, Configurable Sentiment offers key benefits to Corporate Communications teams:

  • Save time with fewer manual corrections for your reporting and analytics.
  • Enjoy improved transparency in your data
  • Demonstrate the impact of your work with even greater confidence

No matter what makes sense for your measurement and reporting on your communications strategy, configurable sentiment will help you be more confident in your reporting in a way that works for you. Take control of your data with transparent methodologies and descriptions for informed decision-making. 

For Signal AI clients who want to learn more about Configurable Sentiment, visit our help center here. 

Learn more about how Signal empowers business leaders to make informed and confident decisions by requesting a Demo here

You may also like

View More
Article

4 Controversies of the Paris Olympics in the Media

As a major global event, the 2024 Paris Summer Olympic Games have received plenty of news coverage this summer. However, the spotlight isn’t only on athletic prowess and global unity. Controversial topics abound, from doping scandals to environmental concerns, casting a shadow over the events.  As the Associated Press writes, “The Olympics frequently have been […]

Read more
Article

‘Come to the table with data’: Insights from Prudential and SAP Communications leaders

The Communications industry, and particularly the role of Communications leaders, has seen a huge transformation in recent years. We’ve gone from a communications world led by intuition and gut feeling to more advanced data-driven decision-making. Reputation management is also pressingly important across the C-suite and key stakeholders. How can today’s Communications leaders better position themselves […]

Read more
Article

Brands Go Quiet this Pride Month, Fearing Rainbow Washing

Every June during Pride Month, the corporate landscape transforms into a kaleidoscope of rainbows and pro-LGBTQIA+ messages. Ideally, it’s an opportunity for corporations to show support for the queer community and advocate for social change. However, it can often come across as “pride-for-profit.” Rainbow washing (a.k.a. pinkwashing or rainbow capitalism) is the marketing practice of […]

Read more
Article

4 Controversies of the Paris Olympics in the Media

As a major global event, the 2024 Paris Summer Olympic Games have received plenty of news coverage this summer. However, the spotlight isn’t only on athletic prowess and global unity. Controversial topics abound, from doping scandals to environmental concerns, casting a shadow over the events.  As the Associated Press writes, “The Olympics frequently have been […]

Read more
Article

‘Come to the table with data’: Insights from Prudential and SAP Communications leaders

The Communications industry, and particularly the role of Communications leaders, has seen a huge transformation in recent years. We’ve gone from a communications world led by intuition and gut feeling to more advanced data-driven decision-making. Reputation management is also pressingly important across the C-suite and key stakeholders. How can today’s Communications leaders better position themselves […]

Read more
Article

Brands Go Quiet this Pride Month, Fearing Rainbow Washing

Every June during Pride Month, the corporate landscape transforms into a kaleidoscope of rainbows and pro-LGBTQIA+ messages. Ideally, it’s an opportunity for corporations to show support for the queer community and advocate for social change. However, it can often come across as “pride-for-profit.” Rainbow washing (a.k.a. pinkwashing or rainbow capitalism) is the marketing practice of […]

Read more
View More