Glossary

Sentiment Analysis

What is Sentiment Analysis?

Sentiment analysis is the use of natural language processing (NLP), text analytics, and artificial intelligence to identify, extract, and classify the emotional tone expressed in text data. It determines whether a piece of text — such as a survey response, customer review, or social media post — conveys a positive, negative, or neutral sentiment.

In the context of employee and customer experience, sentiment analysis transforms unstructured text feedback into actionable insights at scale.

How sentiment analysis works

Sentiment analysis typically follows these steps:

Data collection: Gathering text data from sources such as customer feedback forms, pulse surveys, support tickets, reviews, and social media.
Text processing: Cleaning and preparing the text by removing noise, handling misspellings, and tokenising words.
Classification: Using machine learning or rule-based models to classify text as positive, negative, or neutral. Advanced models can detect more nuanced emotions (frustration, gratitude, confusion).
Analysis and visualisation: Aggregating results to identify trends, patterns, and areas requiring attention.

Applications of sentiment analysis

Sentiment analysis is used across many areas:

Customer experience improvement Analysing customer feedback at scale to identify common pain points and satisfaction drivers. Read more about data analysis.
Employee engagement monitoring Understanding how employees truly feel through open-ended survey responses, supplementing quantitative engagement scores.
Brand monitoring Tracking public sentiment about the brand across social media and review sites to manage reputation.
Product development Mining customer feedback for feature requests, complaints, and suggestions to inform product roadmaps.
Early warning system Detecting shifts in sentiment that may indicate emerging issues, such as rising DSAT or declining employee morale.