Metrik.world
Sentiment
Sentiment assesses the emotional tone of user interactions on social media. It determines whether content about a brand is perceived positively, neutrally, or negatively. It can equally assess the tone of the content the brand itself publishes.
alternative names
- tone of communication
- polarity (positive/neutral/negative)
units
- Category
- % (ratio)
Additions, clarifications, extended definition
Sentiment can be tracked by category (positive, neutral, negative) or as the share of each category within the total.
Interpretation
Positive sentiment – the audience responds favourably (satisfaction, recommendations, enthusiasm). Negative sentiment – the audience expresses dissatisfaction, criticism, complaints. Neutral sentiment – purely informational messages without emotion. Tracking the sentiment trend over time shows whether communication is building trust or generating negative perceptions.
Watch out: limits and risks of distortion
- Irony, sarcasm, memes – automated sentiment analysis algorithms often misread them.
- Context dependence – the same word can carry a positive or negative meaning depending on the circumstances.
- Representativeness – only publicly available content is analysed (e.g. Facebook groups or direct messages are not included).
- Automated (machine) sentiment analysis is fast but less accurate. Manual analysis is accurate but demanding in time and cost.
Where and how to measure
- Can be analysed manually via the native analytics tools of each network (Meta, YouTube, TikTok etc.) – e.g. by listing all reactions and comments, assigning a tone and expressing the share of each.
- You can also measure the data via third-party tools (e.g. Brandwatch, Emplifi, Hootsuite etc.).
Formulas
share of positive sentiment = (number of positive comments / total number of comments) × 100
Example from practice
A TikTok campaign generated 8,000 comments. Of these, 65% were positive (users praised the product), 20% neutral and 15% negative (criticism of the video's length). Although overall sentiment came out positive, management was alerted to a growing number of negative comments from the younger part of the target group, which led to content adjustments.
