Guide
What is online sentiment analysis?
Online sentiment analysis measures whether public online posts about a topic are critical, supportive or neutral, and how that mix changes over time. It shows what people say online, not what everyone thinks.
What it can tell a newsroom
- When and how the online mood around a story shifted
- Which arguments dominate on each side
- Where the conversation happens: which platforms, communities and channels
- How online mood lines up with news coverage and events
What it cannot tell you
- What a whole population thinks. People who post are not a representative sample.
- A margin of error. That requires a random sample.
- Exact numbers per post. Automatic labels are checked on a subset, so trends are directional.
How to read the numbers
Look for three things in any sentiment analysis: how many posts were analysed per platform, the period, and the caveats. A statement like "critical posts rose from 14–41% a week to 63–73%" is meaningful; "63% of fans think" is not.
How to write about it
Say "online sentiment turned after…" or "63% of analysed posts were critical", and cite the platforms, period and number of posts.
Get the First Analysis Free Delivered within 3 working days.