All Collections
Analysing results
How to use Sentiment Analysis?
How to use Sentiment Analysis?
In this article we’ll tell you more about what sentiment analysis is and how to use it.
M
Written by Melanie Pieters
Updated over a week ago

What is Sentiment analysis?

Sentiment analysis is a natural language processing (NLP) technique used to determine whether data is positive, negative or neutral. It is often used to analyse open text responses and can help brands understand how people feel about their brand, what their opinion is on something or what the customer needs are.

How does it work?

  1. For all of your open ended questions (on English surveys, send after 1/1/2021) you will see the option to “View text analysis”

  2. When you click on this link you’ll be taken to the text analysis page of your survey. It can take up to 30 sec for the analysis to be run, but this will only be the first time you access this page.

  3. On this page you will see:

    1. On the left: a list of all the keywords that the model has recognized and grouped together, including a count and the % frequency, which is based on your total number of responses.

    2. On the bottom right: a word cloud generated on your top keywords

    3. On the top right: the option to display sentiment analysis for this question.

    We don’t automatically show sentiment analysis for all open ended questions, as for a lot of questions such as “list the brands you know”, doing a sentiment analysis isn’t relevant. It’s mostly useful for questions that asks about a customers opinion or feeling.

  4. When you switch on the sentiment analysis, you will see the total sentiment of your question. This is calculated based on the individual original responses that your users have given and will show how many were positive, negative or neutral.

  5. You will also be able to see the sentiment for the original responses. You can see the original responses by clicking on the arrow next to the key word. You’ll see a green dot for responses that were classified as positive, yellow for neutral and red for negative.

  6. You can choose to hide a certain key word by hovering over it and clicking on the eye icon. This will update both the word cloud as well as the sentiment analysis.

Limitations

There are a couple of limitations that are important to point out.

  • Text & sentiment analysis is only supported for English surveys that have been purchased on or after 1/1/2021

  • Text analysis is not included in the export functionality

  • You can’t make any manual changes to sentiment yet

Why is this labelled as “Beta”?

Text & sentiment analysis are features that are still actively being worked on, so you might see a few things change. As we want to continue improving our text analysis tool, we’re looking for your feedback! If you have any thoughts to share (what you like, what you dislike, what you're confused by) please contact your CSM or let us know via the in-platform live chat.

Did this answer your question?