Improved sentiment detection model with a new sentiment "promoter"
We have released a new sentiment detection model with improved performances as well as a new sentiment,
promoter is designed to detect extremely positive messages from your customers. Some examples of use cases that this sentiment covers are messages from customers expressing:
how much they love your products
how many compliments they received thanks to your products
that they are actively sharing their love for your brand among their friends.
that they are likely to be good promoter or advocate for your products
How to use it?
You can create a new rule to auto-tag these messages or integrate it in an existing one!
For more information about how to use the sentiment detection feature, have a look at Automate your work using Customer Sentiments.
⚠️ Upcoming changes to
Quick heads up, we will also be making some modifications to
sentiments in the near future to make them easier to use and understand:
🗃 The sentiment
need-humanwill be deprecated.
By default it will be replaced by the
urgentsentiment in your rules.
If you don’t want
need-humanto be replaced by
urgent, we recommend that you modify the rule to trigger on keywords instead.
✍️ The sentiment
menacingwill be renamed
threateningreflects better the meaning of this sentiment: a message containing a business threat. The threat can be a lawsuit, a complaint with BBB, leaving a bad review…
🔂 The intents
feedback/negativewill be merged into a single
They will be automatically replaced by a combination of the new
feedbackintent and either the
negativesentiment in your rules.
Did you like this update?
Leave your name and email so that we can reply to you (both fields are optional):