In this guide, you’ll learn how to implement and create Sentiment Analysis with Inbox.
Sentiment Analysis is an Inbox feature that allows you to get the emotion of any piece of content. Agents will be able to visualize the sentiment score of each Inbox ticket, facilitating them to solve and handle tickets in the best way. For example, if a customer is angry, the tone, choice of words, and perhaps the urgency to respond to that ticket are different than that of a happy customer.
How is it possible to automatically interpret the emotion of text?
To automatically interpret the sentiment of a message, we’ll use machine learning—we take away all the complexity involved in building and deploying such a model and let you focus on building your communication flow.
- A MessageBird account
- WhatsApp for Business account or any other channel from MessageBird's offering
- A workflow to handle the messages—Flow Builder
- An Inbox account (optional)
How to implement Sentiment Analysis?
Step 1: Set up the flow
2. Select a flow (or create a new one). For this example, we’re going to create a new custom flow with WhatsApp as a trigger.
3. Add the Analyze Sentiment step from the steps menu on the left. Drag and drop it in the flow. The Analyze Sentiment step form will be displayed on the right side of your screen. From the available variables, drag and drop incoming message to the content to analyze field. Ready? Hit Save.
4. Add a Branch step from the Step menu on the left. Drag and drop it in the flow after the Analyze Sentiment step.
5. From the options, create one branch that checks for the value of our sentiment variable. The branch condition should be checking if the sentiment value equals negative. All the messages that are not determined to be negative, fall under the else branch. In case our system is not certain about the polarity of the message, the sentiment is set to neutral.
6. For this example, we want to notify our lead support agents about negative messages via email. To do this, add a Send an Email step in the branch that checks for negative sentiment.
Step 2: Connect to Inbox
1. All messages should be forwarded to our support agents in Inbox. To do that, go back to your communication flow and add a Create Inbox ticket step after the branch step.
2. It’s possible to see the sentiment shown on each ticket in the Inbox queue by enabling Sentiment Analysis option. Done? Awesome! The final flow should look like this:
3. Publish your flow in the top-right corner of your screen, and you're good to go!
4. Done! To try the flow, initiate a conversation with your WhatsApp or chosen channel.
How to use Sentiment Analysis?
In the Queue, you’ll see an emoji next to the name of the client displaying the sentiment score of that Inbox ticket. For our example, the ticket of John Smith has an angry emoji 😡
Once you open the ticket, you’ll also see the sentiment score next to each reply of the client.
In our example, John Smith is asking for a refund. Instead of us having to spend time analyzing the tone and choice of words of the client, the Sentiment Analysis does this for us, so we can focus on how to best handle the ticket.
It may be best to start with the obvious, an apology!
Sentiment Analysis is here to help you by displaying valuable visual insight that reflects the mood of messages, so your team can effectively decide how to best manage each ticket.
And that’s it! 🎉 Now you know how to automatically analyze the sentiment of customer messages and notify the right people on your team using Flow Builder.