Text analytics

The Text Analytics section shows the topics most frequently mentioned by your customers in reviews and survey responses. For example, “Staff” and “Location”. How Does It Work?

Our advanced text analytics feature combines rule-based modeling of human language with statistical, machine learning, and deep learning models. These technologies work together to break down a piece of text into its parts, identify each phrase that communicates a sentiment, and assign a score to each one.

This enables the computer to process the text and understand its meaning, complete with the writer’s intent and sentiment.

The Text Analytics functionality helps you see the strengths and weaknesses of your business at a glance. It also allows you to look closer at specific topics, better understand your customers’ expectations, and compare how different locations are performing in certain areas.

In order to have better accuracy we built out our advanced Text Analytics based on:

• an industry
• the language of the reviews

We currently have the following industry-reviews language supported:

• Hospitality (English, German, French)
• Healthcare (English, German)
• Automotive (English, German)
• Restaurant (English)
• ECommerce (English)

Main Topics

On the Main Topics page, you can see the total number of times each topic has been mentioned. You can also see how many of those mentions were positive (green), neutral (yellow), and negative (red). At the top of the page, you can filter your results by date or business type.

The “Mentioned In…” column tells you the number of your locations or businesses where the topic has been mentioned in reviews and survey responses.

Actions

The Actions column allows you to dive deeper into each main topic.

Clicking Go To Topics will bring up a list of sub-topics related to the main one. For example, “Room” may have sub-topics like “Bathroom - Cleanliness”, “Room - Size” or “Issues - Noise”.

Clicking Mentions in the Actions column will show you the exact reviews and survey responses that mention the topic. Using the filters at the top, you can narrow your results down by sentiment, topic(s), and date. You can also translate the reviews to your preferred language.

Note: You can find this function on both the Main Topics and sub-topics pages.