Optimization tips > Built-in Sentiment Analysis

An online form with built-in Sentiment Analysis

Using Google’s Prediction Engine to enhance your forms

Ron E, 27/06/2016
Using Google’s Prediction Engine to enhance your forms


If you like using open-ended text elements in your forms, than this tip is for you. If you get lots of daily or weekly submissions - then it's definitely for you.
Some people may argue that it's much easier to analyze and score forms with structured, closed-ended elements like: checkboxes, dropdown menu items and radio button selections. However, in many cases you want to let your customer voice themselves using open text to ask a question, require something or provide feedback. The main drawback usually associated with such input is that your staff needs to manually read each and every entry.

What is sentiment analysis?

What is sentiment analysis?

In general, Sentiment Analysis is about determining attitudes. Some of the common uses of sentiment analysis are for understanding customer’s intent, preferences and feedback. Recent developments in the areas of machine learning, AI, and natural language processing are the driving force behind increasing use of sentiment analysis for a large range of applications.
Joydeep Bhattacharya in his post "Why Sentiment Analysis is a Crucial Part of Customer Acquisition Strategy" claims that "Sentiment analysis is an excellent source of information about audience behavior and can provide us many insights to form a proper and effective marketing strategy".

Text sentiment analysis has been around for a while. It is based on advanced technologies in the areas of machine learning, artificial intelligence and natural language processing. Till now it was reserved for larger corporates who could afford paying for development around these technologies. Read here more about FormTitan’s sentiment analysis for online forms.

What are some use case examples?

What are some use case examples?

Let’s look at two examples where open-ended text elements make sense: First let's consider an online support form in which you let a customer describe the issue they have and ask for assistance. As a support manager, your main objective would be to flag the issues or the customers who are least “happy” about your service and have a high attrition risk. By applying FormTitan’s sentiment analysis to responses you can automatically know if their sentiment is “positive” or “negative” along with a numbered score, without having to read the text itself.
The same applies for online surveys or feedback forms. FormTitan’s Sentiment Analysis automatically scores and categorizes answers as “positive” or “negative”, letting you get a feel for the responses in your aggregate report, before actually reading them.


Does it work?

Does it really work?

Sentiment Analysis works. It's not 100% reliable and it errs but even if you get sentiments with a confidence level of 70 or 80% it’s a lot more than nothing. Google's prediction API which is the engine behind FormTitan’s sentiment analysis is one of the best engines in the market and it keeps improving. And finally, you don't need to totally rely on it, If it helps you deal quickly with customers at risk, it's ok if once in a while you get a wrong trigger. Try it out and decide for yourself. In case you have not done so yet, feel free to try our online form with sentiment analysis live demo

What else can you do?

By using FormTitan’s conditional logic features you can set triggers and automatically escalate customers based on their sentiment. E.g. you can select to email your head of support all customers with a negative sentiment and a score higher than 50%. Read here more about what you can do with our conditional logic.