Text Analytics Changes Big Data into Big Revenue Opportunities

There are two common ways for businesses to increase revenue; ether by attracting new customers, or increasing loyalty of existing ones to improve their purchase rate.

In highly competitive markets, the acquisition of new customers is quite expensive and risky, because not all the investments in marketing have a positive ROI and cutting prices will only get you so far. However, customers who had a good experience with a brand in the past tend to spend 140% more than those who had the poorest past experience. Moreover, 86% of all current customers are willing to pay more for better customer support. 

Those figures give us strong arguments to re-affirm the importance of providing an excellent customer experience.

In this post, we want to show you how different sectors can extract insights and trends by using text analytics tools.

  • Hospitality:

Text analytics tools have been widely used for a part of the hospitality sector, such as hotels or airline companies. However, in restaurants or food and drink chains, the applications of these tools are not that popular. 

One of our clients is one of the largest and trendiest hospitality company on the Western Coast. It had the challenge of analyzing all the customer feedback at a store and a global level. The former approach of manual coding became obsolete and impossible to use as the company grew; It was consuming too much time and human resources in just reading the data. Because of this, the company was not reacting fast enough to the insights presented in its customer feedback.

By combining text categorization with sentiment analysis, the company was able to detect all the topics of everyone's response automatically and classify it into positive, negative or neutral. Vital information that the company was taking forever to record and analyze became available in short time.

Thanks to Bitext’s embedded and interactive visualization, were custom filters can be applied, the company can visualize the global chain results and the results at store level. All with much faster turnaround cycles that allow taking immediate actions. 

If you want to know more about this example download our case.

  • Retail:

Due to the growth of some marketplaces like Amazon, Asos, or Alibaba, retailers are not only able to offer their products to more consumers but receive much more feedback. Before these marketplaces existed, companies were receiving customer’s opinions by surveys or individual emails to the customer support team. Today, around 92% of consumers check product reviews before purchasing any item.

Studies have shown that customers check an average of 10 sources before making a purchase; then how is it possible to manage the information coming from so many different sites?

Another significant rate, is that 95% of unhappy customers will re-purchase from a business if they can solve their issues quickly and efficiently; but again, how is it possible to detect these issues?

By using a variety of text analytics tools, it becomes possible to solve both issues:

  • Thanks to Bitext’s Customer Experience tool it is possible to analyze text coming from different sources just by adding all the reviews to an excel sheet.
  • The Automatic topic and sentiment detection allow retailers to see what are the customers talking about: packaging, prince, quality, fabric and if the comments are positive, negative or neutral.


  • Technology:

Software companies care a lot about their user’s customer experience; nowadays it is one of the most competitive sectors and there is always at least a couple of products that cover the same need. Therefore, businesses must know what their users think about their products and extract actionable insights.

In one of Bitext’s latest projects, a leading software company needed to analyze the answers from thousands of specialized surveys generated by top business clients. As many other companies, to analyze the feedback our client was using manual coding with a specific coding plan.

The main issue with this methodology is that it uses many resources and it is time-consuming. Reading all the answers manually takes time away from discovering new insights.

With our platform and some assistance from our consultancy team, our client transferred their own coding plan in to our automatic categorization platform and achieve 75% accuracy out of the box, and by adding complex rules the company can even achieve rates as high as 90%.  

Bitext’s flexible visualization allows you to extract insights and discover new patterns and trends, so instead of spending your time categorizing you use it on what really matters: analyzing.

PredictiveAnalyticstoday users' have valued our services with 8,2/10 valuating our multilinguality (+20 languages available) and how useful this can be for VoC and survey coding projects.

Download our use case


Subscribe Here!