Use sentiment analysis to improve customer engagement and satisfaction

October 19, 2017

Social Media is the wild frontier of the contemporary business world, a place where huge success and catastrophic failure sit side by side, a shifting landscape which requires constant vigilance and engagement to survive. The key to success lies in understanding your customers’ likes, dislikes, wants, intents and even their innermost feelings, and targeting your social media presence so that it gives people more of what they do want and less of what they don’t. Luckily, social media users are constantly expressing themselves: on Twitter, FaceBook, LinkedIn and Google+, existing and potential customers are generating a vast trove of valuable marketing data, but how can enterprise make use of this chaotic mass of opinion?

Sentiment Analysis is the collective name given to a variety of techniques all aimed at mining the gold beneath the social media landslide. The most modern and accurate techniques make use of software based on Natural Language Processing (NLP), which assesses the nature of feelings and opinions, even taking into account tone of voice, sarcasm and irony, before organising statments into positive, neutral or negative sentiments.

The resulting data offers businesses of all kinds an incredible resource, and a powerful insight into what customers really think and really want. But how does this knowledge translate into value for your company?

The first benefit is a deeper relationship with your customers. Using Sentiment Analysis, you can create a targeted customer service approach which gives customers exactly the service they want. As well as satisfying your customers and promoting positive word of mouth, this kind of evidence-based business structuring can lead to major efficiency savings, generating material value for your company.

This kind of empathetic customer engagement is of course highly desirable, but what about more concrete, immediate benefits?

One of the most incisive aspects of Sentiment Analysis lies in the potential for gaining commercial intelligence. As well as analysing sentiments about your own company, you can discover what consumers think of your competition, and develop ways to differentiate and improve on rival services, ensuring that when it comes to customer satisfaction, you are always one step ahead. For companies who wish to build or rebuild a brand, sentiment analysis is an indispensable source of inspiration and a fantastic metric of brand success.

Here are some tips for improving customers’ experience and how analytics will help in maximising customer satisfaction and market share.

1. Product or Service

This may seem obvious on the surface but requires a good understanding of the differences that will help in managing customer expectations. Gone are the days when the only thing that mattered was the ‘3Ps’(Product, Price and Place) when it comes to growing market share. There are many industries that don’t have physical products or inventories. For many Financial and Telecom service providers, ‘Service’ is their product. This allows for greater flexibility and agility for creating service products at different price points that will appeal to the target market segment.


Path Analysis is a good tool to predict failures of consumer premises equipment and/or infrastructures before they arise and helps to proactively manage customer expectations. Here is an example from the telecom network performance. Path Analysis is also good for predicting ‘path to purchase’ and ‘path to out of stock situations’ during customer acquisition stage of customer life cycle management.


2. Product Development

Creation of ‘service product’ at competitive pricing is relatively easy compared with manufacturing of consumer premises equipment which requires breakeven points to be profitable. However, many organisations create service products that are ‘one size fits all’.

Social network analysis can provide clues about a customer’s calling circle and sphere of influence. When combined with Affinity analysis it will provide strong indications of the propensity for the customer to take up a product / service that fits the customer’s needs.

3. Customer Service

When it comes to Customer Service there is no inventory to keep or product catalogue to maintain, but its success solely relies on the skills of well trained staff / partner network who can deliver quality service. In other words, quality of experience (or lack of it!) is realised at the time at which the customer service is delivered – every single time. Paying attention to this subtle difference in ‘service’ delivery and managing customer expectation is the key to keeping a satisfied customer.

Proactive analysis of customers’ sentiment about new product launches and/or issues with a current product or service could be analysed from social media well ahead of a customer’s intention to contact the Call Centre. Discovery analytics performed using Text Analytics and/or Sentiment Analysis functions on Call Centre contact notes and social media data will enable new insights to be gained about competitors as well as own products/service as perceived in the market. A well trained Customer Service staff can use the insights gained from discovery analytics to help improve customer satisfaction.

4. Channel Management

Cutting costs of channel by reducing retail outlets often leads to creation of low cost channels such as telesales and online without considerations for the customer’s preference. Significant number of customers are not willing to accept offers on telesales channels.

Response rates of online channels can also be generally low leading to ineffectiveness of these channels. Geospatial analytics overcome these limitations to spot customers who frequently visit retail stores and those who don’t. Identifying stores that are closer to where customers live provides the opportunity to find stores that are good at selling particular offer and / or good at resolving problems. Those customers who visited your online channel can also be directed to a close by store to bridge the online-offline gap.

Pin It

Comments (0)
» Blog, Uncategorized » Use sentiment analysis to improve...
On October 19, 2017

Leave a Reply

Your email address will not be published. Required fields are marked *

« »