Reduce your online churn using text analytics

It is likely that most of you reading this already have an online customer experience program in place, with feedback collection tools quantitatively monitoring your website and apps. In other words, you’ve got all the essential structured data you need to tend to your customers’ needs and make the appropriate adjustments to your online sales funnel…right? Well, unfortunately that’s not always the case.

Digital marketers can collect all the quantitative data their hearts desire and still lack the rich insights they need to improve the digital customer experience. Unlocking these insights requires a much more in-depth approach. As unstructured data (or text analytics) becomes more prevalent in the world of feedback, so do our chances of mitigating customer churn.

The moment you decide to start working with things like open-ended text comments, multitudes of unstructured data start filtering in, related to different parts of your website (e.g. the ordering process, service, my accounts, product content, etc.). Analysing this data can be challenging but also very informative.

In this blog we will give you insight into three questions when it comes to using text analytics to reduce churn:

1. What is text analytics?
2. What are the different techniques you can use?
3. How can you apply these techniques to reduce online churn?

1. What is text analytics?

Text analytics is the gateway for getting in touch with your customers’ thoughts and feelings in a more intimate manner. With this type of qualitative analysis, you’re not only assessing small-scale issues here and there, but also the entire online customer journey as a whole.

How, you ask? Well, text analytics can reveal a plethora of information:

• It identifies the root of the problem or source of satisfaction. With open-ended questions, customers are given the chance to identify what is or isn’t to their satisfaction and why.
• It also enables the emerging trends to surface that many surveys often limit or restrict entirely.
• Issues can be prioritised quickly and efficiently. For example, by identifying the most frequently used words or using word pairing.
• Customers’ ideas and suggestions materialise and are elaborated upon, which can lead to an enhanced digital experience.

Implementing text analytics gives you the power to hear your customer’s voice in all its forms.

---

START YOUR FREE TRIAL

Try Mopinion now!

• Collect unlimited feedback
• ■
• Free 14 day trial

---

2. The text analysis techniques

Frequency of words (word count)

This is text analytics in its simplest form, whereby the topics are counted and brought to the top based on the frequency with which they are mentioned.

Word grouping (or word pairing)

Often times a group of words can provide you with more insight than just one word alone. For example, the words “costs”, “expensive” and “monthly” are grouped together. With this information, it’s pretty safe to conclude that there are many customers that think the monthly costs for one of your products or services are too expensive. But to take a closer look you can always open the individual comments.

Sentiment analysis

At this point, you know which words occur the most often and with which other words they are paired, but is the feedback positive, negative or neutral? More often than not your customers are going to provide you with feedback on topics they feel strongly about. Text analytics can gauge the severity of the feedback based on positive, negative and neutral word usage as well as the sentiment associated with commonly used words.

Categorising feedback comments

Using machine/automated learning, comments can be grouped based on the issues they pertain to. To refine this process, the user can advise the system based on whether or not the comments were categorised in the right manner. Over time, the automated categorisation improves based on this user feedback.

3. How to apply these techniques and reduce churn

Now that we’ve covered all of the bases of text analytics, let’s talk about how you can use these tools to reduce online churn. Our advice? Stick to the nuts and bolts of effective digital customer feedback, which include collecting the right feedback online, analysing the data and turning these insights into action.

Bring in the data

When it comes to reducing churn it is pivotal to collect data from the digital points of contact where you want your customers to convert. Take a close look at the online journey of your customers. Pinpoint the places they may find important and which areas you want to excel in when it comes to the competition. This usually translates to online ordering or registration funnels – touchpoints where you’ll be able to collect the most relevant feedback.

Make an analysis using text analytics

When feedback is flowing in – and especially when volumes start to increase – it is important to analyse and make sense of this data. As mentioned earlier, a good analysis is imperative to identifying the drivers when it comes to churn. Why are people leaving my funnels? Why do they run to the competition? Luckily for you, all of this vital information is stored in the open-ended comments you collect. All you need to do is analyse and act upon it.

So…what are some of the top drivers that may be chasing away your customers?

• Non-responsive websites
• Low quality of customer service
• New products/services introduced by a competitor
• Pricing
• Poor site navigation

But how can you tell if these are in fact the same issues your customers are experiencing? You can run through them one by one but when data volumes are high, this can cost you valuable time. This is when those text analysis techniques become your best friend!

Starting out, you’ll probably want to see which topics arise most frequently. A simple word count could do the trick, but most of the time this is too general. However, taking it one step further by using the word count in combination with the sentiment tool could prove to be of more value.

For example, you notice that the word “login” comes up a lot with a negative sentiment. At first glance, you can see that there’s something going wrong with the login process. But let’s dig a little deeper. By using the word grouping tool, the words “e-mail”, “link” and “password” are all grouped together which may be an indication that your customers are receiving an e-mail to reset their password but that the verification link is redirecting to an error page. Upon further investigation within the individual entries, you discover that many of these sentiments are rooting from a technical error within your system and how it builds up the registration e-mail.

This is, of course, a really great way to solve individual problems. However you also want to keep track of changes on a higher level, too. It’s important to know how issues and their sentiments evolve over time. So, what now? How can we make sense of these insights and do something about them?

Take action and “woo” your customers

In an increasingly digital world, we are seeing dramatic changes in the way customers and businesses interact online. Customers are acquiring higher expectations and given more online alternatives. Essentially they want everything on their terms, meaning we need to “woo” our customers now more than ever. Getting customers to stick by you means catering to their needs, and doing so in a timely manner. Every customer is important, so don’t lose their loyalty!

There are a number of initiatives you as a digital marketeer must put in motion to “close the loop”. Turning insight into actions means understanding internal and external actions and following up using specific timescales. Think about how you and your team can fix the problem and keep your customer in the know. This balance of internal and external communication in response to customer insights will put you on the right path to an improved customer experience and lower churn rates.