Customer loyalty is more resilient than you think. It can often seem like someone has decided to stop purchasing from you because of one negative experience, whereas in reality it’s usually the culmination of a series of events - a late delivery here, a faulty product there - and there were signs that would have been easily spotted, but that you failed to because you weren’t paying attention.
Obviously, this is a bit unfair. Any retailer who deals with more than 100 different customers, which basically means any retailer that’s still in business, is not going to be able to keep a check on every customer. It is, therefore, often extremely difficult to pinpoint what caused a customer to become disenchanted, and without knowing the cause, it’s clearly going to be far more difficult to rectify in the future.
In order to better gauge customer loyalty, retailers are increasingly turning to analytics. They are doing this in two ways, through both structured and unstructured data.
Structured data is easy to analyze, and can provide some insights. It’s important to use data to establish signs as to what could be pushing customers away - site visits, survey responses, social media interactions such as likes and followers, and time spent on page. However, this data can only really take you so far. It does not let you hear the voice of the customer, and does not explain why these things are happening.
Analysis of unstructured data, specifically text analytics, on the other hand, can open far more doors to insights. For example, if a restaurant’s customers rank their dinner poorly, looking at what people have tweeted or said in reviews is the only way to identify the underlying problem. According to Ipsos-Mori’s report, ‘Ipsos Guide to Text Analytics’, demand for text analytics grew 70% last year. The co-author, Jean-François Damais, noted that, ‘Virtually all market research projects involve some analysis of text. In the customer experience space in some of our key markets (i.e US, Canada, UK, France, Germany), I would say that 70-80% of research projects require significant Text Analytics capabilities to extract and report insights from customer verbatim in timely fashion.’
A wealth of tools able to analyze unstructured data have arisen in recent years, with firms such as BrainSpace now able to process millions of documents in hours with increasing sophistication, language capabilities and accessibility. There are often thousands or even millions of potentially relevant posts in social media, and these hold a treasure trove of knowledge to those who have the ability to look for it. These can show signs of dissatisfaction before this has manifested itself into someone deciding to shop elsewhere. For example, if a customer posts on Facebook complaining that your page is difficult to navigate, you can jump in to help them before they decide to shop elsewhere. If text analytics suggests that a lot of people are agreeing, it will likely explain why you are experiencing poor clickthrough and poor sales, and the page will need to be rectified.
Analysis of text can also help to identify customer trends, showing the popularity for a new product perhaps before it’s been released and sales figures can be taken into account. Text is a far better gauge of excitement than page clicks as you can identify what it may be specifically that is causing the excitement, which enables marketers to adjust their campaigns accordingly. Ultimately, structured data can provide an outline of your customers, but it is text analytics that colors them in.