Subscribe to Our Weekly Blog Updates
Get the best in Customer Experience content delivered straight into your inbox.
You have successfully subscribed
Customers are people and people are passionate. Whether they are talking about a recent service experience or a product they love or hate, open-ended survey questions and reviews allow people to go beyond a score or a rating and express their passion (or lack thereof) for you and your products. And that feedback can be just what you need to make real improvement to your business.
According to a recent Medallia analysis of over 80,000 English sentences randomly chosen from various sources of customer feedback, 80% of the text-based feedback contained at least one polarized comment, either good or bad. The analysis also showed that the feedback is rich with real insights about specific business issues. Consider the example below from a review site. The text gives the company (in this case Heaven Palace) rich insight about why the customer, while happy with the stay, was unhappy with the brand: a hotel company policy. In this case, the insights can be used to drive real improvement to both process and policy.
The sheer volume of text-based customer feedback that large companies generate in both social media and surveys is daunting. It can run to millions of records a year. When the volume exceeds even 100 records, it becomes difficult to categorize comments and analyze them for recovery actions and improvement opportunities. In large companies, then, it becomes critical to automate the analysis of sentiment and the identification of detailed insights often “hidden” inside text-based feedback.
Text analytics can improve your feedback program in the following ways:
1) Increase insights with fewer questions: Consumers suffer from survey fatigue; research shows that response rates have dropped precipitously in the last 20 years, from about 20% to around two percent today. Medallia has found that shorter surveys generate better response rates but to get at least the same level of insight with fewer questions, companies need to ask more open-ended “what else should we know?” questions that generate text-based responses.
2) Get to the root cause: While feedback scores and ratings give you a barometer of your feedback, scores typically can’t tell you the “why” behind the score. Follow-on ratings questions can deepen your understanding of the “why”, but usually not as well as text-based feedback. Open-ended comments from customers usually provides the detail needed to identify the root cause of an issue so teams know how and where to improve.
3) Get timely insights. Look for a feedback program that offers native text analytics (Medallia does) so there’s no delay in getting insights. Systems with non-native text analytics force you to wait to understand the root cause of an issue: the system that captures feedback needs to send the data to a text analytics system that in turn takes time to analyze the data before sending it back. The added time not only delays insights and actions to improve but also can create more unhappy customers.
4) Identify emerging trends: Most feedback programs and even review sites have a specific set of questions they ask of customers. In order to get a view of satisfaction over time, those questions rarely change and are limited to scores and ratings. While the responses can show trends in opinion about the question topics, they can’t show new and emerging trends that aren’t covered by the questions. Text feedback fills this gap. If a customer doesn’t see a question that allows him to give the feedback he has – for example, if questions ask about hotel check-out times but the guest really wants to complain about bill accuracy – open-ended questions allow the customer to still give his feedback, only in text format. By using text analytics you can catch emerging issues and act on them before they escalate.
To learn more about how Medallia can help you understand the hidden jewels in your customer feedback, watch our recent webinar by Medallia’s Chief Scientist for Text Analytics, Ji Fang.
Photo credit: krystian_o