When it comes to extracting insights from unstructured text there is the hard way and then there is the organized way. Follow these principles and you will be extracting actionable insights and sharing them with CX stakeholders in no time.
In two days my six-year stint in Etuma will be over. We've gotten so much done and I've had a great time. Etuma is a fast-growing company with excellent partners such as Qualtrics and Questback and dozens of large companies as customers.
I've learned a new industry and become somewhat of an expert in analyzing customer and employee feedback. I've written close to 120 blog posts, half a dozen white papers, conducted many webinars and talked to many companies.
I just went through all our customer projects during the past six years: I've been involved with 159 companies' customer or employee feedback analysis. Here is what I have learned about CX text analytics process and CX text analysis business in general.
Donald Rumsfeld imortalized the unknown unknows in a press event during the first war with Iraq.
There are things that only you know, there are things that only the customer knows and then there are things that only your competitor knows. You need to build a data gathering and analytcs system that can capture and uncover the insights for all these three dimensions.
1. Close the loop!
Ensure that the actionable insights 'flow' to the CX stakeholders, and that they can interpret the data and deduce concrete actions from it. In practise this means that you have to create a dashboard, which indicates in close to real-time what kind of problen clusters are emerging and what is the root-cause for those patterns.
A rule-based semantic analysis system understands language structures and the relationships between words. It has a massive set of pre-configured rules that enable automatic and accurate real-time analysis. Developing a rule-based language analysis system takes a considerable amount of work and time, sometimes stretching even up to dozens of years. Very few companies have the skills or resources to develop such a system.
We've seen hundreds of customer experience feedback analysis projects during the past seven years. Certain mistakes or rather misconceptions show up in most projects. One of them is to analyze text in vacuum. Knowing what kind of person left a comment makes the feedback analysis results more valuable. The other is to leave the analysis results and insight extraction solely to the analytics team. You need to ensure that there is an insight distribution system that customer experience stakeholders actually use. They also need to close the loop, that is, fix the underlying issue.
CX text analysis is a an application, which connects customer and employee voice to company's business strategy, customer value proposition and corporate values.
Loyalty management surveys, like NPS, have increased the volume of incoming text comments, and an ever larger share of customer complaints arrive via email, web form and social media. All that feedback is unstructured text.
There are four ways to create the categorization system (Codeframe, Taxonomy). But whatever way you choose, make sure that the system takes into account both top-down (what the management wants to see) and bottom-up (what the text makes possible) approaches. Well working categorization system requires a couple of iterations and is a balance between these two views.
Designing and implementing a uniform categorization system might seem like a daunting task but the benefits are clear. Uniformly categorized customer comments have the power to transform your organization.
Customer feedback taxonomy (aka Codeframe, Categorization system) enables you to report verbatim analysis results in the same way as structured information (like sales figures) is reported. It creates a common language within a company and brings customer’s voice into the decision making process. It also has the power to transform the organization to be more customer centric.
Here are the most important categorization system requirements.
Capture all relevant words, phrases and brands from open-ended feedback.
You cannot analyze customer feedback without categorizing it. This categorization has to be done systematically, relevantly and consistently. Your categorization system (Codeframe) needs to be uniform across the organization otherwise the text analysis results cannot be used in top management reporting.
Signal categorization turns open-text into statistical information, which enables you to
- Detect patterns (trends, weak signals);
- Benchmark organizational units; and
- Distribute the customer comments in real-time based on customer experience stakeholder roles.
There are four ways to categorize feedback:
If you get only few hundred Signals per month, this is a manageable method. With higher volumes this task becomes slow, expensive and the results are inconsistent. Humans can handle only about a dozen categories. This means that e.g. all weak signals and most emerging trends belong to the “other” category.