NPS is under pressure: as the scores start to tail off, some firms are concluding it is not up to the task of meeting today’s rising customer demands. The response from Safelite AutoGlass, USA’s largest windscreen repairer and a subsidiary of Belron International Ltd., has not been to ditch the method, but to focus on that often overlooked ‘why?’ question in its NPS surveys. The result, according to Kellan Williams, Customer & Quality Analytics Manager at Safelite, has been “a complete paradigm shift” in the company’s decision making. This is helping Safelite to maintain and even raise its NPS.

Digging into comments from customers in NPS surveys is a laborious task. The ‘why did you give that score?’ question has been deprecated by many NPS practitioners. As a provider of text analytic software, I see numerous companies who have only ever managed to analyze the reasons given in a superficial way. Too often, the answers simply accumulate, unexamined. At Safelight, this changed when they started using text analytics software and applying sentiment analysis and automated text categorization methods to these open-ended responses.

“From a cross-industry perspective, we have a very high NPS score, so being able to improve on that becomes difficult,” Williams tells me. “Text analytics has allowed us to focus on key areas where we can improve customer experience organically, versus trying to improve a bunch of things and hoping that improves the customer experience.”

Using an automated method was essential. The NPS surveys alone yield over 500,000 comments each year and all of these are read and processed using a battery of different analytical techniques that Williams has developed using Ascribe InspectorTM. However, as he says: “it is not enough just to analyze the text and create sentiment topics. You need to know how to make the data actionable. This is where the skill comes in.”

One method Williams has adopted is to look at ‘over-indexers’. “We might find we are overperforming in some areas when compared with the total company and underperforming in others,” he says, “so this tells us the areas we need to work on.”

Another approach is to look at the sentiment groups that are affecting the NPS score the most and dig into the specific comments made by customers, to provide the full context. This way, individual managers can start to understand why a sentiment category in, say, Baltimore is up or down compared with the rest of the region.

What managers see is a dashboard that shows the topic or sentiment groupings, which can be sliced and diced by other variables such as geography, demographics or the specific services provided. He says: “This way we can really understand what makes somebody a promoter and what negatives drive somebody to be a detractor.”

Unlike some text analytics methods, which essentially perform a new analysis each time, Williams advocates applying the same analytical framework, improving this incrementally every month. “Being able to have a consistent methodology and algorithms that are applying the sentiment is really what has given me the ability to have actual insights out of the data,” says Williams.

He concludes: “We are definitely making decisions based on text analytics that are improving our customer experience. We have the whole picture of our customer and we are not doing anything extra to touch that customer. It’s data we already are gathering. It’s like doing a bunch of research projects or focus groups within one dataset.”

Six key tips for making customer feedback more actionable

  • Capture and process all of the ‘reasons why’ questions alongside other survey data
  • Use sentiment analysis to get to the feelings and concerns of your customers
  • Develop a consistent analytical framework that organizes responses into topics and sentiments, and let this grow organically over time
  • Slice and dice your data to understand which customer segments are the most satisfied or dissatisfied
  • When you have identified the key sentiment groups that affect your NPS, review the exact words of customers within those groups
  • Use dashboards to bring the findings to decision-makers, and let them interact directly with the data

Rick Kieser

Rick Kieser is CEO of Ascribe, a market-leading provider of online software solutions for managing verbatim texts. He is an expert in software as a service text analytics companies and is a seasoned executive leader with 20 years of operating, management consulting and venture capital experience.

About Ascribe

Founded in 1999, Ascribe is the leading provider of verbatim text analytics solutions for the world’s most recognizable brands and research firms. Clients spanning 57 countries depend on Ascribe to gain real-time, accurate and actionable insights into the feelings and experiences of their customers. Ascribe analyzes more than 300 million open-ended customer comments per year, captured across a broad range of channels and in a myriad of languages. With Ascribe, companies make better, more informed decisions through a deeper understanding of their customers and markets. Visit Ascribe at

Post Views: 1359