Contacto

ING

ING Direct Spain

ING DIRECT leverages unstructured feedback to improve user experience

MeaningCloud is integrated into the bank’s customer experience management processes to get a complete view of its
customers’ opinions.

How can a big company know what customers say at all touchpoints? ING DIRECT Spain, the bank most recommended by its customers in Spain, needs to understand open-format comments coming through multiple channels such as surveys, contact centers, and personal interviews. For this purpose, they have integrated MeaningCloud’s semantic analysis APIs into their customer experience management processes and systems to be able to detect problems and opportunities for improvement in any source of user feedback.

ING provides retail and commercial banking services. It operates in more than 40 countries thanks to its nearly 51,000 professionals and has more than 35 million customers. ING DIRECT is the largest online bank in the world and offers financial solutions for individuals, freelancers, and SMEs.

 

THE CHALLENGE: understanding customers’ unstructured feedback coming from multiple channels.

As a company committed to improving customer experience, ING DIRECT has been employing tools to perform structured analyses of the voice of the customer for a long time. These tools typically make use of customer satisfaction surveys with closed answers, which are easy to analyze and understand. However, the evolution of these channels has increased the weight of unstructured information such as answers to open-ended questions in web forms or transcripts of contact center interactions and office conversations. This feedback has a special value: customers express it freely, without the limits of closed answers, and they talk about the issues that really interest them, using their own terms and with a higher degree of spontaneity. When processed, open-format comments provided by customers help determine the root causes underlying numerical ratings and hence can generate highly valuable information for businesses.

ING DIRECT used to analyze this information through manual processes, but when the volume increases, these processes become slow, inefficient, and expensive. For this reason, they needed a technology able to analyze unstructured comments coming from different sources that would easily integrate into their existing customer experience management processes; they chose MeaningCloud.

 

THE SOLUTION: text analytics engine adapted to its scenario and integrated into its existing systems.

ING has an integrated system that collects its customers’ multi-channel input, including open-format comments. MeaningCloud’s APIs have been integrated into this system, providing a text analytics engine that enables the system to extract information automatically from those comments: mentions of brands, competitors, etc.; conversation topics; and their related polarity (positive, negative, or neutral sentiment). In particular, to adapt MeaningCloud’s functionality to ING DIRECT’s scenario, we developed a multidimensional classification model with five axes related to the company’s business: channels (web, office, telephone, etc.), comments, products, operations, and sentiment.

Besides this type of analysis with predefined categories, text clustering technology was also used to discover non-predefined (and unexpected) conversation topics, which enables ING DIRECT to identify the “new voice” of the customer (including unknown topics and emerging trends).

Thus, the unstructured comments turn into structured data, which is easier to monitor, analyze, and even include into the corporate CRM (like they did).

Interestingly, in an initial phase the integration was carried out manually using the MeaningCloud add-in for Excel. This tool allows you to perform text analytics directly in a spreadsheet: open-format comments were pasted into an Excel spreadsheet and sent to MeaningCloud’s APIs. The APIs extracted the most valuable pieces of information and returned them to the same sheet, where they were analyzed and used for decision-making. This process served to refine the analysis processes until they were valid enough to enable automatic integration.

 

THE VALUE: the real-time collection of actionable information from a more complete voice of the customer.

MeaningCloud’s semantic APIs allow ING DIRECT to turn its customers’ open-format comments into actionable information (topics, mentions, sentiment). This allows them to detect situations related to a specific customer (e.g. risk of losing the customer) or a specific channel (e.g. accessibility issues in an oice), and opportunities for improvement that can be applied to all customers (e.g. web contracting processes or new product needs).

 

CONCLUSION:

MeaningCloud’s semantic APIs allow ING DIRECT to turn its customers’ open-format comments into actionable information (topics, mentions, sentiment). This allows them to detect situations related to a specific customer (e.g. risk of losing the customer) or a specific channel (e.g. accessibility issues in an oice), and opportunities for improvement that can be applied to all customers (e.g. web contracting processes or new product needs).

 

ABOUT MEANINGCLOUD:

MeaningCloud is the easiest, most powerful and affordable way to extract meaning from any kind of unstructured content, from social conversations to internal files. Use its plug-ins to easily and accurately perform text analytics in spreadsheets, classify texts graphically and perform sentiment analysis on websites, and embed semantic analysis into your applications risk-free through its pay-per-use web-based APIs. MeaningCloud is a Sngular company.

 

MEANINGCLOUD CUSTOMER CASE STUDY: ING DIRECT SPAIN

Company Profile:

  • Bank

Semantic Analysis Problems:

  • Understanding open-format customer comments coming from different channels
  • Integration into the existing analysis processes and systems

MeaningCloud APIs Used:

  • Text Classification
  • Text Clustering

Amount of Content Analyzed:

  • Tens of thousands of comments per month

Type of Data Analyzed:

  • Unstructured customer feedback: transcript of person to person or telephone conversations, open answers to web forms

Results:

  • Their existing systems and processes have become capable of analyzing text automatically
  • No voice of the customer item is lost
  • The customization to their scenario allows extracting insights that are relevant and actionable