Innovative Capabilities

We have a rich experience helping our clients fundamentally transform certain aspects of their operations based on developing, prototyping and implementing a number of new capabilities. These have ranged from next-generation matching engines, innovative models, developing comprehensive customer platforms, next-generation data strategy and others.

Sample Cases

Customer Data Platform – 360º Customer View

Our client needed a more robust data platform to centralize and organize their massive amounts of customer data. The also required a useable interface for marketers to plan a campaign’s initial counts, review past campaigns, and view the contact history across all channels for specific customers. The overall goal was to provide marketers with personalized views on their customers – starting at contacts. More specifically, the objective included:

  • Creation of a Big Data Repository as a central source of customer data
  • Development of user-friendly modules for marketing campaign results reporting, contact history
  • Delivering cross-channel communications prioritized by individual customer relevance
  • Messages are delivered to the right CM through the right channel at the right time

We accomplished all the above objectives on time and with quality, developing the Big Data repository in Hadoop framework, designed all the modules, prototypes and wrote requirements and produced Big Data Email Model to support their cross-channel communication objective.

Merchant Data Quality Enhancements

We were part of an engagement with a top issuer with a closed-loop merchant network where we focused on enhancing the quality of the Merchant Database with a number of customer-facing applications, such as e-Statementing. We were tasked to evaluate, track and improve the quality of Merchant Database with emphasis on merchant names, addresses and others. We also focused on Identifying Client’s merchant share for applications like merchant acquisition, suppression identifications, merchant pricing and others. We conducted extensive analyses (segmentation, stratification, online cross-validation) and implemented a number of capability modules such as developing Data Enrichment Algorithms, Enhancing the Data Matching capabilities and others. As a result, we were able to:

  • Produce and set-up monthly reports for merchant data quality and disputes on merchants
  • Improve data quality of key data elements for e-Statement and disputes reducing efforts
  • Identify Client’s merchant share at merchant level

Next-Gen Data Strategy

Our client, a top card issuer in the US and globally, was leveraging numerous data sources for key decisions across the customer life-cycle (acquisitions, underwriting, customer management). Over time they had been accumulating more and more information and we were brought-in to create transparency of usage, data sources value, redundancies and overlaps. We conducted extensive diagnostics of applications, down to model /coefficient level in many cases and built comprehensive usage maps. Leveraging matching we evaluated overlaps between sources, within and especially across vendors and based on diagnostic models we evaluated the incremental value of data of a number of sources. We created a blue-print of the Next-Generation data strategy and assisted in contact re-negotiations with data vendors based on model-based detailed data source value assessment.

Customer Experience: Feedback Classification Model

Our client, a top card issuer, receives thousands of feedbacks / complaints each month through three channels – phone, chat and mail. We were tasked to build a text-mining based model to classify complaints into appropriate themes. Our client aimed at improving the customer experience, based on sharing insights with different business stakeholders and driving change. Another application was to produce report for regulators (CFPB). We built complaint classification model using text cleaning, text parsing, text structure/pattern analysis. We conducted extensive model validations. Improved around 30% in model coverage and 20% in model accuracy compared with existing rule based classification system. Created a process for classification and distribution of customer feedback to relevant business units / groups / stakeholders.