Customer Service, E-commerce, Automated Natural Language Processing

Updated: Jun 20

One of the most important differentiators of value in the retail industry, especially e-commerce, involves delivering a high quality customer service experience, consistently.

Client Requirements

Our team of data scientists worked closely with the customer services team for one of the fastest growing online retailers of bespoke furniture and home-ware in Europe. Since the client began operations in 2010, they have increased sales by over 49% CAGR (compound annual growth rate), growing from a small online retailer to an international company that generates more than GBP £173 million in sales revenues per year, employs over 489 people, and has operations in 11 European countries. The senior leadership team wanted to find a cost-effective way to build capacity into their customer services operations by automating routine tasks in their after-sales services activity.


Our Work

Our team began by analysing the customer services database, which contained over 17 million records, 46 thousand fields, and 782 billion data points. The analyses involved a variety of feature extraction and feature engineering exercises to understand workflow activity and communication patterns in the database. By combining a combination of text mining, classification algorithms, decision tree analysis, and business rules, the team came up with a multilayered solution that could accurately classify up to 53% of all incoming emails into 2 categories:

  1. Simple to solve; and,

  2. Complex to solve.

All the emails that could be classified as simple to solve were analysed even further using Natural Language Processing (NLP) techniques and matching algorithms that mapped customer requests for information to predefined email response templates. For example, when a customer sent an email asking when order would be delivered, our solution would classify the customer request as simple to solve then analyse nature of the request and automatically collect the relevant information from the customer service database. From there, the solution would insert the date into an email template for delivery date requests.


Our Results

Within 5 weeks, the team delivered a complete solution that identified over a 100 categories of email communications that could be automated. This helped reduce customer waiting times by 39% and generated efficiency savings of GBP £250 thousand per year. The solution also translated to better employee experiences and greater customer satisfaction ratings because the customer services team had more time to solve complex queries for customers.


Topics Associated with this Project

#CustomerService

#ECommerce

#StatisticalDataMining

#TextMining

#NaturalLanguageProcessing #NLP

#ClassificationAndRegressionTrees #CART

#DecisionTrees

#RandomForests

#LogisticRegression

#NaiveBayes

#SupportVectorMachines

#CustomerLifetimeValue

#QueueingModels

#OperationalExcellence

#ProcessAutomation

#MachineLearning

44 views

®

EVIDENCE BASED SOLUTIONS

ABOUT

+44 (0) 203 920 9378

info@authenticevidence.com

278 Northfields Avenue, Ealing, London, United Kingdom

SOCIAL
  • White LinkedIn Icon
  • White Twitter Icon
  • White Facebook Icon

© 2020 by Authentic Evidence Limited.