Improve Customer-Experience Outcomes with AI

Mike Romeri, CEO       September 2019                                                                   Download pdf version  

As we all know, pleasing customers is the goal of most businesses. And marketing organizations have sought to understand customer “needs and wants” according to segments and personas. Customer service has championed “360-degree views” as a way to understand customer experiences (CX) better and identify how to deliver a “seamless” experience across all channels to improve satisfaction and retention. Despite all of the past strategies and the investments made in IT, however, an in-depth view of how a customer experiences a business remains elusive for many reasons.

Identifying and collecting the right data remains a challenge even for large enterprise companies. Many struggle to understand what data they need to improve customer experience and how to collect it in order to apply analytic platforms and models. To complicate matters even more, new data privacy laws, a growing reluctance to “join” or become “members” of loyalty and other programs, and the monetization of data by data brokers have slowed the acquisition of data … but not stopped it.

Fortunately, valuable customer data that are available via internal customer touch points are growing exponentially, thanks to digital solutions like online shopping, brand apps, integrated POS systems, digital marketplaces, and more. Companies also have an opportunity to augment their internal data with external data such as location, local events, and weather to provide context to every occasion that a customer interacts with their business along the customer journey.

The increase in the use of digital channels and the growing availability of external data for context, combined with the next generation of CX analytics and AI, are beginning to allow for the development of additional and market-differentiating insights on customer segments, personas, and behaviors.

Currently, most companies are still more focused on collecting rather than using data effectively, since the reality is that the vast amounts of data and technology needed to analyze data overwhelm the time, skills, and resources of their IT departments—resulting in little to no tangible value received from the CX investments that have been made.

What can CX data analytics and AI do to help businesses develop a 360-degree view of their customers and provide a seamless customer experience across channels?

The answer is automation, precision, and optimization of solutions for your complex business use cases.

 

As companies collect more and more data about their customers, analytics and AI can be used in a number of ways to inform decision making, including these examples:

  • Customer personas are used to classify customers by relevant attributes like age, gender, and occupation. Analytics can be used to identify groups of customers whose purchasing behavior is similar even if they don’t fit neatly into any traditional classification.
  • Data about these “analytic personas” can be mined to precisely identify the common characteristics associated with favorable or unfavorable outcomes such as which brands they prefer or do not like.
  • Analytics can predict the outcome of different price/offer scenarios and indicate the percentage probability of acceptance.
  • Optimal price/offer scenarios can be deduced from the “what if” scenario predictions by prioritizing the scenarios based upon current, high-priority business parameters (such as revenue, margin, or profitability).
  • AI can be used to automate the entire process as well as to continually recalibrate the analytic models used to predict and prescribe outcomes.

Short List of Requirements

The following is a short list of what you will need to get started on your digital personalization initiatives:

1. Easy access to well-curated internal and external customer data

What most non-data scientists don’t realize is that data sourcing and curation represent a majority of the work effort for any data science project. To be really clear, you must have access to the data you need in order to succeed.

2. Well-balanced personalization team

Like all transformation efforts, digital personalization initiatives need active support and direction from the CMO and from sales, marketing, merchandising, and customer service executives and senior managers in order to be implemented effectively and achieve lasting results.

Digital personalization efforts attempt to optimize customer interactions with data. Marketing and, where applicable, merchandising personnel play a very important role as domain experts who coach the data scientists and IT specialists to ensure that all current insights are captured in analytic models and not left behind as automation proceeds.

3. Skill-set requirements

The business-domain experts must have a thorough understanding of existing customer experience and campaign-management techniques. Usually, channel marketers (e.g., email and mobile app) are the most helpful to the data scientists because of their insights into the current customer journey.

Don’t forget that customers respond to the look and feel of your online and in-person customer experience interactions and the content you create to help them navigate your offerings. User-interface designers and content developers and writers are essential. They ensure that the analytic insights are translated effectively into messaging that works before it is sent across the appropriate channel for each digital persona.

Getting Started

Today, most companies are at the beginning of creating their digital customer experience. Here are a few ideas that can help you get started and also achieve some early success.

  • Evaluate how customer experience executives, managers, and thought leaders work today.
  • Focus on their decision-making process. What are their decision drivers? How well can they predict the outcomes of their decisions?
  • Capture all of the decision drivers for one key decision made, whether related to pricing, customer offers, promotions, media, etc., even if they appear to be in conflict.
  • Validate the driver assumptions and work with the end-to-end team to ensure that current practices consistently follow your organization’s own best practice. Compliance to current best practice will drive improvements.

  • Ask the data scientists to apply analytics and relevant external data to optimize outcome predictions. Use AI- driven, optimized decision recommendations to further improve your results.

Many companies will want to move quickly and will try to adopt a single, comprehensive enterprise solution that may have been designed as a best-practice platform for their industry. Before you follow this approach, use caution. New enterprise solutions usually have no way to capture the good insights and effective techniques you already have. You must adopt their vision of best practice.

Experience has shown that these solutions often fail to finish on time or deliver the expected benefits since they were built to solve pre-determined business problems; your business problems will be unique to you and your customers. The step-by-step process we recommend, focusing on improving one decision at a time, is less risky and brings benefits in months, not years. Caveat emptor!

Analytics2Go Analytic Microapps are designed to optimize your business decisions related to reducing customer churn, improving sales forecast accuracy, identifying operational problems, and diving into their root causes. Our Analytic Microapps include XSell (Cross-Selling), Sales Rewards, Short-Term Demand Prediction, Weekly Demand Prediction, Market Basket Analysis, Customer Segmentation, Next Purchase Prediction, Text Sentiment Analysis, Targeted Email Campaigns, and eCommerce Product Recommendation.