Intelligent Scoring: The Modern Approach to Evaluating NPS and CSAT

By: Ryan Murphy, Courtney Jones

July 22, 2020

Tags:
Intelligent Scoring
Clarabridge Analytics
Contact Center
Artificial Intelligence

This year’s unexpected pandemic has challenged companies to cut costs. Expensive customer satisfaction surveys are being scrutinized as one cost-cutting measure. Asking about “willingness to recommend” in a time of such tremendous uncertainty is out-of-touch with the current consumer mood and doesn’t provide much actionable insight. At the same time, it’s never been more critical to keep a pulse on shifting customer concerns. That’s why companies are beginning to explore alternatives to traditional post-interaction surveys.

Clarabridge’s Intelligent Scoring is a breakthrough in interaction analytics that can automatically evaluate and score every conversation your agents have with customers, whether that’s over the phone, via email, or on live chat and social channels. Clarabridge analyzes the actual voice of the customer – the words they use in conversations – to automatically evaluate attributes such as effort, loyalty (NPS), and satisfaction (CSAT, OSAT). Intelligent Scoring distills multiple aspects of the customer experience into one easy-to-compare score. Think of it like a credit score, which consolidates many disparate factors into a single data point. It’s all automatic, so there’s no latency and no survey fatigue!

With Intelligent Scoring, you define your own set of evaluation criteria and weight the relative importance of each variable. Clarabridge uses Natural Language Understanding to automatically analyze and score every conversation. Then (unlike a survey), you can use Clarabridge CX Analytics to drill into root cause and outcome drivers, assign tasks to take action, and track continuous improvement. You get immediate and actionable insight.

 

Assess Dual Sides of the Conversation with Survey-less CX Scoring

Surveys often solicit feedback solely from the customer’s perspective (for example: are you willing to recommend?). With Clarabridge, you can assess each interaction more holistically. For example, you can base an agent’s score on the presence or absence of desired behaviors such as empathy, problem resolution, or script compliance, while also scoring the customer on the presence or absence of CX indicators (runarounds, complaints, expressed effort).

This approach surfaces instances where proactive outreach may be needed, or the company’s protocol may need to be revisited. Additionally, this bilateral scoring approach mitigates situations where an agent may have a low NPS associated with a call because the customer was upset about other aspects of their experience with the company, rather than due to any fault of the agent.

 

…………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………

 

Agent performance compared to Customer’s experience

 

Agent
Client

80 of 100, based on exhibiting desired behaviors and core competencies

42 of 100, based on mixture of poor CX triggers and CX wins

 

Client is friendly towards the agent, but obviously upset at the problem.

 

 

………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………….

 

This customer’s situation is increasingly familiar. She has been unable to maintain the payments on her car and her car was repossessed. She expressed a lot of negativity about company policies and was clearly not satisfied. This interaction scored low for “customer satisfaction”. The customer service agent, however, scored well for quality of service, nailing many of the KPI behavior categories identified as important.

Automated scoring of both sides of the conversation achieved two goals: rating the agent’s performance for QA purposes and rating the customer experience for CSAT purposes (while also revealing opportunities for proactive outreach or policy adjustments) — without using surveys. Intelligent Scoring scores 100% of interactions, therefore giving a more holistic view of CX.

Wrap-up

The utility of one-dimensional customer experience (CX) indicators like NPS or OSAT is dwindling. Automated scoring and analysis of interactions using Clarabridge is a more modern approach to understanding customer experience. Ditch the surveys and start taking a more automated, data-driven approach to listening and acting on voice of the customer data.

 


 

Ryan Murphy is a product manager at Clarabridge, where her not-so-inner linguistics nerd can run free. She is a northern Virginia native, but crossed the Potomac to attend Georgetown University, where she earned her B.A. and a Regent’s Award for her studies in linguistics. She concurrently studied philosophy and cognitive science, which heavily informed her post-grad work as a federal consultant with a particular focus on AI ethics. When she’s not immersed in Intelligent Scoring use cases or brainstorming novel CX Studio features with her team, she can be found trying out new recipes in her much-too-small-for-experimenting kitchen, exploring the myriad neighborhoods of D.C. on her bike, or training for marathons and triathlons.

 

Courtney Jones is a business consultant at Clarabridge, and for the past year, she has diligently supported the development of Intelligent Scoring from beta through its release. Originally from Charleston SC, Courtney received a BS degree from Clemson University, as well as two master’s degrees from the University of Virginia. Her further education and shift out of a career in Finance led her to Washington DC – and subsequently, to Clarabridge – where she has worked to find companies data-driven business insights across several verticals such as healthcare, finance, retail, and insurance, as well as to steer this innovative beta project to a successful release. Outside of work, you will likely find Courtney tailgating and cheering on Clemson football (she is from the south after all), traveling (she’s been to 26 countries so far), and training for marathons or Spartan races.