How To Know How Well You're Supporting Customers Via Self-Service

Customers have more choices today: more products to buy, more information to influence purchasing decisions, and more channels for customer service. What they don't have is more time. Here's some data from Forrester Research that shows how important it is to provide fast, easy service: 66 percent of consumers say that valuing their time is the most important thing a company can do to provide them with good customer service; 47 percent say that they are very likely to abandon their online purchase if they can't find a quick answer to their question.

Customers live on their screens and start customer service interactions online, via self-service. Tracking a package, getting a simple price estimate, returning an item, or paying a bill online are all examples of simple self-service tasks that we all do regularly. These interactions succeed because they are simple inquiries or processes that can be easily automated.

You can do much more with self service than just these simple examples. You can use customer, journey, and contextual data to make self-service more personal. You can offer self-service not only on a website out of band of where the customer is engaged, but within an application, in the flow of customers' work. For example, customers using Xero's accounting software find answers to their questions from within the product. Answers change depending on what the customer is trying to do within the product. Xero also proactively displays answers to most common questions based on prior user behavior.

But most complex self-service interactions don't do a great job at meeting a customers' need for service. They tend to place too much cognitive load on the customer, like navigating complicated decision trees or steps that lead you to the correct answer. Customers get frustrated with these poor experiences and ultimately pick up the phone.

How do you know how well you are doing at supporting your customers via self-service? Most companies start by measuring call deflection. This calculates the number of calls or live agent contacts that you can deflect by providing self-service answers or processes.

Companies also track customer satisfaction by trying to engage customers with simple thumbs-up/thumbs-down or 5-star content ratings. Others send out post-contact satisfaction surveys. The problem in both these approaches is that customers don't fill out post-contact surveys and the results that you get might be skewed. An alternate technique is to embed short micro-surveys in the flow of a self-service experience to get a pulse of customers' satisfaction and gather quick feedback.

You can also look at behavioral metrics to give you a clearer picture of the success. For example, the dwell time of customers on self-service content gives you an indication of how engaging or useful the content is. You can also track customers' next actions. Did they reach out to you over a live-agent channel within an hour or two of reading self-service content? This might indicate that the self-service experience was not effective at fully serving customer needs.

Self-service works. But getting it right is a journey. You need to constantly measure and improve your operations. And realize that as you get better at self-service, where your operations grow to handle more complex tasks, your contact center agents will increasingly address escalations and exceptions. This means that the nature of work within a contact center will change, and you will have to reskill and retrain your agents to support this new work.


Kate Leggett is a vice president and principal analyst at Forrester Research.