Self-service is often posited as either a transparent means of cutting costs, or a way to dramatically improve customers’ lives. This dichotomy should not exist, though—companies should be achieving both. And while a few industry leaders are, many are still struggling to make self-service more than just a quick and easy way to cut costs.
Where Self-Service Falls Short
Oftentimes, self-service is a base way for companies to avoid hiring real people for menial jobs. Instead of having 200 employees there to buy a plane ticket for you, you can now see flights online and buy tickets yourself. Instead of waiting in a massive line at the grocery store to buy one bottle of sparkling water, you can now use the self-checkout machines. The premise of self-service is that through new technologies, you can eliminate much of the hassles and slowness of person-to-person interactions.
The problem is, in most people’s day-to-day lives, self-service only works about 50 percent of the time. You accidentally hit “pay by card” on the checkout scanner, and then realize you wanted to pay with cash. There’s no way to go back, and the self checkout manager is busy helping another customer. What should have been a painless and rapid task, now feels even more laborious than waiting in a normal line, because your expectations have been let down.
The thing is, expectations are relative: they are formed based on easily manipulated sets of circumstances. If you are expecting to wait in line at the grocery store, the fact that there is a line will not anger you. If you are expecting to go through self-service in under two minutes, the extra minute and a half that it will take to work with the machine will anger you. Even if ultimately you save time by going through the self-checkout process, you will feel more disappointed than if you had waited in line. When consumers are using highly technical machines to perform tasks, their expectations change. You move from expecting the inherent slowness and tedium of interacting with a person, to expecting the lightning-fast response of the mobile phone. That’s why, when offering self-service, it’s crucial that you prepare the customer for the new technology they’ll be interacting with. This means offering comprehensive onboarding and setting up ways to change people’s habits.
Customer service is not just about exceeding expectations; it’s about setting up expectations.
Rory Sutherland talks about this in the context of advertising. He says, “Most problems are actually problems of perception.” By that logic, solving a customer’s problem is just as much about actually fixing it, as it is about making the customer perceive their problem not to have been a problem.
Self-service has the potential to do this. When customers can solve a “problem” themselves, it ceases to be a problem, and is instead simply a
task they need to complete. For instance, say that you have to cancel an Uber ride because the driver called and said he doesn’t want to turn around and go to your house. Upon cancelling, you are charged a $2 fee— not a ton of money, but still, it seems unfair. Before Uber had in-app support, this would have been a problem. Now, however, directly under the ride summary, you have an option for being incorrectly charged a cancellation fee. The problem evaporates, because Uber has preempted it with self-service. “The interface fundamentally determines the behavior,” says Sutherland.
Self-service vs. Self-support
The issue with self-service arises when it is confused with self-support. Self-service allows people, using technology, to streamline their experience of the product. Self-support, however, allows people to solve a problem after it has arisen. In other words, self-service is preemptive, self-support is reactive.
Good customer support technologies make self-support as easy as not doing self-support. Reading a knowledge base article or watching a video should be as easy as initiating a chat conversation with a customer service agent. The flow from problem to solution must be easier and more intuitive than just contacting an agent would be. But when a customer is already at the point that they need help, there has to be a very, very accessible option to contact a person. To use Uber’s support as an example again: if you have an issue with a ride, you go into one of their suggested knowledge base articles. If you cannot find your answer there, you can type your problem at the bottom of the knowledge base article, and hit submit. In my experience, you typically get a response within two minutes.
Uber has successfully married self-service with self-support. Everything in the app is done by the user, including problem resolution. However, when a problem cannot be resolved by the user, there is a simple and intuitive way to interact with a support agent. In most cases, for Uber, self-service and self-support are indistinguishable. But when good old fashioned (or as old fashioned as you can call in-app chat) human help is needed, you can easily access it within the app.
The Potential for Self-service is Great
Self-service does save money. And oftentimes, self-service is what consumers want. But technology is not infallible, and where I see many companies failing is through the idea that self-service is inherently “good,” without examining the means by which they are providing self-service.
For instance, Sutter Health offers an online portal for booking appointments, paying bills, etc. In theory, this is awesome. You can book an appointment online while you’re at work; you can eliminate waste via paper bills; you can view your health records in one repository. In practice, though, the interface is cluttered, and there are gaps in user flow. For instance, it’s difficult to communicate with the nurse/doctor on the appointment about which you have a question. In Uber, you go into the ride you had an issue with, and from there have myriad options from which to choose. In Sutter Health’s portal, you have separate tabs for billing, communication, and appointments.
Self-service is not inherently good. It is only good when it is better than not doing self-service. And even then, it must be as good as the expectation it has set up. As in the grocery store example, expectations change with medium, and you must not just make self-service exceed non self-service expectations, but also have it exceed self-service expectations. Where many companies are failing is by employing self-service without using sophisticated technology, which is even more frustrating to the customer than not having self-service at all.
Self-service has the potential to be incredible—to save customers time and providers money. But that potential is only being met by a few select leaders, who are continually increasing expectations for self-service across other companies and industries. The challenge I pose to companies is not to provide self-service; it’s to provide the level of self-service and self-support that customers crave.