More than any other department in a business, customer service is subject to a slew of competing priorities and pressures. Control costs and making customers happy. Uphold policy while being a customer advocate. Try to upsell but stick to your average handle time. The list goes on and on. Unfortunately, attempts to shore up performance on one end of the scale can often lead to disaster in another area. Here are a few of the most common examples of where things can go horribly wrong.
Treating process as the outcome. If you always do A, then B, then C, the customer will be happy, right? Maybe sometimes, but people are unique, situations are different and emotions often come into play. Research consistently shows that an emotional tie is the key to a strong relationship. Trying to script out every aspect of the relationship is about as effective as going on a first date and spending the whole night reading pre-written sentences off of a stack of notecards. It’s okay to have a toolbox of processes and approaches, but representatives should be given some level of empowerment to adapt to the specifics of the situation.
Cramming too much into the service interaction. There is nothing wrong with upselling if it is appropriate given the particular circumstances of the interaction. If a customer has a clear need that could be met by the company and appears to have a few extra minutes, offering to describe an additional service can be a true win-win opportunity. But it has to be done in the right context. Offering a balance transfer to someone whose interest rate was just raised, talking about a new account to someone who just had a bad experience with an existing account, or trying to describe a new product to someone who is clearly in a hurry is never a good idea. Whatever gets incented will win, which can often mean the service part of the interaction (the reason for the call) may get glossed over so the rep can offer three or four different add-on products, if that’s how they get extra compensation.
Automating the customer experience. People are hopefully smart enough to know when something is real, and when it is not. My son and I recently hit a fast-food drive-through. A pleasant female voice asked if we wanted to try the new special, and when I said no thanks, a gruff teenage male voice came on the speaker and continued to take the rest of the order, making it pretty obvious the initial welcome speech was a recording. My son started laughing so hard that the cashier was clearly offended. There is a time and a place for automation, but companies need to remember that machines don’t really smile. Only people do.
Ignoring the need to defend the brand. Service representatives need to act as customer advocates, but they also sometimes need to say no. A critical part of their job is to solve customer problems, but also to defend the brand. Unfortunately, many performance management systems put representatives in the no-win situation of having to choose between a good score on a survey or upholding policy. Instead, representatives need to treat their customers with dignity, regardless of if they say yes or no.
Becoming a slave to a small number surveys. Companies don’t make their money from survey respondents.They make their money from customers. Once the survey enters into the dialog, sometimes that all changes. I was very loyal to a particular coffee shop until I heard an employee tell a customer, “You have been selected to take a survey. Make sure you give us great ratings or we all get in trouble.” Since then I have started going somewhere else. I would have given perfect ratings—until that moment.
If used properly, surveys can be a very effective tool for looking at patterns or trends, and getting a representative view of how the service experience can be improved. However, over-reacting to a small number of surveys (or even just one survey) leads to attempts at manipulation, a constant game of whack-a-mole, and an overall degradation of service. Beyond that, small sample sizes can be dangerous. A score of 50 percent on a sample size of 10 has an error range of plus or minus 31 percent! In other words, the “real” score is anywhere between 81 percent and 19 percent. A manager cannot say with certainty where in that range the true score lies. Incentives would be based on random sampling variation, not on performance.
The key to avoiding these mistakes is to keep common sense in the equation and not try to over-react or over-legislate the service interaction. If you have hired smart, dedicated and friendly people, create the right environment and let them thrive. Your customers will notice.
Tom Rieger is the author of Breaking the Fear Barrier, and is president/CEO of National Business Innovations LLC, a member of the NSI/NBI family of companies.