I’m not sure there will be a difference either way. Customer service jobs spent decades trying to train humans to act as much as robots as possible. Of course replacing them with a shitty bot seemed to make sense, they were already pretending to be shitty bots.
Any “quality” in customer support comes from individuals circumventing company rules to provide actual support to the customer. AI can’t do that.
You’re right, absolutely spot on, about several things but ONE IN PARTICULAR is this:
A human being helping a customer is, quite literally, an act of circumvention. Customer service EXISTS, SPECIFICALLY, for scenarios that require exceptions and skilled, knowledgeable internal maneuvering within, between, and around the cold mechanisms of machinery and policy. We tend to think of, say, purchasing items at a store as standard operation, for instance. But really, from the perspective of the business, its objective is to RESTRICT access to goods and services. The cashier manages exceptions to this goal. If the company has its way, it would take your money while relinquishing NOTHING.
I do not agree with the idea that humans are being trained to act like robots. Any company with a customer service department is likely tracking the root causes of their customers’ issues. With enough data, they can identify the most common problems and their solutions. If the goal is to resolve a customer’s issue as quickly as possible (which seems like a reasonable assumption), it makes sense to guide the customer through the most common solutions first, as that will likely solve the problem.
If someone works in customer service and repeats the same script daily, it’s understandable that they may come across as robotic due to sheer boredom. A skilled customer service representative can recognize when to use the script and when to deviate. However, if a company fails to hire the right people and does not offer a fair salary, those best suited for the role are unlikely to take the job.
it makes sense to guide the customer through the most common solutions first, as that will likely solve the problem.
And this why you have to suffer through those lengthy recordings that tell you about a bunch of shit that generally doesn’t apply to your situation before you can even use the menu, much less talk to a person. I am disabled, I have had to be on the phone with the Social Security Administration, Medicare, my insurance company, and various state benefits agencies probably 15-20 times a year for the past ~14 years, and I can count on one hand the number of times those ‘common solutions’ were even remotely applicable. I don’t even need fingers to count the number of times they have actually contained the solution to my issue, because it has literally never happened.
Once you get to a person who can make an assessment about what’s going on it makes sense for them to cover a few basics (I used to do tech support, I know how much time a simple, ‘Are you sure it’s plugged in?’ can save), but replacing customer service with AI means you’re pretty much stuck in those recordings for your entire call. Now to be fair this can be done better than most places do it. I shop on amazon a fair bit (can’t drive so I order most things online) and when I have issues I honestly prefer dealing with the livechat AI than calling because it’s a much faster and smoother experience and they can quickly bump you over to an actual agent when there’s a weird thing going on that’s outside of its scope. But most companies don’t have Amazon’s customer service budget to do shit like that well, so usually what I get is ‘If you’re calling about XYZ, hang up and dial this number. Did you know that if your birthday is on an odd-numbered day blahblah-ad-blahblah? If the crescent moon is waning and the distant hills are draining and the watchful eye is straining…’ etc.
I’m not sure there will be a difference either way. Customer service jobs spent decades trying to train humans to act as much as robots as possible. Of course replacing them with a shitty bot seemed to make sense, they were already pretending to be shitty bots.
Any “quality” in customer support comes from individuals circumventing company rules to provide actual support to the customer. AI can’t do that.
You’re right, absolutely spot on, about several things but ONE IN PARTICULAR is this:
A human being helping a customer is, quite literally, an act of circumvention. Customer service EXISTS, SPECIFICALLY, for scenarios that require exceptions and skilled, knowledgeable internal maneuvering within, between, and around the cold mechanisms of machinery and policy. We tend to think of, say, purchasing items at a store as standard operation, for instance. But really, from the perspective of the business, its objective is to RESTRICT access to goods and services. The cashier manages exceptions to this goal. If the company has its way, it would take your money while relinquishing NOTHING.
I do not agree with the idea that humans are being trained to act like robots. Any company with a customer service department is likely tracking the root causes of their customers’ issues. With enough data, they can identify the most common problems and their solutions. If the goal is to resolve a customer’s issue as quickly as possible (which seems like a reasonable assumption), it makes sense to guide the customer through the most common solutions first, as that will likely solve the problem.
If someone works in customer service and repeats the same script daily, it’s understandable that they may come across as robotic due to sheer boredom. A skilled customer service representative can recognize when to use the script and when to deviate. However, if a company fails to hire the right people and does not offer a fair salary, those best suited for the role are unlikely to take the job.
And this why you have to suffer through those lengthy recordings that tell you about a bunch of shit that generally doesn’t apply to your situation before you can even use the menu, much less talk to a person. I am disabled, I have had to be on the phone with the Social Security Administration, Medicare, my insurance company, and various state benefits agencies probably 15-20 times a year for the past ~14 years, and I can count on one hand the number of times those ‘common solutions’ were even remotely applicable. I don’t even need fingers to count the number of times they have actually contained the solution to my issue, because it has literally never happened.
Once you get to a person who can make an assessment about what’s going on it makes sense for them to cover a few basics (I used to do tech support, I know how much time a simple, ‘Are you sure it’s plugged in?’ can save), but replacing customer service with AI means you’re pretty much stuck in those recordings for your entire call. Now to be fair this can be done better than most places do it. I shop on amazon a fair bit (can’t drive so I order most things online) and when I have issues I honestly prefer dealing with the livechat AI than calling because it’s a much faster and smoother experience and they can quickly bump you over to an actual agent when there’s a weird thing going on that’s outside of its scope. But most companies don’t have Amazon’s customer service budget to do shit like that well, so usually what I get is ‘If you’re calling about XYZ, hang up and dial this number. Did you know that if your birthday is on an odd-numbered day blahblah-ad-blahblah? If the crescent moon is waning and the distant hills are draining and the watchful eye is straining…’ etc.