Sarah Hatter (00:00):
Maybe he's like a pickle with a hat on and it's like you can ask Mat our Pickle Mat, any questions that you have? Well, if Pickle Mat doesn't have all the answers for me
Mat Patterson (00:14):
Pickle Mat has no answers. Human. Mat is here to introduce you to this episode, which is an interview with my oldest support friend for, to be clear, she has asked me to remind you. She is a young and vibrant woman, Sarah Hatter. We talked about how we met, what it was like to do support right at the start of the SaaS support era and AI of course, and community, and a bit about pickles. I am always so energised by talking with Sarah. I think you will be too. It's the supportive. Let's go.
Sarah Hatter (00:46):
I'm Sarah and I don't ever think I've been on one of your podcasts before Mat Patterson, but you've been online several times for the past. Oh gosh, going on almost 15 years now. That's how long we've known each other.
Mat Patterson (00:58):
Actually, I figured out it's at least 17 years, but again...young, vibrant.
Sarah Hatter (01:05):
I help run Elevate CX, which is a community of customer experience professionals that runs the gamut from people who are managing support teams, people who are working for outsource teams around the globe, community managers, even some products. People are in there listening to all of our secrets and trying to make our lives better and our customer's lives better. It's really awesome to be in this space and see how much we've accomplished in customer experience, but also how much more work we have to do together. So thanks for having me.
Mat Patterson (01:37):
When I first met Sarah, I was the only support person at the email marketing application Campaign Monitor.
Sarah Hatter (01:44):
Yeah, you were a customer of mine. That's actually how we met. I was a support team of one working at what was then 37Signals is now, I think they're called Basecamp.
Mat Patterson (01:54):
I think they're back to 37Signals.
Sarah Hatter (01:55):
See, that's what happens when you go so far, you just time as a flat circle. Yeah, you were a customer of ours. I was doing support for a million users all by myself in a Gmail inbox. This was before Zendesk existed. This was before Intercom existed. We had no tools
Mat Patterson (02:09):
Help Scout. Might want to mention them
Sarah Hatter (02:10):
Help Scout. Help Scout wasn't around. Literally no one cared about us or our jobs except each other. That is what happened and so you emailed me as a customer one day. I wrote you back a response that most likely wasn't the best response I've ever written. I couldn't imagine. It was very hurried and you wrote me back and you were actually so kind and you said, I do your job across the world for a company that's kind of like yours because the kids will never remember this, but back in the day, cloud software services were new. People were downloading and installing and saving things on hard drives on their iBook. There was still so much to it and we were trying in vain to roll into that the lessons that we've learned from companies like the, or even Jeannie Bliss who had talked about customer experience and hospitality.
Jeanne Bliss: Hi everybody, this is Jeanne Bliss.
Mat Patterson (03:06):
Oh, hi Jeanne Bliss. Sorry for invoking you there. Jeanne has written a lot about customer service and she's probably most known for being the first to talk about chief customer offices. It's worth looking up
Sarah Hatter (03:18):
And it wasn't really ever going to work for us the same way, so I think that you and I being on the forefront of that, working for two pretty large names in the SaaS space very, very early on, got to be a part of dictating the path and creating those guardrails. We were the ones who said, we're going to do things differently. We're going to be kind, we're going to be friendly, we're going to be fast, and we're going to give as much resources as we can to people. We're not going to say things like, thank you for your feedback and we apologise for you inconvenience.
Mat Patterson (03:46):
Yeah, I was too scared to say it at the time, but I'll reveal it here. I know how much Sarah hates the word feedback, but I still sometimes use it. Sorry Sarah, please don't hate me. I look forward to your thoughts when this episode drops.
Sarah Hatter (04:00):
Here we are 20 years later trying to do the same stuff.
Mat Patterson (04:02):
I do remember looking, trying to find other resources, how do people do this and everything. I found it was like, well, if you're trying to set up your call centre in the Philippines, this is the model to follow. I'm like, this is not helpful to me. I don't want to do that. I don't want to go to a conference that's full of the CEO of telecommunications companies that are the ones talking about it
Sarah Hatter (04:24):
Or people who were working for companies like HP and Samsung who had 800 people under them answering phones and faxes and whatever it might be. It just wasn't in our realm. I didn't go to the Disney Institute for training, just I want to put that out there. We also remember that we're having to create new policies around things like refund requests, nonprofit discounts, user seat licences. We have to think about can someone cancel an account or gain admin control of an account without being the named person on that account because they got fired and we didn't think this through. There were a lot of rules and boundaries and like I said, guardrails that we were part of having to start creating from the ground up because it wasn't transferable, if you will, from the real world. So there wasn't a model and it's nuts to look back on that and think like, yeah, how did you do a background? What would I do? And the support person making $45,000 a year, year who has no technical background and gets to decide whether we let him in or not or what test he has to pass in order to get access to stuff.
Mat Patterson (05:37):
What was your background?
Sarah Hatter (05:39):
Oh gosh, my background was not in tech. My background was deep in customer experience, customer service. I never worked in food service, which I'm kind of bummed about. I wish I had. I feel like that's a real trial by fire for a lot of people In our line of work
Mat Patterson (05:54):
On that fit state of mind to respect what little customers you have, you shouldn't be in here.
Sarah Hatter (06:00):
My very first job was answering the phones at a call centre, and you might remember this if you're of a certain age group, if you're a true millennial, you remember the ads on TV that were like buy a dumb thing on TV and then it would go to a blue screen with a PO box and there was always one of three addresses or phone numbers or whatever. Well, I worked at one of those places when I was 15 years old taking calls from people who were buying sham wows on tv
Mat Patterson (06:29):
The Sham, wow sells for 1995, but you get one for the house, one for the car, two for the kitchen and bathroom,
Sarah Hatter (06:35):
All this shit. It was still like radio advertisements too, and we'd have a headset on and we'd take the call and someone would dictate their credit card to a 15-year-old on the other end of the line who would print it out on a dot, make tricks printer and then walk it over to the shipping department, plain text. That's where I come from, Mat.
Mat Patterson (06:57):
Now at this point we started talking about why small companies are often able to provide higher quality support than the large ones can, partly because of all the rules and the processes that you need to have as a large company with a huge customer base, and Sarah suggested that she thought AI might drive a real split between companies at different sizes.
Sarah Hatter (07:17):
Is the future of this then that those giant companies, the Disneys and the Hulus and the Samsungs and whom ever might be out there in the world listening, are they just going to default to we're just going to let the robots make all these decisions and then the smaller companies that are staying small intentionally staying more in control, empowering their consumers as well, they're going to become like the artisan customer support customer experience, the small batch customer experience, if you will.
Mat Patterson (07:47):
Yeah, I think we will definitely see human support as the luxury option that you get if you can afford to pay for a company which is willing to employ people to do it.
Sarah Hatter (07:59):
Agreed. I've always thought that premium support should be premium anyway. Maybe we replace that with the idea of customer success or onboarding or whatever, success managers, hand holders, people who are there on the spot for you, but it's worth it if you're using a complicated piece of software or an expensive piece of software, it's worth it because we know that charting customers is very expensive. Acquiring customers is more expensive than it ever has been. There's so much demand for perfection and people are so fickle with their attention span too that no one thinks twice about I'm just going to cancel this and try a new thing on the market. So as much human contact that you can offer even at the premium price if it's effective I think is still a supremely smart thing to do for your company. I don't know. I feel like sometimes we have these ideas, we talk about the startup on the quarter or the new guy in the block and we think of that there are nine people in a room building software or they're like this small group of people, but in six months time they're going to be 150 employees and 90% of those are going to be engineers.
Mat Patterson (09:13):
Hey, you shout out to the engineers who really care about customer experience. I know a few of you, you're all awesome,
Sarah Hatter (09:20):
So I don't know, Doug, your odds of getting back to that mentality is it's so slim, so how are we going to influence the VCs and how are we going to influence the 25-year-old CEOs coming out of Y Combinator to just say, look, we've got to focus on people still. You still have to do something that tells your customer, we see you. We know you have a braid, we know you have a beating heart. We do too. Let's try to work together.
Mat Patterson (09:45):
There is potential. I think there is one future timeline where all that artificial intelligence stuff is actually helpful in extending the amount of time that a small group of people can support customers like people. Because to your point, I think a lot of companies start off and they have good intentions and the CEO is answering emails or whatever, but then six months later, CEO's got zero time to be talking to customers and they've got hundreds of people and all these people are onboarded and there's no system in place, there's no training, there's no culture built to support all of that stuff. It just takes time and you and I think both probably started in companies which were quite small and didn't have millions of dollars to spend at the start and just had a long time to figure out what kind of company do we want to be and how do we want to treat our customers, and then it was so embedded by the time that more people came along that it sort of was able to carry on at least for a while. Some of the people that I employed at campaign monitor 15, 16 years ago that only recently left that stuff, it does carry on for a long time. Even as a company grows, it gets harder to do, but if you're a company that starts and then you grow and grow and grow immediately, what is the culture? It's just whatever comes from the top and that's probably board driven.
Sarah Hatter (10:59):
Yeah, it's board driven, it's VC driven. We forget now too that our lifecycles with VCs is sometimes 10 years versus two years and having B series A, series B, that kind of like now we are profitable, we don't need them anymore. We're seeing companies that are taking those on as a board that is their directors for the existence of the company, right. Process is so important when it comes to customer experience and everybody, like you said, being unified on that process and understanding that without customers, we don't have a company, we don't have a product, we don't have sustainability of that product or company. We have to have people who are willing to give us money for it. We could have the best idea in the world, we could have the best at GPT, the best ai, but if people aren't willing to pay money for it, then that doesn't get us anywhere and if we don't nurture the relationships that we have with those customers, the people who are giving us $49 a month for it, it's a constant churn and acquire situation which is not profitable either.
Mat Patterson (12:00):
No. And miserable to work in as well. So I want to go back briefly to AI and that idea of the human luckiness of having contact with a person. I think that is probably for a lot of people certainly listening to this, that's the best case scenario for what is going to happen is that yes, absolutely some jobs will be lost, but the interesting jobs and the more nuanced and the more complicated stuff is going to need people. We were both in London for your conference recently ish and I think we were in the same hotel were we were in
Sarah Hatter (12:35):
The same hotel. Yeah, we all were. I think a lot of us were right there down the street. Yeah,
Mat Patterson (12:39):
You walk in there, there is no concierge person. There's just a bunch of kiosks, right.
Sarah Hatter (12:43):
Actually was, I don't know what time you checked in, but there was a guy in the main room whose job it was to tell you that you couldn't check in until three. That's all he did. He couldn't check you in, he couldn't do anything when it was time. He showed you how to use the kiosk to check in.
Mat Patterson (12:58):
Yeah, that makes me think about, I think this is a very good example of where we're going to end up, which is there'll be a lot of companies which are, you get the kiosk version of service here and if you want to pay the lowest price, that's what you're going to get. And that makes sense because they're not having to pay for a nice friendly person who can check you in and shows you what a television is as if you don't know. And sometimes that's all you need as well. A lot of the time, yeah, I don't need you to be giving me the handheld service here. I've been on a plane for a thousand hours and I want to get in that room. That's literally all I need from you. Is that one key? To be fair, sometimes I need them to explain how the incomprehensible shower works.
(13:39):
Why do they never label the things properly so fancy, so annoying, but there are still going to be hotels where yes, I walk in and a person tells me, okay, you should go for a walk here and if you go down this street, it's much easier to get there and all of that kind of stuff that will still exist. And I think we need to personally as individuals, we all need to decide do I want to be okay, well my job really is how does this kiosk work and what happens when the kiosk stops working and how does it get escalated? There is going to be a lot of those jobs of like, well, yes, you are doing service design, but there's a lot less humanness in it and the humanness really is applied at the level of designing the interfaces and the processes and the systems or do you want to be, well now I'm working for the four star hotel and I'm on the front desk kind of job
Sarah Hatter (14:28):
And in the same conversation, who do you want your customer to be and what is their expectation? Because I don't know if you know about Yotel, it was like Yo sushi, they have their own hotel. It's exactly what it sounds like. It's like tiny cruise ship rooms. They're cheap. You do self check in, there's no food, there's no amenities. It's at four walls, a door and a bed. The idea of it is this is just simple, straightforward, whatever, so if you are a backpacker or a cheap traveller or someone who just needs the last room in the city, you go to hotel, you know what to expect. That's what you're expecting. But if you're someone who saw that this was a new cool hotel that you've never heard of and it has really slick modern branding and you are expecting a four star experience and you are expecting bell hops stick your bags up and you are expecting someone to show you how the TV works, you're going to be very upset with that level of service even though it's not meant to serve you.
(15:26):
So I think what ends up happening is we build products or we're building these companies, we're building experience trying to be slick and modern and cool and self-service and easy and we are still inheriting, accidentally perhaps through messaging customers who want a more full service experience, which just creates friction that we didn't really intend. And when you acquire the wrong customers, they are a support headache. We know that already. We know exactly the archetype of who they are. So then the conversation gets even bigger about we can't cast a wide net. We have a very specific ICP. We have a very specific successful customer that stays with us on that three to five year track that we get X amount of a CV from. Gosh, look at us all professional being in our C-suite using our C-Suite words.
Mat Patterson (16:19):
Okay, so an ICP is an ideal customer profile. Basically the people you want to buy from you an a CV annual contract value, how much are they paying you per year?
Sarah Hatter (16:32):
We want to acquire the right customers for the right product to fit our ethos as defined the process for our support as defined, but we've got to make it very clear what that is. And so there's so many times when someone is picking up a trend like AI or a chatbot and because it is not implemented specific to their customer that they're trying to acquire and keep, it is just a bad experience all the way around. Or maybe there's somebody who has, they work for a company that is chronically bloated and overprocessed and you do not have help documentation that they can copy, paste or link to and send. You have videos that have a five second entry splash and they're 90 seconds long because you got to watch the next one to get to the next point and there's no transcript. We have all these slick ideas, but if they're not implemented for the right customer, you're wasting your time on technology.
(17:30):
And AI I think is one of the worst offenders of this. AI has been around for years. As long as we have had modems and routers and connections, we've had the options to be using these things. We've peppered them into our daily lives all of a sudden in the last five or six years, we see this as a way to replace people to do these jobs of people, but if they're not trained, if they're not set up well, if you're not feeding them the right information and if you're not prepping and teaching your customers the usefulness of it, what they're there for and giving them options if needed, it's just going to again cause more of that friction. And we have this running joke that what is the number one thing people ask a chat bot and I'm sure you know what it is.
Mat Patterson (18:15):
Well, I think a lot of people, it's going to be can I speak to a person?
Sarah Hatter (18:18):
Remember that just screaming it into the telephone, look at us now typing it into the keyboard furiously, talk to an agent, talk to an agent, talk to an agent. We're never going to get away from that instinct that someone wants to just get to the source who can help me. And so we have to think about process long-term before we think about technology right now because technology right now is never going to replace the scale and quality that people can offer to customers, but as a support leader yourself, I mean you've had to train people, you've had to train ics, you've had to train people who just don't pick it up. They don't get it. It's almost like we have very parallel issues here between technology and human beings and I don't know which one's superior. Robocop will tell us. I guess
Mat Patterson (19:05):
If you go to check into a hotel and Robocop is there should have gone with a slightly higher brand.
Sarah Hatter (19:11):
Robocop is there to show you how the remote works. Oh my gosh.
Mat Patterson (19:15):
Check out as at 10:00 AM you have five seconds to comply.
Yes, I know it was ED 2 0 9 that said you have five seconds to comply. I'm not a fool, but we moved on to talk about customer expectations.
You need to know what sort of company you are and the customers need to know what sort of company you are because it's almost impossible to have a good experience if they come in expecting one thing and the thing that you give no matter how good it is, is not that thing. So when the sales team is out selling a type of service that they do not fund you to provide, that sucks for everybody. Or when the marketing of the website makes it look like they're going to get a kind of help that there is just not a team there to provide. No one in support wants to be giving bad service, but if there's five of you and there's a hundred tickets to get through and you've got a certain number of hours to do it, it's just not possible to do everything. And so the alignment between what it looks like on the website, it's going to be what the company internally is willing to pay for in terms of service, what the sales team is selling, what the emails, what the support form says they're going to get, all of that stuff, getting it aligned even if what it comes out as is what our service is cheap. It's quick. That's what you're going to get. That's what you're paying for. Sure. Great. And I understand it, it's a yotel of service. Absolutely.
Sarah Hatter (20:34):
You saying I understand that that's been made clear to me. I am the right customer for that right product. When you talk about people having expectations on silly slick marketing things, I think that's one of those problems we've had with the AI stuff is that people think that they can slap a bot on there or a little guy in the lower right hand corner, what is he? He's probably a bear bear with glasses on maybe or maybe he's like a fun pickle. Maybe he's like a pickle with a hat on and it's like you can ask Mat our pickle mat, any questions that you have. Well, if pickle mat doesn't have all the answers for me, this is where it comes back to this is a process, this is a practise. This is not a one and done thing. You don't slap an AI chat bot in the lower right hand corner and let it loose and hope it does its job.
(21:23):
I hope it deflects as they say. Right. Same thing with the help centre. You don't put it up to give the appearance of helpfulness. It's meant to be a living document, something that is updated constantly along with the rest of your product. I think we forget that the idea behind AI and the idea behind any solution that's been introduced into CX is to reduce friction and when all it does is become an impediment and increased friction or increased frustration by the time the customer gets to a person, they're pissed, they don't want to deal with it. That's not a new story. So we've got to find a balance and I think the balance is always going to come back to two things that I've been sort of ruminating on for the last 20 years. One that you talked about, smaller companies have more leverage to empower their support teams
(22:16):
You and I have been preaching this from the rooftops across the world and we are dying for people to listen to us saying you need to empower the people doing this job frontline. So we still have to get that message across oddly enough. And secondly, it's about using the technology and the right way to serve the customer, not to serve your board and not to serve the person who made the AI tool and not to serve trends, not to seem like you're on top of things, whatever. Again, we're going to cycle all the way back again from hot and flashy and cool to artisan solutions and that's just the endless cycle that we're going to be stuck in unless we try to make this a more I guess globally accepted process.
Mat Patterson (23:02):
Yes, there absolutely is. A cycle happens in business more generally. We do go through these trends, but the thing about customer service, customer experience is it is always just the same thing, isn't it? Nobody is coming to you to pay to get help, right? They're coming because they're trying to do something. They think that the product you're selling them is going to help. They hope it is. They throw some money at it to try and make it. There's usually a gap between that point and the point where they actually get the value and that's where customer service sits in the middle trying to bridge them from one side to the other side. And that's all we do really is do that in a thousand different ways and maybe some AI help will help get them across. Sometimes it's absolutely not going to help and the trick of support is always like what do we need to give this person and and in what format and at what cost to us? That's the balance of support.
Sarah Hatter (23:52):
That's the balance. And there's also like we grow these teams of ics who are like, okay, your talk at London was so brilliant. I'm going to invoke it right now.
Mat Patterson (24:01):
Oh, well that's very kind. You can hear a version of that talk in episode six of this podcast or on the Help Scouts YouTube channel link in the show notes
Sarah Hatter (24:11):
First day on the job training, here we are the new support team 2028. This idea that we have to just set the standard. No one wants to talk to us, no one wants to talk to you, okay, you are here to solve problems quickly, efficiently, and just accept the fact that people are at their last resort coming to you. Okay, so that's how we think about customer support. Now we can now trace this back through the process of all the things we could do to make that not happen. I always say the best support is no support. I found an answer. I got what I needed. I didn't have to engage anyone or anything, any sentient being did not have to come into this with me. I got the answer. We then we hope 15% of the rest of those customers who have a question and have a need have to then have a sentient being helped them with that.
(25:05):
So if we're going to go into the future putting AI in front of everything that AI has to know all the answers, get it all right, all the language, get all the typos, predict all the needs right there. It should not be a gatekeeper to try and like I said before, deflect or stop people or hope that they just get frustrated and forget to cancel and we charge them for the rest of their lives. When we get to the human being again, short, quick, easy, friendly, happy, we've repaired the relationship with the customer who's had to spend all this time trying to find the answer or whatever it is, we try to do whatever we can to invest in the services that we're giving the customer and all of these different ways that they get that service has to be as fast and quick as possible and efficient as possible.
(25:54):
We have the means to do it and I think what's going to end up happening is in two years time, we are either going to condition the world to just accept really shitty terrible service and this is just it because we've outsourced it all to ai, we've given up or companies are going to say just like Richard Branson did in the seventies, I can actually create a better experience for people in my little niche industry and I'm going to do it and I'm going to become the gold standard that then that next company, that kid in their garage who starts their company in 2028 is going to say like, we're not going to do AI because we're going to invest in people, we're going to invest in community, we're going to invest in education, and that's what's going to set us apart. So companies can choose to do that right now. They can choose to do that, some of them do.
Mat Patterson (26:43):
The reason that we don't see a lot of it I think is because if you have a board and they're trying to get back there X millions of dollars, everyone wants to see ai, they just want to see that. That's just a requirement. Even if you wanted to be a human first company now and you had investors, you cannot talk about ai. You just won't survive. We just have to get through that hype cycle.
Sarah Hatter (27:05):
Yeah, I think we're at the tail end of it. Here's a business proposal, Mat, you and I ready for this. We have enough context in the world. I think we can get this off the ground running. We start a company, doesn't matter what it's software, some kind of software pickles are involved somehow, right? I do want to use that Map the pickle. That sounds good. We go get money. We tell them check out our amazing AI and we show it to 'em and we do the demos and they're like, this is incredible. It's teaching me everything I need about this app. I'm going to be the most successful customer and I'm never going to cancel and you're going to make money off me for the rest of my life, and they give us more money to build more of it so that we can acquire more customers. The trick is there's no ai, it's actually people
Mat Patterson (27:46):
We're mechanical turking,
Sarah Hatter (27:47):
Yes, but ethically
Mat Patterson (27:49):
Ethical turking, okay,
Sarah Hatter (27:50):
Ethical.
Mat Patterson (27:51):
I mean don't Google that. That's probably a thing, but yes, essentially that is what's happening with a lot of AI already. It's just that yeah, they're paying those people nothing and it's miserable. One benefit of crappy AI support is people realising, oh, there's a lot more to customer service than we thought, and the positive outcome of this is one, realising that actually a lot of customers are, I mean a lot of companies are already providing terrible service and that mediocre AI service might be better than that. Great brings up the absolute bottom to somewhere mediocre, but it also will show us that oh, good service is a lot more than what this thing can do because it's not just providing an answer to a question. There is so much more nuance involved in delivering good service and figuring out who this person is and what they're trying to do, which is maybe not the thing that they're actually asking you for and helping them. All of that stuff is just going to require people and so companies picking out, yeah, sure, we use ai but it's behind the scenes and it's like an assistive technology to our agents to provide that service. Absolutely.
Sarah Hatter (28:58):
It's part of the empowerment of the people that are doing the job and unfortunately really, really, really good experiences when people invest in those experiences. There's a huge upfront cost, but there's also a huge upfront cost to churning out customers and users because the systems are so terrible or your cancellation policy requires me to call you, which I think is illegal now in the United States because it's such a bad policy or I can't get answers or this doesn't make sense and I don't know what to do. I don't know why. In the world of business customer experience expenses, the cost of people working for you, building the processes or whatever it is, is more scary than 700 grand a year for 10 developers to hack away on all the same problem. We don't ever think about that cost or that cost doesn't scare us as much. It doesn't put us off. We think that there's still something disposable and convertible and transient about customer service and about people who do that job. They're still entry level or they're still unskilled, which is just simply not true.
Mat Patterson (30:13):
I feel like we should form a union with all of the other people whose jobs are considered soft skills because it's exactly the same problem that they all face, which is they're underpaid and undervalued because people think that they could do that job because they've never actually done that job. And that's why companies that are founded by people who have actually had some service backgrounds always have better customer service because they value it and they invest into it.
Sarah Hatter (30:37):
And not only that, but they also, they have happier employees. The other side of this too is the longevity of support employees. Support employees are an incredibly valuable resource to have as part of the ecosystem of a company, and when you get someone who is happy to do that job and to perform well for your customers and has empowerment to do that and make changes or implement process, they will stick around. Employees are also very expensive to replace just like customers.
Mat Patterson (31:05):
You can't go straight to happy customers without going via happy customer facing stuff. It is just not possible and you can't get rid of all the staff and just have AI because it's not that good at doing that. Even in the best case, even the best case, you still want someone in the company who is like, this is what good service looks like for our customers and at our company and doing the very hard work of making the AI approximate that that's also going to be a complicated job and you're going to need the person with the skills.
Sarah Hatter (31:36):
But I do think that one of the biggest mistakes people need to stop doing right now is gatekeeping through ai. I think, like I said before, it just creates way more friction than it's intended. People are too smart when they're putting this stuff together to think like an unintended consequence is it creates friction. They know it creates friction. They know that if someone is typing in, cancel, cancel, cancel, and they don't find an answer for it, they can't figure out how to do it, it's not something that's accessible to them and chat says, maybe we'll get back to you in three days worth an answer. That person's going to likely forget about it or move on or whatever until the next six months when it triggers their memory, oh, I should go get back into I tried to cancel and I didn't do it or whatever it is. There's human behaviour and psychology is also behind a lot of bad decisions. We can't chalk it up to ignorance all the time.
Mat Patterson (32:24):
There are definitely companies who use all sorts of technology as an intentional, desirable friction to prevent expensive interactions with people. We've all experienced service from those companies and that's miserable.
Sarah Hatter (32:38):
Once when I worked for a company, I won't say the name, I would get a lot of emails from people saying, we forgot to cancel this. Can you refund me the last six months? You can see that we've never used it and these are businesses saying this. It wasn't like an individual, well, I guess they're still humans of their businesses, but whatever. I remember bringing it up to my boss at the time saying Maybe we don't cancel 'em. Maybe we pause 'em so they don't lose their work and their stuff, but it's kind of a goodwill gesture, don't you think? And maybe even if we sent them an email saying, Hey, you haven't used your account for a while, so we're going to pause it, it's going to trigger them like, oh yeah, that product, I should use that again. I should get back into it.
(33:19):
And his determination was doing the math. They made so much money off of those accounts that had forgotten to cancel that it wasn't beneficial to their bottom line to do that, to implement that kind of goodwill structure to be like the better company or to gain reputation points because it just made them a lot of money and that was 20 years ago. So imagine the expansion on that kind of thinking now it's nuts, but we have to not be naive in thinking that some of this stuff is unintentional when it comes to that, the poor customer experiences that are out there and not just intentional in the way that we don't want to pay agents. I hate that word. We don't want to pay employees to do support. We don't want to invest in education, but sometimes an apathetic customer who forgets that they're paying for us, paying for something that doesn't care is more valuable to us than all of those other things. Or sometimes I think that we've gotten apathetic as a culture just saying we understand that where 10, 20 years ago the baggage was having to call Verizon or Comcast or Utility. It's now ruled up into the Netflixs and the Disney pluses and whichever other softwares is embedded in our lives, they're also going to be a poor experiences, so why deal with it? I don't know. I have optimism about the future of customer experience. It doesn't sound like it.
Mat Patterson (34:42):
No, I also, I think there is lots of potential to do interesting things. I think if you're a person who's starting today, your first job in customer support is today, there are all these new tools, there are ways that we can deliver better service at lower costs than ever before. I think there's huge opportunities for the right people coming through. We definitely don't want to be the old curmudgeon who are like, ah, I'm not going to use these AI tools because I could do it myself. That's just not going to be the way of the future. There's going to be new and interesting ways to deliver really great service that use all of these new technologies that didn't exist,
Sarah Hatter (35:18):
But as you said before, use them as supplemental to the work that you're doing not as a replacement model
Mat Patterson (35:26):
For the agents doing the work. There's a whole bunch of stuff that you answer in support, which is ultimately all I'm doing is being a slow search engine to find the knowledge base article that is already there and so you having to wait three hours for me to go, yep, it's this article here it is. If the search engine can be better because it now can understand your terrible question asking and match you up with the knowledge base article, fantastic. That's great use of technology. The AI can be on the front end. You want to have a short conversation in which it finds that for you and gives it to you in a second. That's such a win, I think huge potential for lots of that stuff to be built into great service, but what it still needs is somebody who's saying, this is important to us and we are going to put in the time to craft the prompts or to rewrite our knowledge base in a way that works better with the way this thing can find stuff so that it can provide that great experience, so much work to be done there.
(36:19):
And we want the people doing it to be like the customer service people and not the tech AI people who don't care in the same way and who aren't going to have the same priorities of how this thing works
Sarah Hatter (36:30):
Or presume to think they know how people will behave. And I think that's one of the big curses about siloing out your tech teams versus, I call them people teams, which is like the customer support, customer success, people who actually do the work one-on-one with customers. You can't leave them out of that conversation when you're building based on presumption because they actually know behaviour. I also think that to be a really, really great successful person in customer experience, as you probably know, requires an interest in human behaviour in psychology, requires an interest in where are we going and where have we been? What have people been through before? What baggage are they bringing into the room? Having an understanding about the humanity of things, it doesn't just make you a better human, but you understand humans more so you can better presume how they're going to behave and better presume what their needs are and implement stuff to actually solve those problems for them before they have it
Mat Patterson (37:29):
And make a judgement . So many judgement calls and support
Mat Patterson (37:33):
What is the appropriate thing, what's the best thing for this person. Sometimes people asking for a refund, the best thing is a refund and sometimes it's actually frustrated because you can't do what you want to do and actually I can just help you do that. You won't need a refund. We don't need to pay it. Like an AI is not going to make those judgements.
Sarah Hatter (37:49):
It's not going to make the judgement for you. And it also takes us back to the conversations we were having in 2008, 2009 about things like feature requests and how people were saying shut 'em down. Don't even answer 'em. Just saying we don't answer feature requests. Like no feature requests are telling you somebody is trying to use your product and give you money for it and you can accommodate them. Forcing people into your version of how to best do work or how to best use a product is only going to create more of that friction, and that's something that has influenced how we think about serving people and serving them customer education on things. So I don't know. I think very deeply about this because where we are as a culture, you can ignore crazy stuffs happening and it's influencing people and it's influencing people what's going on inside of them.
(38:41):
And you and I also know is we're protective of our teams because we know that it's just a barrage all day long of angry people questioning people, people with needs having to serve those people one after another, after another another. And that wears on you. It wears on the customer, but it wears on the person doing the job. So wouldn't it be better if we empower those agents? We got to come up with a better word for what to call people who do this job. We empower them to make the right decision on behalf of the customer. Our policies are empowering that employee, but we also empower our customers to find out this is the right product for you. Here is your needs are met. We are servicing you all of this education in various formats. If you're a visual learner, here's gifts. If you like a video walkthrough, here are those right here.
(39:30):
You need help. Well, you can start with our help section that we update weekly or you can chat with the bot if you don't want to talk to a human or you can email a human. You can use this conditional form that's going to help you categorise so that it goes to the right person. That shows a thoughtfulness about everybody's encounter, the person who has to answer that email or whatever it is, and the person who has to ask it. It gives so many options Instead of just funnelling everything into this is what I believe is going to be the best experience and the cheapest experience and the easiest experience and just instal and go, that's never going to work long-term. It's not a sustainable solution for anyone. Eventually you're going to have to circle back and hire humans to redo this process
Mat Patterson (40:17):
And you won't have any of that mental, that kind of internal history of how we talk to our customers and what's happened in the past. All of that stuff will be gone because you fired them all and tried to replace them. Yeah, I agree. The best customer experiences and the worst customer experiences in the next few years are all going to involve AI and people in some form, and we just need to be on the right side of it. We want our companies to be the ones that are providing better experiences because of technology. Not worse. Hello, it's non pickle. Mat, here again, originally this was going to just be one episode, but it was getting really long and this felt like a great place to stop. So what I'm going to do is keep the rest of our conversation for the next episode. Part two, Sarah and I really getting into talking about the support community and that has played such a big part in my working life and I'm really so appreciative of what Sarah did to help it grow, so I'll put that second part out in the next couple of weeks.
(41:14):
Please stay subscribed for that one, and if you would like to help this podcast reach more people, I would love for you to leave a comment or a rating on Spotify or on Apple Music or you listen to it. It is the easiest way to help this get in front of other people, and I'm going to try to read out some comments in a future episode...depending on what you say.
