Customer research
is one of the most vital
things you can do
for success

Andrea Palten
Hello, today we have Dave Seaton. He’s currently the Vice President of Service Level Transformation at nThrive. Thank you so much for being here. So, Dave, that’s a long title, Vice President of Service Level Transformation. Can you tell us exactly what you actually do for nThrive?

Dave Seaton
Absolutely. So, in a nutshell, I transform customer experiences largely with our service and support organization, but there’s also a lot that goes into that from running a voice of the customer program, to doing Lean Six Sigma process improvement, all around, being able to deliver exceptional customer support experiences at a great cost.

Andrea Palten
Awesome, and then what does nThrive do?

Dave Seaton
nThrive provides revenue cycle software to hospitals and large health systems. So, we’re not on the clinical side, we’re actually on the financial side. Everything from registering patients and checking their insurance to making sure that the right codes are on the bill, to getting the bill to the insurance company, to following up with collections and denials processing. So, everything that a health system needs to make sure that I get paid on time and accurately by the insurance companies.

Andrea Palten
Okay, which everybody wants.

Dave Seaton
Absolutely.

Andrea Palten
So, what kind of titles levels of people are calling in? Who is your client, the actual person?

Dave Seaton
Yeah. So, where we’re in the B2B space, we’re selling software as a service to these health systems. So, the people that call in, that access support are the end-users, and depending on the product, it could be a biller that’s getting claims out the door. It could be the VP of the revenue cycle that is accessing our analytics platform, trying to do their monthly reporting. Usually the buyers, the decision-makers are health system CFOs, VPs of the revenue cycle. But depending on what they’re buying, they may not actually log into the product themselves. So, this concept of a customer is really multifaceted as you think about the end-user versus that person’s manager versus the department head and then ultimately the CFO. They’re all our customers, but they have very different experiences along the way.

Andrea Palten
Got it. So, that makes it a little bit harder. You have all these different types of people calling in. So, let me ask you, what are you doing that promotes great customer service at your organization?

Dave Seaton
So within the last 18 months, the first thing we did as I moved into this position and establish my team, we did a very intensive customer research project, and it’s kind of the first law of customer service to know the customer. So, we went out and we combined qualitative and quantitative research to understand what our customers value in a service experience and we learned several things. One of the things we learned is that our intuition was wrong. We were very, very focused on improving handle time and turnaround time and that was important, but it wasn’t as important as some other things. So, we identified five service elements that were all slices of the pie, different size slices that were all important to what are our customers’ value in a service experience. After we knew that, then we knew where to focus with our targeted initiatives, with our Lean Six Sigma process improvement, with our experience engineering initiatives so that we could design those customer service experiences that are going to resonate with our customers.

Andrea Palten
Oh, wow and if you’re allowed to share, if you’re not, don’t share. But if you’re allowed to share what was out of the five pieces of the pie, what was the most surprising?

Dave Seaton
Yeah, I probably shouldn’t share that in case we’ve got some competitors listening.

Andrea Palten
Alright, no insider information. Got it. That makes sense.

Dave Seaton
But I can share generally that today’s customers are seeking customer service in very different ways than we’ve thought about it traditionally. The telephone is not the number one way that customers are seeking customer service. In fact, a lot of customers don’t want to talk to you at all. So, especially if you begin and look at the B2C world, one of the top channels for customers to get service is Google, another is YouTube. So, if your brand’s not engaging in the place that the customers value, you’re going to miss those opportunities to design and deliver the experiences that are going to keep your customers around.

Andrea Palten
Oh yeah.

Dave Seaton
Yeah. So, we had some similar surprise learnings. Today we’re very, very focused on reducing customer effort. So, when customers do have to contact us, how do we make that as easy and hassle-free as possible? So, some of our initiatives are around:
The process improvement. Let’s take as much waste out of the process as we can.
How do we design those digital experiences on our portal that are just super easy and engaging. Then when we do have humans interacting with customers, how do we engineer those experiences so that customers walk away saying, “Man, that was easy. It was really easy to work with nThrive.”

Andrea Palten
And that’s such an important thing. People want it easy, especially now where there are so many companies that didn’t implement or aren’t implementing all the stuff that you guys have. You’re on hold for four hours, or you’re trying to get to a chatbot, and nobody answers anything. So, I love that you said easy cause that’s so important. So, let me ask you, Dave, for 99% of the people that I interview on this podcast, the number one thing that they’re struggling with is limited resources. How is your organization managing the lack of time or the lack of people in customer support?

Dave Seaton
It’s all about making those strategic trade-offs and when you start with knowing your customer, then that informs where you’re going to prioritize and where you’re going to scale back. That could be with channels, that could be with certain issue types, that could be deflecting the self-service or communities. Having that understanding of your customers lets you make those strategic trade-offs. So, for example, we started 2019 pretty sure that we were going to invest in live agent chat. I love live agent chat as a customer. Many of my peers do as well. We had acquired some technology that could do live agent chat and all that was left was to actually run the project and implement it. But there was this feeling kind of in the pit of my stomach. I wasn’t sure that it was the right thing to do and I didn’t want to spend a lot of money and resources and time implementing live agent chat only to have customers not use it.
So, one of the things that we did in our customer research is we talked to them about their channel preferences and followed it up with a survey. What our customers told us is they said, “I’m sorry, Dave, we’re actually experts in your system.” All of our customers said this. “We’re super users and the kinds of things we need your help with don’t lend themselves well to a live agent chat. We know you’re going to have to go away and work on this for a while and come back to us. So, we would rather see you invest elsewhere than in live agent chat.” So, we didn’t do it.

Andrea Palten
Wow! See, that’s really good that, first of all, you have that repertoire anyway with your customers for that. Let’s face that, a lot of companies don’t have that and that you’re listening to them. That’s really good. I love it. So, let’s talk a little bit about the successes that you guys have had, especially in the customer service department. How do you actually measure success at nThrive?

Dave Seaton
So, regarding our customer service and support, we’ve got three main metrics. The first is operational. It’s our handle time, our turnaround time, are we meeting our SLAs? If we say we’re going to handle a critical issue in this amount of time, how often are we meeting that commitment? So, we’re tracking that, monitoring that. That’s an easy one to track. My team over the last really this year has rolled out a refreshed and updated voice of the customer program for support with new transactional surveys. We’ve got the operational metrics with the turnaround time, and the handle time, and the quality, and all that.
Now we’re bringing in the experience metrics and we’re measuring the customer effort score and the customer satisfaction score. They’re very similar and they’re absolutely correlated together but there are some subtle differences that our clients are helping us uncover where, for example, I can be ultimately satisfied with the result of my customer service interaction, but I can feel like it was really hard, it was a really high effort. So, back to that idea of making it easy we know from our research that there’s a good correlation between low effort and customer retention, less so than customer satisfaction.
So, we’re really focused on how do we reduce that customer effort? How do we make it easy to do business with us? It all fits together. Back to your last question of strategic trade-offs, if I had to go for something that was going to improve customer satisfaction or something that was going to reduce effort, I’d pick effort based on what our customers have told us.

Andrea Palten
Yeah. So, when you’re measuring, are you looking at high effort versus low effort? Who gives you the feedback? Is it your agents or is it the customers that you send a survey to?

Dave Seaton
For those, it’s the customers that we send a survey to.

Andrea Palten
Okay. Okay. Got it. Got it. Alright. That’s really good that you guys are doing that. I love it. Let’s switch a little bit. Now let’s talk about AI. Let’s talk about artificial intelligence. So, since you are a VP and you’re a senior thought leader in your industry I want to know from you. So, this is not nThrive. This is a Dave question.

Dave Seaton
Okay.

Andrea Palten
How do you think artificial intelligence is going to change customer service?

Dave Seaton
I think it’s going to be huge, but I don’t think it’s going to be in the way that most people are focused on right now. So…

Andrea Palten
Oh, tell me more.

Dave Seaton
Yeah. Most of the conversations that I hear about AI are what I would call the front-end, chatbots, interacting with customers, and really putting the virtual agent or the AI right upfront interacting with customers. I think that there’s an absolute future there, but there are several gaps and milestones that we’ll have to cross to be able to get there. For example, we’re still having the debate about should you let the customer know that they’re talking with a bot or should we pretend that they’re interacting with a real person? My view on that, Dave Seaton’s view is to tell them it’s a bot. Call it a bot. Say, hey, I’m the nThrive bot. I’m here to help you because customers know.

Andrea Palten
Yeah. Being deceived as what they would think, and a lot of people do will give them names like Lisa the nThrive bot so you can make it personal too. But yeah, love that you’re telling them right off the bat, this is what you’re dealing with.

Dave Seaton
Yeah. Yeah. We don’t have a bot; we do have some automated messages that are generated by the system. Customers know they’re automated, we’re not fooling anybody. If they open 10 cases and they get 10 same verbiages back, they figure it out real quick. I think the bots right now should be considered very experimental and you should be careful about deploying them because there’s some work to do to get them to be truly helpful. I was talking to an industry peer at a CX round table last week and he had come in and he was reading these chat logs of their bots, and they had designed this bot. They said they were a utility.
So, they said if the customer’s moving, they need to change their address, what better place to deploy a bot? It’s straightforward, it’s easy. The bot can ensure quality and handle the transaction. So, they did, they had this bot, but then the last step was they transferred the customer to a live person to actually verify the information and then make the change in the system. So, what the customer experienced was I spent all this time with the bot, and I typed in all the information and now I’m talking to a person who is asking me to confirm everything that I just typed in. It was a total experience fail…

Andrea Palten
Oh yeah.

Dave Seaton
Even though it started with good intentions. So, what I’m saying most excited about in the near term is AI on the backend, and this is using AI to accelerate and automate the customer service processes that the company is using. Doing things like intelligent case routing. The AI can understand if it can detect customer intent and customer sentiment and match those up with an agent who is good at that particular issue or good at dealing with that particular customer sentiment, and doesn’t have too many cases already, and isn’t on PTO, and all of these other decisions that you’ve got to consider the AI can do that. It can do it very accurately and very quickly, and it can learn. It can try out different agents and different scenarios and learn from that and get very, very good at intelligent routing. So, I think that’s exciting.
I think detecting that customer sentiment and predicting when issues are going to get escalated so that you can avoid all of that firefighting and escalation and a VP gets involved. Now you’ve got a conference call and there are a project team and executive updates. All that overhead, if you can have an AI that can detect those indicators and get that in front of somebody and say, hey, this case is likely to be escalated. Let’s jump on it. I think that has huge potential for reducing all that firefighting, all that overhead, and cost.
So, I’m most excited about I guess AI on the backend. Also, it’s not technically AI but the robotic process automation and just automating some of those repetitive manual tasks that our support agents get stuck with so many times. I think that has a lot of potential in the near term. 

Andrea Palten
Yeah. Yeah. I like the way that you think about it, the backend part, it’s really good. Yeah. It’s interesting when a lot of people think AI will fix everything and then they will implement it incorrectly and it doesn’t work and then they blame the AI, but it’s actually done incorrectly. We didn’t talk too much about our companies before we started hitting record on this podcast but Inbenta, we have a natural language processing way of doing things. So, when people put in their address or they say the wrong thing, or they use slang or they have an accent it all picks that up and it makes such a difference than those old fashioned, clunky AI where you write something in, and the chatbot doesn’t understand you. So, it’s so important, pick the right vendor, ask the right questions, and have the right process in place so you don’t get into something like that example that you gave. 

Dave Seaton
Right. Right. 

Andrea Palten
So, Dave, I want to ask you, what is the number one thing, you can only say one thing. What’s the number one advice that you have for customer service departments?

Dave Seaton
Know your customer. 

Andrea Palten
Yes. Start with that and end with that. Know your customer.

Dave Seaton
Be very skeptical of your own opinions and your own intuition and your internal stakeholders’ and departments’ opinions. One of the most dangerous things that I hear is phrases like, well put your customer hat on. If you were a customer, how would you want this to be? Well, many times the people in the room have never been customers of the thing that we’re selling. So, I can think what I would want as a Netflix customer or as a pizza delivery customer, but I don’t buy healthcare financial software for hospitals and I don’t send out bills. My assumptions are likely to be very, very poor about what those customers value in their experience. So, number one, know your customer, and everything else will follow.

Andrea Palten
Oh, I love that. I like that, how you wrapped it up. We started with that and we’re ending with that. Dave, thank you so much for your time. I appreciate you being here and happy holidays.

Dave Seaton
Yeah, you too, Andrea. It was fun.

Thanks so much for tuning in. This podcast was brought to you by Inbenta. Inbenta Symbolic AI implements natural language processing that requires no training data Inbentas extensive lexicon and patented algorithms. Check out this robust customer interaction platform for your AI needs. From chat, bots to search to knowledge centers and messenger platforms. Just go to our website to request a demo at inbenta.com, that’s I N B E N T A.com. If you liked what you heard today, please be sure to subscribe to this podcast and leave us a review. Thank you.