Southern Philippines. The country's military said he believed the attack was retaliation after a military operation against pro Islamic state groups. A pair of once unbeaten teams are in and an undefeated team is controversially held out of the four team New Year's
Day college football playoff field. It was announced today Number one in unbeaten Michigan will face number four Alabama in the Rose Ball in one semifinal, while unbeaten second ranked Washington takes on third ranked Texas in the Sugar Bowl in the other semifinal. The winners will play for the national championship. The Florida State Seminoles at thirteen to zero, were left out, becoming the first unbeaten Power Five
Conference winner to ever miss out on the college football playoff. I'm Chris Karragio, NBC News Radio or on board kcaa's Inland Extress KCAA Comlinda ten fifty Am, the station that leaves notice year behind. The information economy has a ride. The world is team with innovation as new business models reinvent every industry industry. Inside Analysis is your source of information and insight about how to make the most of this exciting new era. Learn more at Inside Analysis dot Inside Analysis
dot com. And now here's your host, Eric Kavanaugh. L all right, ladies and Doleman, welcome to the future. Indeed, your host Eric Kavanaugh here for another episode of Inside Analysis, the only Coast to coast radio show, all about the information economy and what's the big deal of the information economy these days. The experience. It's all about the experience, folks, the customer experience, of course, the partner experience, the user experience.
That experience had better be good if you're going to keep your customers, if you're gonna keep them happy. There are lots of options these days. It's very easy to leave a certain service provider. You can just get tired and go to a new service provider. So we're going to talk about customer experience and how that gets done, all the interesting stuff that happens behind the scenes and under the covers in order to get you that high quality customer experience.
Now, I can tell you I remember a number of years ago, one of the first companies to figure out that they could grab my record from my phone number. I was like, hooray, they figured it out. They know this is the number associated with my account, they could bring up my call record when I call it to the call center and do a good job helping me. What a fantastic innovation that was. Well, the gentleman on a call today know all about that. We're going to be hearing from Dan
Bodner. He's the CEO and co founder of a company called a Varrant that does all kinds of good stuff and customer experience. And Andy Roberts, who is with the Sabio Group out of UK. They do similar things. And we're going to talk about call center as a service. So what does that mean? So call centers a service is a new kind of technology and basically, as the name implies, it's a service provided to companies that want call
centers. What do you have to do? You have to line up all these different systems, get the data there, make sure you have good people of course on the phone. All that stuff comes in handy. But the experts are going to tell us what's going on. So first of all, Dan Bodner, welcome to Inside Analysis. Tell us a bit about yourself and your amazing journey at Varian since what nineteen hundred and ninety four, that's like the last millennium or something right, tell us what's going on? Yes,
Hi Eric, and hello everyone. So it's all about the ex customer experience, and today it's all about the ex automation. With AI coming to play. Finally we can use bots to help the agents to improve customer experience. So happy to talk about that. But a little bit about variant. Yes, we started in nineteen ninety four. At that time it was all about taking unstructured data, mostly voice and speech data, and analyze it to find insights in the data. Ten years later, in the two thousand and six
timeframe, we created workforce optimization. This was about giving the workforce tools so they can perform their their work better because you know, conduct center, as you mentioned, are very labor intense, intense, there's a lot of people working. So we kept that journey with marrying data mostly unstructured data, initially speech but also text and video and giving workforce engagement tools, you know, more capabilities. But really what's exciting is a few years ago we started to
put together very da VINCAI at the core of the platform. So the workforce is starting to use AI and of course today with Jenyi, the workforce is no longer just people, it's really people on bots working together, and that increase the capacity of the workforce tremendously and allow the workforce really to get the more enjoyable and elevated customer experience for the consumers, which is really what the
industry has been trying to do for many, many years. But because it's very expensive to hire people, it's also very hard to find them and train them, and then they don't have time really to delight the customers. So now is the help of bots working side by side with people. I think we're getting to now to a point that we can actually achieve more with the same resources, same budgets, so our customers can delight their consumers with better
customer experience through the consumption and boughts. Yeah, I think you hit the nail on the head there with the same budgets line, that's the kind of thing that makes the coos at your clients very happy in the CFOs because you typically hear about new, fun, interesting things that cost more money. So if you can deliver a better experience, a more streamlined experience, for the
same budget, that makes people really happy. So the vendor experience is pretty too, right, Absolutely, our customers are happy because they can contain the budget, but also they want to do better job. They know that they can differentiate from their competitors by giving better customer experience. So customer experience is
not lost on brands. It's just that it's expensive and with new technology and people on bot's working together, they can now afford it and do more with the same budgets and at the same time delight to their consumers so they can get more leyalthy, they can get you know, better revenue generator from a
generation from their customers. So it's a win win win. The consumer obviously wins, and of course we all consumers ourselves, so we love to be on the call where everything goes very smoothly and we get a contextual and quick responses. So obviously it's a win for the consumers, and it's a win for our customers because they can do more with the same budget. And it's
a win for Variant because we get to sell new technology. Yeah. Well, and just real quick, before I bring Andy into comment on this, I'd like to share with their audience there are so many ways in which machine learning and AI can help and the automation component in particular. For example, skill path routing is something I learned about some twenty odd years ago. Over a course of time, if you have machine learning in place, you can
track which call reps do better in certain scenarios. It could even be with certain demographics, if someone's calling in from Georgia or New York or New Jersey or wherever. It could be the location, it could be the age group, it could be the topic at hand that someone's calling about. And to be able to ascertain that kind of thing while the call is coming in and then route that path to the appropriate person. That's just one of many ways
that you can really optimize the experience. Right, Absolutely, you hit the nail on it. It's all about data, right. If you have the right data, then your AI can do machine learning on the data. And
data power is really the very open platform. So basically, this data that we have in our platform because we have been recording agents for many, many years, and that agent behavior is really the source of the data that empower the bots to do machine learning and emulate agents and try to do the job that and take some of the functions that humans are doing replace it with automation. So yes, it's data and there's lots of use cases for the data.
You mentioned one, and I'm sure we're going to cover a few more on the show here, because the data does power the AI, and the AI powers automation. Yeah, it's really wonderful stuff because what you're doing is you're leveraging, first of all, the organic energy and success of your team members. So everyone knows when you dial in, this call may be recorded for quality monitoring and assurance. Right, everyone understands that. But these days
you can automate that process. Now. Twenty years ago, even ten years ago, most of the time there was just some manager who was listening and it would take manual notes. But now with the power of AI, there are all sorts of technologies. Gone comes to mind. There's some others out there that are very interesting, and I know that you guys have a whole bunch of good stuff too, where you can do sentiment analysis even in real
time listening to someone to understand, Oh, this person's getting upsetting. Hit a red flag, bring in the higher level service. Oh it's a high value customer too. Uh, let's go all out make sure this person is taken care of. You can trigger all those actions because you have this foundation of data, a real world data of what people said, how they said
it, how timely the conversation was. All these little tidbits fold into a recommendation engine that then gets the job done for you, right real quick, Dan, Yes, absolutely, So the data is the foundation. We call that the gym. That the bots need to really go to the gym every day twenty four to seven, twenty four x seven and they train on that data so they become better and better and more accurate to augment the workforce. So, yeah, the data is key, and real time obviously is very
important. How do we help the agent in real time, whether it's suggestion knowledge or as you said, pointing out the sentiment that needs to be addressed because the customer is frustrated they don't really hear what you say, right,
so slow down, take to time, show some empathy. These things can be done in real time because now all this technology is running in the cloud and you have elastic process in power, so you can actually apply posting power in the cloud to create real time AI outcomes business outcomes for the agents. So the technology of cloud technology, AI technology coming together empowered by data is really the Sickeret sauce for increasing c exformation. Yeah, that really is.
It's the confluence of all these factors. As you suggest, cloud computing AI, which is really mature now. I mean AI has been around arguably for forty five fifty years. We went through a couple AI winters when the hype got too high and there just wasn't the compute power that we have today. But now it's everywhere. We have many different cloud providers. You could do some of this stuff on prem you don't have to go to the cloud.
But it really does help to have this marshaling area because guess what, those folks are busy making sure those servers hum as best they possibly can all day long. That's their job. So your job as the company using that service
is to optimize what you're doing. And as you suggest, having that wealth of information, of rich unstructured data of audio files, text files, process models, all that fun stuff, it all comes together to create this springboard basically, which is now in action, right Dan, Now, right now, that kind of technology is happening when people call the call setters today,
right Dan. Absolutely, we have hundreds of boughts deployments across many the largest brands in the world are now adapting AI powered bots, so it's no longer just a high when you bring the bots into the platform and you embed it into existing workflows, so it doesn't disrupt the workforce, right, it's part
of the natural flow of what the agent does. Because look, contact center agents don't have time to search with AI tools, right, They're not going to get on the internet and search because that means they have to put you on hold, and obviously that will take a longer call and a very poor customer experience. So the goal of today's platform is to bring that AI to the fingertips of the workforce, so it's embedded in their workflows and it can
really power them to do to be more productive. And the way we think about bots here at Variant is we created many, many bots. We have thirty five bots right now in our platform. Each one is doing only one thing, but they do it very well. So again I give you one example. One of the things agent needs to do at the end of the call when the customers sign off is they have to summarize the call. Now, they don't like to do the summaries because it takes them time. They're
not really good in typing and they always been criticized. Hey, the summary is really terrible English. So guess what you know? They call the rapprobot and instantly they get the summary prepared and it's better summary that they can do themselves. So the agents are happy. You can cut sixty or ninety seconds of the call, which is a huge amount of savings, and obviously it's a win win win for everyone. So this type of Jenny I is not
just technology. It's really technology that is the finger tips of the agent when they most need it. And this way you can really start to create business outcomes. Yeah, that's wonderful stuff. And just to help our audience understand actions like putting someone on hold, I promise you no one likes to be put on hold. Okay, everyone to deal with it. They were like, okay, I'll get put on a lot. It's fine. You go and go back to your computer, start typing or doing something else as you're
weighting. But no one wants to be put on hold. So if you can lower that number in particular, that's going to greatly improve customer experience across the board. And I see Andy nodding his head in the background. There, let's bring in Andy Roberts from Sabo Group has been working with the folks at Variant for like decades, for a long long time. And Andy, I saw you smiling there. So you know that this whole concept of open
call center as a service, we'll talk about that today. It really reflects an inflection point I think in the industry right because and you see this across the board in terms of it, you see all sorts of things as a service, managed service providers. That just makes life easier for the customers who are just trying to get their business done. But tell us about yourself, Andy, and call center as a service. Up, you're on mute. I think you're on mute. No, I can't hear you. It doesn't
say your arm on mute, but I can't hear you. Let's go back to Let's go back to Dan while he's figuring that out. Something must have happened with your microphone. We'll get you back on in just a second here, But Dan, I'll throw it back up to you. Call center as a service in particular, open call center as a service. Talk about what that means. I mean, you talked about this open varied platform. What
do you mean by open call center as a service. Yes, So you know the technology now is moving so quickly that customers are really not want to be disrupted by a transformational change, right, so you can just throw away everything you have in the contact center and start with the next technology and guess what, a year from now, it will be even better technology. So
he can't really restart every year. So you need an open platform, one that integrates into the ecosystem that you have today and that you can plug and play with pieces of mute technology that you can add to your legacy solution and evolve over time. And especially when it comes to open data, because we agreed right, data is the second sauce here, so you really need to bring all the relevant data into an open platform so that all the applications running
in the platform can benefit from this data. So open data being the ability to collect silo data and behavioral data. You know, like we discuss agent how agents behave is really locked up in silos across the enterprise. So bringing all that together to an open data hub using open da Vinci, which is our AI engines, so you can leverage commercial models, not just propriety models.
Because there's so much innovation industry around AI engines. So the ability to bring in the latest AI engines into the platform, quickly train it on the data in the gym, and then embedded into workflows. That's creating an open environment where customers can evolve, and you know, we have customers that really move quick, but other customers need to evolve at their own pace because you know, running a connecenter with five thousand agents, it is not a small
headache for the COO and you can't just disrupt yourself. So we designed the platform to be open with that in mind, and also hopefully Andy can comment when when you get his mic back. But hoping is great for partners because partners can really develop all kinds of value edited services around the platform to help
customers to consume AI more effectively. And there's so much they can do, whether it's data practices and getting insights from data so they can improve you know, the to bestic class operation, or or taking bots and integrating these bots with the customer environment so they can create more connectivity. Because eventually the bots, you know, even if they understand the human being, their ability to
respond is really based on how well they integrated into the customer ecosystem. So a partners can can leverage an open platform in order to really differentiate themselves at services and delight our joint customers. And and uh, you know, uh, we've seen a lot of activity for partners taken advantage of being open.
Yeah, and that's that's a really key point about open standards and open systems, open platforms, because we are dealing with an ecosystem, and when you try to do everything yourself, you're never going to do the best job. If you can enable your partners to work within your ecosystem and leverage their expertise, that's when you really get the power of a whole environment, of a
whole ecosystem. Well, folks, don't touch that. That will be right back talking to a couple of experts in the field of customer experience, Dadminitor and Andie Roberts. Don't touch that, doll. You're listening to Inside Analysis, Expected Recording and Progress. Welcome back to Inside and Analysis. Here's your host, Eric Tavanaugh and take this to all right, folks, back here on Inside Analysis talking to a couple veterans in customer experience. These folks really
know what they're talking about. We're talking to Dan Badner's CEO and co founder of Variant, founded in nineteen ninety four. I could do some quick math. That's thirty three years ago. That's pretty impressive. You've been working with lots of different customers and you got thirty three bots. I think you said thirty three or thirty five each, thirty five each two different things, and a lot of people are now figuring out what this jenai stuff is all about.
And it's not just being able to spin up fun, little creative articles. There are lots of other things it does. Summarizing is one of the amazing things that it does. You can take a big, long paper, feed it into one of these engines and say give me a two page summary, and bam, it does. I actually did a fairly complex ETL job on Friday, scraping a bunch of content from the web and then using chat GPT, actually using bard by Google to create an ordered list, and it
was like magic. It would have taken a couple hours for a good ETL person took me five minutes with these new tools. So there are very cool tools fueled by AI that are changing the game. And I'm excited to bring Andy Roberts now in from Sabio Group. Andy, you had some technical difficulties, but we've got that solved. So I always blame the Russians for hacking. By the way, I'm pretty sure it's their fault. But Andy, tell us about your perspective on contact center or call center as a service and
how it changes the game. Okay, good Athian everybody. I'm calling in from London. It's actually not raining. It's beautiful out there at the moment. I can see some pools over my shoulder. So yeah, I'm at Saber twenty five years, so not quite as much as Dan, but we've been working with Variants since two thousands and been on the journey with them.
Sabby are a specialist services, expert services company that really absolutely understand how to be able to deploy the technology and to be able to deliver return on investment and excellent customer experience. Everyone here gets up every morning trying to make the customer experience brilliant. As far as the journey is concerned around contact centers call centers, the journey is quite an evolving one and it's going at pace.
The relationship that we have with Variants going back twenty years was very much around on premise contact centers with all of the workforce engagement management technology that was on top of it. What we saw probably about seven or eight years ago was a move to cloud, and certainly the move to public cloud has really accelerated, and that's where the SEACASS development and be able to move to the public cloud has really gained pace. And what we've seen is that organizations of delivering
and moving to the cloud at different speeds. Digital native organizations are moving very quickly and larger scale enterprises are moving a little bit slower. As far as open seacats is concerned. What we're able to see is we're able to see the ability to be able to not just move to the cloud, but it's actually to be able to augment, to be able to bring other parts of
the customer experience to life. And that is the whole of the variant portfolio quality monitoring, speed, channeltics, workforce management, but also all of the AI components as well. And Tabby have had a relationship with automation as long as we've been in existence. It started off with IVRs and it moved to
natural language speech recognition, and more recently it's been around conversational AI. So when we actually start talking about the ability to be able to make a difference for the end customer, there's one key thing that I think is important, and that's about the appropriate service at the right time. So we've all experienced issues with automation when the experience has been poor and I think you're smiling there,
Eric and you're thinking, oh, I can remember that time. So actually looking at how you want, how organizations want to ensure that they provide the experience that's really the right time for the right type of transaction is the heart of what we're looking to try and do, and the whole of the
variant portfolio allows us to be able to do that. But it's about thinking about the right transactions to be able to automate, and those where you need a live agent, being able to have the right context, the right information to be able to delight the end customer. Yeah, that's all really important. And you hit one nail squarely on the head there, which is knowing
what to automate, how to automate, and when to automate. And of course you always want the manual override correct, like when things are going bad, you want to be able to pull the handle and slow it down and bring in the agents, and that all really goes back to the data and the architecture too. Do you want to talk a bit about the architecture of these solutions, because as you said, you started off on prem. Now I'm sure heavily in the cloud, but I think on prems legacy will live
on for a long time. As they say, the rumors of on Prems demise have been somewhat exaggerated. But tell us how the architecture is changing from your perspective. I'll give you an example of a very large mobile provider in Europe and the architecture that they have. They've got an on prem architecture and they've but as far as all of their AI and automation, that's done in the cloud and that is utilizing well. So this specific provider is he's been
able to automate fifty nine percent of all the traffic. Now that is a total of thirty six million calls a year are being automated. But the key thing about that is being able to drive the net promoter score up. So over the same period they've increased their net promoter score by twenty seven points. So being able to look at the two things in conjunction, I think is
really important. The other thing you mentioned, which is around being able to migrate to the cloud at the right time, I think is really important. What we've seen is people moving to the public cloud, but they're replacing a lot of the technology that was on premise, and they're replacing it on the
cloud without getting any of the real benefits. And I really think that the types of solutions that Varrant and Sabio are looking to try and put together is being able to actually extract the value for the enterprise customers that we work with. But we've got to think about two major components and two sets of populations
as we do. So. The first one is the end customer. So it's about having user centered design engineers sitting on the customer side of the fence to actually sit in your shoes when you're actually engaging with an insurance company, an airline, or a utility. And then secondly, it's about having speech scientists about being able to tune the box and being able to tune the AI engines to allow us to be able to maximize the operational run for customers.
And what we're seeing is that that's where the next phase of automation and workforce engagement management is sitting in that space. Yeah, we're really going to see a lot of change I think over the next probably two to five years in terms of who's doing what in organizations. And I think the real success factor is going to be getting team members who are willing to use the new technologies, who are willing to embrace new ways of doing things as opposed to fearing
the use of these technologies and trying to work around them. What do you think? Yeah? Absolutely, And we were talking a little bit at one of the management meetings I was in this morning. Dan spoke about summarization and being a really really good benefit from a generative AI perspective, and I agree totally. But there's also auto translation, which is incredibly useful, especially in Europe. But also there was a really good use case, which is when
somebody sends her an email. Typically the SLA on an email is one day, two days. So if a customer ends up sending through an email and wants to cancel their hotel booking for two days time and they send it by email, that might not get looked at. However, we've designed a use case specifically for this for one of our travel customers, and specifically, what I was looking to do is to be able to understand the email that comes
through that if the data cancelation was the twenty ninth of November. It actually uses the knowledge that's in the email to be able to root it directly answer the question, wow, present an email and go back to the customer. But we could also have human intervention into that where you could actually root that to a live agent. Now I believe that'd be amazing customer experience. And that's the type of way in which we can use I believe generative AI to
really really move the dial. Yeah, that's amazing. I mean really you think about it. The power of AI, the power of analytics, the power of automation, these things come together with appropriate data. Of course, a baseline of data is requisite. I mean, you can use adversarial models. I don't know if we'll have time to get into all that, but to be able to capture to receive the email, understand the syntax, the
key messages inside there, and then take action to respond to that. Wow, that's what we've been hoping for, like the last twenty thirty years, right, absolutely, And if you think about that recording starting from progress. Sorry I was trying to hit mute there hit the wrong button. Go ahead.
So if you think that from an agent perspective, if you're actually having the message being typed for you and suggesting you're just checking, you know, I think that reduces repetitive work and it allows you to really go and help
the customer, which is what we're all here to do. Yeah, I mean you think about the different component parts coming together and being able to serve the purpose of helping out the client and also have a record such that when the person who works at the company comes and run and checks in on that, they can see what happened, they can score. And that's really the curation side of this whole storyline, right, is being able to monitor.
I've seen some of these really good tools lately, monitor what the GENAI is coming up with and being able to curate that and say yes, no, or maybe tweak this a little bit until you get it right, and then you let the automation do its thing. Right. Absolutely, and I think that you know, the individual that was demonstrating this was from Glasgow, all
right, and there's he actually was able to get personality. So we saw the first version, which was a very polite response, but when he actually changed it and said could you have it in glass region all right so you can pick it literally, you had a completely different email that was being written, which took which put real personality into the response, which I think for various different for certain companies and certain populations. I think it's a superb way
of being able to give the box some real identity. Yeah, that's that's also amazing that they can pick up on dialects and different ways of phrasing things and then sort of reflect back to the user what they wanted to hear in their own language. That makes people feel comfortable. I mean, we heard Dan talking about it earlier. This term delight It is delightful if you feel you've been heard, if you feel that your situation has been handled properly,
that's when you're a happy customer. That's when you come back. And that's the whole objective of customer experience, right Andy, Absolutely, absolutely, you want to ensure that when an organized when somebody is interacting with a company, it's an appropriate service that dealt with quickly, swiftly. Twenty four to seven through automation where it's simple and it needs to be relatively easy to use. But when it is a nine to nine to nine call, your car's broken
down, you need road side assistance, you're in distress. Sentiment analysis was mentioned by Dan earlier on all these things. You want to if somebody is getting angry, getting frustrated, you want to try and diffuse the situation, and you want to be able to use the appropriate service, which quite often
is a live service. Yeah, and we should point out too, it's not just humans or bots, it's really humans with bought assistance or augmentation, right, Because you're going to have a human on the call with someone, but the AI engine is in the background listening to things and can figure something out and then boom, make a suggestion or you know, come out and
say, hey, offer this offer. That the idea is that you get this team basically working with you as a call center operative talking to the customer, and behind the scenes, all these little bots are doing their thing to give you advice and to help you along the way. Right. Yeah, But being able to provide assisted service and being able to take out any after cool work, being able to ensure that the repetitive task for the agents as
well is being reduced. All these things will make people's life much better. Yeah, I love it. Maybe we'll bring Dan back in here to comment on some of this stuff too. Dan, you know, when I think about the coalescing of these different technologies in different workflows, once again, if you have the data underneath to map it, we're dealing with a completely different
environment than mean were just two or three years ago. What do you think, Dan, Yeah, you know, we focused on the bots that are helping the agent in real time, but we have boughts that are helping other personas in the connectenter for example, the compliance team, right, the people that their job is to make sure that all the interactions are in compliance,
especially in healthcare and financial services. And these people what they need to do is actually they need to sample a small number of calls and randomly listen to this calls to make sure that all the mandatory statements have been made and the call is basically compliant. And you know there's sometimes you know, there's industries where you have to make the mandatory statement within the first thirty thirty seconds of
the call, otherwise you ought of compliance. So our compliance bots will actually listen not just to a random sample, but compliance what we'll we'll listen to one hundred percent of the call automatically fine noncomp line's issues. So it really is another bought that helps the compliance t The bots are helping the supervisors,
the bots that helping managers to make better decisions. And so when you think about all these bots helping different roles around the conduct center, now you can get really the whole machine to do what end said right, which is to focus on how do we actually do the better job for our end customers. And it's always in the same budget and resources. Yeah, that's a really good point too. And as I've learned over the years, these bots,
they do have to have very specific tasks. I mean, I'm sure we're going to be getting better with these things to where they can be more versatile and do different things, but at the point in time we are right now, you want them to do very specific things and to do them well and to measure all of that, and folks, don't touch that down. Will be right back. You are listening to Inside Analysis. Welcome back to Inside Analysis. Here's your host, Eric Kavanaugh. All right, folks, back
here on Inside analysis talking to a couple experts in customer experience. These are the folks behind the scenes who make the experience good. And I promise see there are many many other folks who are working on this stuff. They're writing code, they're building systems, they're checking things, validating, they're curating. That's we talked about a moment ago. That's one of the big functions in the modern world. And you know, really, this Jenai stuff is amazing.
So I'll ask each of you to kind of comment on how Jenai plays a role in customer or call center as a service, as a technology space. I've been playing around a lot with these Genai tools, and I had the CTO of Boomy on the show a couple of weeks ago, and he made a really good point, asked, and what's the most important aspect do you think about this nu machine learning technology, this AI? And he said learning and he meant human learning. And so Dan, I'll throw it over
to you first. You know, we as humans, we need to understand this is like a new kind of surfboard or roller blade. It's a new way of doing things that has not really been around it to any great extent in the past. So it takes a lot of practice just to play around
with it and see how it operates. What do you think, Dan, You know, we we spoke before about some bots of the system agents in real time, and you know, one of the important things in real time assists is that the boss will be really, really accurate, because agents don't want to be disrupted by bots of giving them some stupid advice. So and I remember no longer, you know, three years ago, we've sold some systems with AI that certain agents just turn them turned them off because they said,
no, we don't need that. You know, we know better than the bot, right, and that disrupting me with some coaching that actually is not helpful. So key to to UH agent assist is is that the boss will be really accurate. And the only way they can become accurate is if
they train on the right data. So when you develop an AI model, obviously you need data that's the initial training, and that initial data can be captured and it's a limited amount of data and you can create an AI model that works, But in order to be really accurate, you really need the boss to continue to train on fresh data. And that's twenty four to seven
training in the gym on fresh data. Because the data changes, the discussions between the customers and the consumers and the agents are actually changing because you know, if you are from a circle company and you announce a new job, obviously the conversation about the side effects will be different than new So it's not
just static data. It has to be fresh data. And that's why this boss need to live in a platform where the data is being constantly collected into a single hub, where it's all unified and available for the bots to get trained and get better and better all the time until the point that they really are doing what they're supposed to do very well. And you know, Eric, you mentioned before the break that maybe over time bots will be able to
do more and more and more. But you know, if bots can really do one thing and do them well, that's okay because then you can create another bot to do something different and again make sure that that new bot is very good at what they do. So the approach we take it variant is we have today thirty five bots, and you know, we said we're going to have another fifteen bots in the next couple of months, so we're creating a lot of bots quickly. Because the data is only did the platform,
and we bring new AI morals, not just at variant development. We bring it from any company in the world that creates new including a lot of open source. We bring that new AI model and we train them on our unique data, this behavioral data, and then very quickly we turn them into something that is really helpful to automate the single test and our approaches want tusk of
the time, you can make the whole conneccenter be really really better. Yeah, division of labor, right, so you have different bots to do different things. And to your point, and maybe I'll throw this one over to Andy, situations do change. I know one of the things that we learned during the COVID period of time is that many of the models that were doing very well before COVID did not do well after COVID because behavior changed, because
people stayed home a lot more, they were buying stuff online. The whole humanity out there changed behavior significantly, and that required a lot of effort from developers and from companies like yourselves to be able to re understand and reconfigure and sort of realign what the algorithms we're doing. What do you think about that? Andy. So as far as the contact center and customer experience, all the data that it's been talked about by Dan say that the contact center becomes
the eyes and ears of the organization. If you think about the amount of data that comes in through all the relevant channels, and you think about the sentiment analysis, and you think about the areas where customers are really excited or really frustrated, what you have is and this is through the contact center, it's through all the workforce engagement platforms CRM and everything that's captured by AI.
There's an enormous amount of data. So when you start talking about how to be able to look at generative AI data is the new goal, all right, So being able to ensure that you can access that data and mine it for the information that you're talking about about when things happen with COVID and when people's working patterns change, you can see it happening through the data that's collected and in real time. Yeah, and you make a really good point about
the quality of data and the context, the appropriate context of data. So one thing that I think a lot of people learning is that these large language models like CHADGBT and BARD, they were trained on the corpus of data that
was vast and really whatever was available on the web. Twitter. Elon Musk was talking about how he did rate limiting because they realized that people were scraping massive amounts of data from Twitter in order to ascertain trends and then come up with models to train them around particular topics, etc. The more focused that data is, the better a chance you're going to get of this spot doing
the right thing. If you have very noisy data, if you have data of all sorts of different things, it's going to be a lot harder. It's kind of like a It really is kind of like raising a child over time. Like if you have a good structured environment and you're good to your child and you're caring, you're going to have good results. But if the child lives in a chaotic environment, there's all kind of weird bad things happening, all that is going to reflect through the child's eyes at some point in
time. The same is true for these bots, So you want to be very careful about how you train them, which data you give it access to, and that's really important stuff. I'll throw that over to to day and to comment on what do you think there, Yeah, you need to unified data hub. You need to bring all these data, whether you physically bring
it or you just link to the data. But all that data that is behavioral data that is being captured by recording if you will, the human agents talking to real customers, that all that data they could be talking, could be chatting, could be social media, right, it could be surveys. All that data, all the interaction data, is the behavioral data that you need access for the bots. And and and if if you only have partial data or you have noising data, yes you're going to get partial results or
just bad results. Right. You don't want bad results. You don't want your bots annoying people. That's pretty much the last thing you want these things to do. You want them to be understanding a situation and being able to act in a particular environment. And that takes time. You curate it over time. I would tend to think, right, you deploy a bot for a particular customer, and you want to be able to have this unified data hub so you can train that butt in their environment so you can learn.
And we get back and real quick, I throw this over to Andy. We get back to training it on data relevant for the company and for the use case, right, and using your data, your historical data, and then figuring out who are the good call reps, who are the people who have good reports and good results. Let's train the data, train the bots on the data from those calls, not the data from the bad calls. Right, Andy, Yeah, I think that two points. The first one
is good prompt engineering, I think is really really important. Being able to
accurately present the right prompt to get the right outcome is absolutely essential. I also think that as far as being able to provide the best customer experience, the way to be able to get the box that Dan's talking about is spend time with customers, understand their problems and to be able to understand what the good agents do, and then trying to replicate that with being able to use the data that's been collaptured from all the various different sources to allow us to
be able to ensure that we're delivering the best possible outcomes to automating utilizing the best behaviors that have come from from those agents. So you go ahead, we talk about best practices, right, We've talked about best practices for like forty fifty years, probably longer than that. But what you want now is for those best practices to be codified. And what's a beautiful thing is you can score all these behaviors. You can score them, you can understand.
We have net promoter score that we understand. There are metrics that you use to be able to gauge where you're getting. This happens all the time in call centers. We want to reduce customer return by five percent, we want to increase engagement by five percent at targeted, metric driven goals, and then you work toward those goals and that works very very well. I can tell you that pretty much anyone in the working world, if you have a job somewhere, if you know what to do and how to do it, that's
the first key to success. And then if you're measured on what you do, and you're measured honestly and transparently, that's wonderful stuff. I mean, sporting folks can do that all the time. They know when they get a first down, they know when they get a touchdown. It's pretty obvious, and in the business world it's not quite as obvious, but it's still pretty clear. When you get a good customer engagement. You feel good about that.
That's what people want, That's what everyone wants. That's the key to exceptional customer experience. Folks who've been talking to Andy Roberts and Dad Bonder. What a couple of great experts you've been listening to Inside a oaus. All right, folks, time for the podcast bonus segment here on Inside Analysis. What a fun show talking to Andy Roberts of Savio Group and Dan Bodner of
Variant, the CEO and co founder. And Andy, you've got a fun story about a very large insurance company and the CECAST in particular, which is call center as a service. Go ahead and tell that story. It's a little bit unusual. So it's a customer that we've actually been working with for a long period of time, specifically around automation. Anyway, they had a five day outage. It's a global insurer, a five day outage, which
was pretty traumatic for them. We were able to spin up a seacast platform from scratch in twenty four hours, all right, and it literally they had a huge issue. It was in the US and in the EU, and we were able to take four and a half thousand calls in the first day,
six hundred agents. When people talk about the ability to be able to look at seacasts as an environment, now this was a very basic, you know, dr as a service solution that we put in, but we were looking on the customer side of things and being able to say, right, you need you need telephony, and you need it really really quickly. The guys basically took the solution overnight and we're able to replicate it back and give it to the client. Now that is how quickly a seacast solution can be
deployed. However, what you then need to be able to overlay is all the really really interesting technology which allows agents to be effective. It allows for the customers to be able to maximize customer experience. They need to look all the automation engines at the top as a site. But as far as being the speed of which things can be deployed in the SEACAS public cloud environment, that that's an example of how things have changed enormously. Now, is it
going to be able to offer a world class service. No, but what it can do is it can change things rapidly to be able to improve customer experience. Yeah, that's a fantastic story that just amazing stuff. Herculean effort is I guess what I would say, and you know, Dan, I'll bring you in for some final comments here. You know what really excites me about all this stuff is the ability for these engines to identify success and failure
and even just interesting stuff. I mean, you work with a Jenai tool, you can say, give me the most interesting three parts about this twenty page paper. It will go find interesting stuff. And I've been surprised at what a good job it does. So my point is that you know, for so many years now in business management, in sales and marketing, we've kind of been going by gut instinct. I mean, we'll use data from
dashboards and different things like that. But I think what we're seeing right now is a see change, is an inflection point, It's a j curve, whatever you want to call it, where all of a sudden we can leverage the power of real world data at scale to understand what are people really buying, what are people we're really excited about. What does make for a successful phone call? We had to guess for many years we had a pretty good
idea, but not like we can do today. When you can when you can capture, analyze, measure, report, and then optimize that whole cycle is really powerful these days. And if you have this unified data hub, as you've suggested, you're kind of off to the races. What do you think about all that, Dana? So, yeah, that the data is the key, and the bots that are working and using this data to deliver
business housecomes. I think that's what's really excite our customers. So you know, bots are not just helping people, and we're now at the point already where bots are helping bots and and I'll give you an example. So when you call, let's say you want to change the quality on your order, and the containment bots will be able to say, yes, mister Cavanaught took
care of it, so you're done. But then if you want to change your payment schedule, and the containment ball will actually, I'll have to transfer you to the proper agent. At that point, the transfer bot is coming to help the containment bot and then transfer the call with all the right context so the agent does not have to authenticate you again, does not have to say how can I help you? Right, you can go back to the point and say, oh, so that you're trying to shag your payment schedule.
Let me help you death at that point. So all these bots now working on the data, as you said, and fitting from the same data, and understanding the context and the flow of your call, of your prior calls, of your history, all that coming together now in a platform that can really eventually increase the extoltomation, which the industry has been looking to do for many, many years. Yeah. Well, there are so many tedious
tasks involved in any job. And if you can get these bots to hammer away at the tedious things you achieve, and I'll make this the final point I get each of your comment on it, you will improve morale in your organization. I believed for a long time now that morale is the single most important characteristic of any organization, because when morale is high, good things happen.
When morale is low, you can have all the money, tools, the best people, and good things are not going to happen because morale is down. And these bots, when orchestrated effectively and efficiently, First to Andy and then Dan, just comment on this, that's going to do wonders for morale, That does wonders for customer service. What do you think Dan or first, Andy, go ahead. I totally agree. If you're sitting in a contact center environment, you want to have a really really strong agents are
motivated. If you look at any reports that comes out around the biggest issues that customers experience for the contact center world, it will be being able to retain quality agents. Yea, the hardest thing to do, all right. You'll see reports, you know, eighteen ninety percent CEOs will say that the big issue they have is with training, with attracting and retaining quality people. If you're able to remove the boring, repetitive taskt then you're on the right
path to be able to move forward and give them more interesting work. They want to get out of bed every morning, and to be able to excite customers and have happy customers at the end rather than grumpy ones that complain. That's right. Good help is hard to find. To be keep them around, you're going to be in really good shape. Final comments from Dan Bartner a variant. So yeah, so ex employee experience is the driver for customer
experience. And since today I'm talking about bots, i'll give you the latest part that we're just announced, and that's a time flex spot. So if you're an agent and your child is sick and you need to get them to the doctor. You call your supervisor and you say I need to get out of the shift, and the supervisor will say, nope, cannot do it. We need you. So flexibility for the workforce is a key UH, a key aspect of morale. They really want to have more flexibility in their
lives and they need flexibility in terms of changing the schedule. So the time flex bot that we created is using something like the Uber system, so you can trade points or coins. UH. Basically you can opt in into a shift where there's really shortage of people, and when you opt in, you gain coins. Now you need to take your key to the doctor, you pay with points, and you get out of that shift, and then somebody else will take your points, and the bot is actually recalculating the schedule so
that overall the customer queue is going to be at the right level. This is something that the industry has been trying to do with people for years. But you have to hire enormous amount of supervisors to change schedules all the time and find an optimal schedule in real time, and that's just not possible. Using that Uber system and you know, and you get points and payments points
based on whether it's a desirable shift or not. So if you opt out to the midnight shift, you can gain a lot of points, and then when you have an emergency, use the points to opt out of that shift that you need. You need flexibility, so you can see the pots. Bots are there to help people in every aspect because at the end of the day, people are making customer experience. I love it the bots helping bots, bots helping people. We've been talking to Dan Bodner of Variant and Andy
Roberts of Sabio Group today. Look these folks up online. What a great show customer experience. That's what you want. We'll talk to you next time. You've been listening to Inside Analysis. Thank you the legacy Southern California's k c a A, the number one talk radio station in the Inlet Empire. Now here's a new concept, digital network advertising for businesses. Display your ad inside their building. If a picture's worth a thousand words, your company is
going to thrive with digital network advertising. Choose your marketing sites or jump on the DNA system and advertise with all participants. Your business ad or logo is rotated multiple times an hour inside local businesses where people will discover your company.
Digital network Advertising DNA a novel way to be seen and remembered. Digital network advertising with networks in Redlands and YUKAIPA call in the nine to nine area two two two nine two nine three for introductory pricing that's nine o nine two two two nine two nine three for digital network advertising one last time Digital network Advertising nine oh nine two two two nine two nine three. T Hebot Club's original pure power d RCO super Ta helps build red corpusoles in the blood, which
carry oxygen to organs and cells. Our organs in cells need oxygen to regenerate themselves. The immune system needs oxygen to develop, and cancer dies in oxygen. So the t is great for healthy people because it helps build the immune system, and it can surely be miraculous for someone fting a potentially life threatening disease due to an infection, diabetes, or cancer. The tea is also organic and naturally caffeine free. A one pound package of tea is forty nine
ninety five which includes shipping. To order, please visit to hebot club dot com. T Hebow is spelled tea like tom a, h ee b like boyo. Then continue with the word T and then the word club. The complete website is to Hebot Club dot com or call us at eight one eight six one zero eight zero eight eight Monday through Saturday, nine am to five pm California time. That's eight one eight six one zero eight zero eight eight to Hebot Club dot com. With sixty years of fascinating facts. This is
the man from yesterday and back in time for this time. In nineteen sixty five, actor Frank Sutton, who portrays Sergeant Carter on former Pyle USMC on CBSTV, is made an honorary sergeant by the Marine Corps and Special Ceremonies. The honor coincided with Frank Sutton's participation in the Marines Toys for Tots program. What are you Nurse? Pile? A private comer? Pyle USMC, Hi, Sergeant, you see how you don't have to tell me what I see? Pile? I know what I see? What's the matter with you?
Knock? Go ahead? You learn all list in boot camp and from this time. In nineteen sixty eight, the rock band Rascals announced a new concert policy, effect of January first, nineteen sixty nine. The Rascals will not appear in concert anywhere unless half the acts on the bill are black. You should see Lovely, Lovely Wells and back in time to this time in nineteen
fifty eight. The fastest moving chart record of this year here in nineteen fifty eight is the chip Monk Song, with sales so far at two point five million. That was very good. Simon's very good, Theodore Alvin. If you were a little slat, watch it on Alfon Alvin with more at man from yesterday dot com. It's that time of year again, No, not the holidays. Medicare open enrollment and if you have questions about Medicare, you
should talk to the local experts. Paul Barratt and Associates, all of his agents are certified with plans that are accepted by most of the medical groups in our area. Call nine oh nine seven nine three oh three eight five. Their service is free and after forty two years in the business, their agents
are trained to help you pick the plan that's right for you. I'm Rick Smith, host of the Rick Smith Show, inviting you to listen to my show during the noon hour every weekday right here on KCIA My show is sponsored locally by Teas nineteen thirty two, a strong union with fourteen thousand members in the IE. Our message is clear, unions improve the lives of working people. You have a right to form and join a union, So go to
Teamsters nineteen thirty two, dot org and get started now. Killistina KCAA Lomalinda at one O six point five FMK two ninety three, CF Brino Valley, NBC News Radio. I'm Chris Garagio. Israel says its defense forces are now operating in the Southern Gaza Strip and intense fighting was reported there today. The chief of the IDF General Staff said that the focus is on targeting Hamas commanders in a very strong way. Witnesses told Reuters that hospitals were struggling to keep
up with the number of wounded coming in. The Israeli military ordered Palestinians to immediately evacuate half a dozen areas in South Gaza. National Security Council spokesman John Kirby says it's unclear when talks aimed at resuming a truce between Israel and Amas will restart. We would like that to happ in today, but honestly, I just don't know. Appearing on NBC's Meet the Press, Kirby said the US is working really hard to try to get both sides back to the table.
French authorities say one person