Dialpad Brings Enhanced AI to the Contact Center

welcome to zcast everyone I'm Zeus caraval from ZK research and I'm joined today by Jan O'Connell the chief strategy officer at dial pad download you say hi to everybody yeah great to be here thanks for having me before we get started I just want to give a quick shout out to you my media partner all zcast are done in conjunction with you he speaks program but uh uh before we get into our topic of artificial intelligence in the contact center uh give us a brief rundown on yourself and dial pad what you guys do and what you do there yes I joined dial pads some of our chief strategy officers you mentioned I joined uh four and a half years ago had acquired uh the startup that I was building and CEO of called Talk IQ which is a real-time speech wreck in NLP startup which is a fancy way of saying if anybody uses Syria and Alexa we were doing leveraging similar Technologies to go do things like real-time transcription uh suggested answers to questions map sentiment and then for those of you that might not be familiar with dialpod we're an AI power communications platform we do voice video messaging and contact center on any device anywhere in the world uh our Founders came out of Google and are heavily invested in how do we provide just a really unified beautiful experience as I said across voice video messaging and contact center yeah all those uh critical tools today in the cloud that make hybrid work work it is it is yeah so uh I know you talked to a lot of customers as part of your job and uh what are some of the trends that you're seeing that's driving uh contact center transformation today yeah so I think like a a number of Trends and I think some of them have obviously been accelerated through the pandemic and obviously pretty much all of us now working um in a hybrid way uh one is just a unified platform um I think that's one of the trends that we've seen at least from from our customers has been look like people want to have a single piece of software uh they want to be able to go and manage that in the cloud and provision new users in a really seamless way and so we've spent a lot of energy and a lot of time of how do we go and unify voice video messaging and content as I said contact accelerate into this one seamless application and I think obviously the next one gets into like the contact center opportunity which I think is obviously you know I call it like a legacy space but I think there's been these these large Legacy providers that have been around and suddenly every business found themselves figuring out how to support and both support users or customers but then also sell uh and how do you actually do that suddenly when you can't show up in the office uh and so I think there's been a heavy investment in contact Center Software the shift to the cloud and then I think like the last Trend obviously gets into AI which is you've got these Technologies which are not new technologies you know speech recognition has been around for decades but suddenly has these really nice applications uh and obviously we've got the benefits of like cloud computing and speed and internet access to suddenly be able to do things like real-time transcription route calls based on what is being said or infer customer satisfaction and that just opens opens up a host of new opportunities for businesses out there yeah what I've seen just this big ship to customer experience-led organizations has has driven a lot of innovation in contact center my research shows that uh you know customer experiences don't talk brand differentiator one of the interesting data points of my research last year was two-thirds of Millennials admitted in changing loyalties to a brand because of one one bad experience right so you know there's no Mulligans given here right when when it comes to business so yeah um now one of the big trends in both uks and ccas and I know the combined offering is what you call crew cast but uh has been the use of artificial intelligence and what are some of the things customers are coming to you looking to do with AI yeah so I think there's like the you know to your point of like people want to have these really great experiences uh with Brands you know a lot of that now seems to start to try to or at least starts with look people want to start online to get their answer so it starts around self-service capabilities of hey how do I go and interact with a virtual agent so something that people come to us is around virtual agents for hey I want to Simply go and get a really quick answer and uh and solve that myself without actually helping on a phone and talking to an agent so that's something we've recently released and are thrilled about um other things get into things like uh what we call AIC sat which is I recently announced this just a couple of months ago but it's really about look you and I have a conversation if I'm an agent in a contact center I shouldn't send you a survey at the end of the call or send you an email saying easiest uh you know how did you feel about the interaction like what happened we should be able to we've got technologies that can go in for that so I can just do my job pick up the phone have a conversation and we'll infer whether you're thrilled or incredibly frustrated with that and so you get 100 coverage and uh obviously you kind of democrat I talk about it in a way that it democratizes kind of the data that's there uh but then obviously you get into the things like agent assists or suggested answers to questions but I think the Unique Piece for for dial pad is we've heavily invested and look building our own Ai and what I mean by that is we build our models in-house we do all of our own speech Rec we do all of our own NLP and so that means that it's domain specific so we think it drives a lot more accuracy and a lot more use because it's really thought about of look how do we go and build AI features that are specific for contact center use cases or specific to as I said like conferencing use cases that are out there yeah I think um uh you know AI has actually made some of the things that the promise of some of the things we enroll on contact center work I know early on when people rolled out Bots and things you want to talk about frustration I mean a a bot without AI is nothing more than a text-based ivr yep right and people hate ivrs and they hate textbooks I have yours anymore yeah right so and the introduction of conversational AI actually makes it usable so um so uh I use the term truecas uh can you um just Define what that is from the dial pad point of view and then talk a little bit about the long history you've had in AI I know it has a I don't want to say it's a it's not a it's a new topic for this industry but it's actually been around a long time and but you guys have actually been doing AI under the covers for a lot longer than most of the industry correct yeah so true cast for us as I said if you keep hearing me kind of do this tagline and click it's voice video messaging contact center on any device anywhere in the world but it's really that it's look you shouldn't have three different or four different pieces of software to manage your communication and collaboration needs so it's like we're going to do all of that in one seamlessly integrated application and then we spend a lot of time you know you see our tagline also work beautifully which is look I'm a big believer in our and and a lot of us here at dialpad are huge Believers that look you want software that looks and looks uh and what I mean by that is you can probably build the best application in the world but if it doesn't look good people are not going to have this great wonderful experience with it so we care a lot about the bringing kind of a consumer design to Enterprise software um and then when you when you talk about AI as I said I mentioned I've been here at dial pipe for four and a half years prior to that I was building a startup that had to do with speech Rec and MLP and so you know we view ourselves as really being innovators in this market uh talk IQ we built really the first real-time speech recognition engine which was we want to go and do a live transcription for Enterprise conversations meaning these are not conversations like you have with Siri or Alexa or or Google which is you know you say Hey Siri you know uh play The Beatles or hey Alexa what time is it um these are 45 minute or longer conversations we're talking about words that don't show up in the dictionary product names company names and we could go and understand that with Incredible accuracy and that opened up all of these different opportunities about look if you can go and understand a conversation as it happens what does that open the door for and it opens the door for things like inferring customer satisfaction in real time it opens the door for suggested answers to questions it means that you can go and Route calls it means you can go and understand when there's a network issue it means we could go flag spam that's happening so all of these different opportunities have been opened up for us because we do this ourselves that have been so heavily invested in it and then you know to I know I go on these long Rambles but the last one I would highlight is you know we do our own speech wreck we do our own NLP and then last year we acquired a business called creative knowledgeware that uh is really an innovator and a Pioneer in semantic search and so again those are really the three areas of AI that we're heavily invested in speech Rec and LP and then semantic search yeah I want to drill down the point too if you do it in your own because if you look across the ukfccas industry there are a lot of pure play um AI specialist companies in their conversational AI ones that do search and almost every one of your competitors partners with those companies uh in order to bring AI capabilities and so you know why are you doing your own yeah and and what what does that give you for competitive advantage and how do you translate that into customer facing benefits yeah so great question a question I got literally every day from investors and and to this day still get uh pretty much on a daily basis so there's really there's a couple different advantages right that we View and I think are important in in in doing this yourselves number one is like I'm a big believer in being able to controller and Destiny and so I think at times if you're leveraging third parties if you're leveraging other apis that are out there your Lego you're just buying the Lego box and so you're buying the Lego blocks and that suddenly everyone else's application looks the same because perhaps your competitors are all using the same partner undergo and power things so one is I think it provides this Innovation and differentiation that nobody else can replicate and I think that's really incredibly important especially in a market that look our competitors are some big names people are aware aware of them and so we have to find ways to actually stand out and I think that gives us a key way to stand out number two gets into gets into cost so those are obviously things like our CFO and the finance folks we'll care about is that look doing this ourselves at this scale now um gives us an advantage in terms of cost now obviously the pricing for even things like like uh speech recognition or transcription or are are different for the vendors that provided as opposed to the public pricing they charge a customer but I would say look like there's there's significant cost advantages for us to be able to go and do this and then the third which actually might be the most important thing gets into accuracy which is look we don't want to use an off the let's take sentiment for example when we're talking about sentiments specifically for contact centers we need to go and build models those sentiment conversations are customer satisfaction conversations look different for a bank than they do in healthcare and somebody in healthcare might be talking about pain and frustration because of the medical condition that they have not because of the experience that they're having with an agent or a brand or a company and so these are all of the things that allow us to drive differentiation is that accuracy stems from we can go and build these models and train these data specifically to these domains and industries and that again provides an advantage to the customer that uses us because look it's not some generic model that's out there it's it's purely tailored so those are you know ultimately the three things of like why do we do this ourselves is look speed accuracy costs and then it provides us a way to actually go and differentiate fortunately we made that bet as I said you know for dialpad made that bet four and a half years ago we've been working on it prior to that so we view this as like look we've got a five-year Head Start of of all of our all of of our competitors and we need to continue to heavily invest in r d on this side to to continue to drive Innovation and differentiation yeah and one of the actually the AIC set feature that you talked about before is one of my favorite ones that you rolled up because even the best organizations would get maybe maybe a 10 hit rate on the email I mean I mean they 10 is I guess Nirvana right yeah I would say probably high they got two two or three percent yeah well your product you can actually do that for every call now yes yeah and and I think those are the you know these are the things where it's like look and you don't have to the the best part of this stuff is you don't even have to change behaviors right you're already gonna pick up the call every day and talk to people uh right if you're if you're an agent in a contact center instead of the the beauty of it is like look you just do what you normally do and we're gonna go give you this data point that one you've never had that volume of data um and so suddenly you're not making decisions around customer satisfaction on what I call like the heroes and zeros which is like you've got a really great experience and want to celebrate Dan because he was helpful or you're like Dan was terrible and I want to tell the company how bad that experience was um and so it becomes this really polarizing data point yeah and what there's another feature you just announced uh you touched on it a little bit but as part of your your larger Asian empowerment launch uh UNL something called AI agent assist can you talk about what that is yeah so aih an assist is the ability for uh you can think of it as a virtual assistant for for our agents that are sitting in a contact center so at any point in time they can go again I I talked about kind of the power of semantic search and and the investments in AI that we've made but it's the ability for an agent to literally go and query unstructured data points to get an answer instantaneously for them and when I say answer this is different than the agent that's got to put you on hold spend 10 minutes looking through 10 different systems right it is so this is the beauty of all of this stuff of this these Technologies right they all come together in real time and we talked like no latency so you're talking milliseconds of latency but suddenly these are the things that we want to go and Empower agents with which is like you should never have to put somebody on hold and that's going to drive faster resolution times we think it improves customer satisfaction for the the customer that they're engaging with and so these are the the kind of the touch points and kind of the beauty of the platform is it's like hey it's it's solving these issues the Asians have faster and that has a positive impact on the customer and we think that's why you know we obviously want more and more people to come and use and leverage style time yeah this is uh and that's an interesting tool because whenever I talk to agents about AI a lot of them are going to take their jobs and it's like no no no you're looking at this wrong it is it's um you know I use the analogy it's like a an autonomous vehicle like down the road years from now sure we might have cars with no drivers in them and one day years down the road we might have contacts and there's no agents in them but today ai's been used to make people better drivers parallel park assists Lane change alert and ai's been used to making agents much better agents by removing a lot of that heavy lifting and so now you don't have to worry about um you know customers screaming at you because they're unhappy and think about it from the company standpoint right you're going to go out there and find our new talent and the agent insurance is always high if you give those eight if you put those agents in a position to be successful your ability to retain them is much higher than those Legacy models correct yep exactly and I think you bring up a point which is like looks like it's it's not just about how do you provide a better customer experience there's this Asian experience right where um even today you look like people can leave their jobs pretty quickly and move to another one because of just the demand in the labor market but these are things that matter it's like you want to go and empower the agents to make sure they have a really wonderful good experience doing the work that they have to do every day which is which is hardwood you know working in a contact center is hard work trust me yeah yeah I wouldn't want to do that now I know David has a group within it called AI labs uh and you use that to drive AI Innovation I I guess across the whole product line but can you talk about that group and some of the things that it's come up with yeah so this stemmed out from um from you know a lot of us here are our Founders and myself included came from came from Google and so Google has had this um Gmail Labs is like a really pertinent example of it and it's a way that you can kind of opt into things that people are just experimenting with whether it's research projects uh some features that may never actually make it to GA and so we took those ethos and really brought it specifically to to our Ai and product and Engineering teams which is like we want to go and talk about the Innovative features that we're working on we want to provide we want to share insights around the team we want to share insights Around The Cutting Edge research that we're doing whether it's around next Generation models for speech recognition or things that we're working on on the NLP and nlu side that have to do with with churn prediction or revenue or kind of purchase intent um and so these are areas that look we want more customers to engage with us we want to work with them and help them well provide them with a unique feature but also get their feedback as early on in the process as possible the other piece that opens the door for is really is really around data sharing so anytime you're talking about building models for nlpnl you you have to go to it you in order to drive an accurate model you got to have a vast amount so it also opens up this door that we can go and engage with a customer to say hey we're building this model specific to perhaps Healthcare or or the medical field and we'd love to partner with them to go and and see if we can leverage some of the data and opt into it to build a specific feature for them to see if it opens up Avenues and opportunities for other things so it's really about just us right continuing to drive Innovation sharing more about what we're working on and again another touch point to go and engage with customers that like to test out things right and as I said some of that stuff will never actually get to GA right and that's the intention of it is to kind of push the boundaries of things yeah um and that's uh one of the interesting things about that is you're not taking advantage of the fact that Communications in the cloud you couldn't do that with a legacy on-prem system right and the reason Google can do that with Gmail is because they are cloud-based and then if things stick they stick and if they don't no harm no foul nice PLM back or put that we you know we put a lot of things but what we call behind experiments which is you know just for a sub small subset of users gather this feedback uh all of those touch points again are unique to kind of cloud software it's funny like growing up in the like growing up in the club like I never think about things about prep like it's just a foreign kind of notion to me of like well why would you you know I smile laugh and I'm like well why would you build software that way there's obviously like I understand there's some some moments where that makes sense uh but I think for the vast vast vast majority of software you're like look it's got to be Cloud native like if it opens up so many opportunities and seamless experiences yeah yeah and so uh one last question uh given all the AI stuff you're working on you mentioned a whole bunch of things uh without getting into any kind of ndas can you share any kind of road map of what customers can expect in the in the near future yeah so so we did have this this uh this nice announcement uh just the other day around Asian empowerment and something I'm really really really excited about which I think is just a huge opportunity gets into what we call aiqa scorecard so obviously every business I think that manages um you know a contact lender of a contact center of any scale is uh doing playbooks and scorecards if they're agents and we think there's a massive opportunity to not automate that process but hope go and do some directional scoring to really go and save time for the managers and the supervisors that are grading calls but also for the agents to get much faster feedback and drive accountability uh and performance uh and so again there's a massive operation well we think it's a massive opportunity there are Standalone businesses that just do scorecarding and we think that look that's a feature that should be that that sits in our our wheelhouse of AI superpowers that we think opens the door for for your and just another opportunity for us so I'm thrilled about that we'll get into things like churn and revenue prediction as I said we've recently announced AIC SAT but we think we can do some some inference on the churn and revenue side which I think are going to be game changing and really useful for look every business in the world is figuring out how to sell more and sure and less so uh it's a great data point to go and solve yeah that actually addresses a an issue that's come up when I talked to contact center managers is to look look now that we're moving people home we need better tools to manage because the old way of managing was well you work in the context industry is a long time you literally walked around people and you listen for things yes they're right right and even even if even if that was you could do that you can only hit a subset of calls and like yes which can listen to every call the score card can listen right and so now you're getting a complete set of data so and you see and you see it you know I talk about you know the rise of conversational intelligence and and look like I I think it's it's nice to have to poke fun at people and like it's all it's all it's it's all fun and games but like um but a lot of these businesses are tape recorders I'm right which is hey I can go have a recording I can go do a transcript and like here's the tool for you to go Data Mining and figure out what to listen to and you're like look reviewing game tape is super useful nobody's gonna say that but it still takes time and you can still like before you could never even do that you had to you had to do the sidecar and literally sit down next to the agent to do things um so I think like there's been like the the first generation of like nice uh kind of better tape recorders and we think in real time and and again leveraging these really unique NLP and nlu models we can go and all it's kind of automate Aid and and and uh just improve the overall process and experience for both managers and agents well uh thank you very much uh Dan McConnell from uh dial pad uh it's nice to hear that the AI is actually uh making contact center interesting well before I like you know honestly you've been with a period of 20 30 years where there wasn't a whole lot of exciting things coming in this place yeah absolutely yeah so it's nice to see some Innovation coming into this so uh again on behalf of Dan I'm Zeus caravalla don't forget to click the Subscribe and I will see you next time on zcast foreign

This transcript was generated automatically from the video's captions and may contain errors.