← The Agile Brand with Greg Kihlström®: Expert Mode Marketing Technology, AI, & CX
The Agile Brand with Greg Kihlström®: Expert Mode Marketing Technology, AI, & CX
#701: No-Code as a Strategic Weapon, with Aytekin Tank, Jotform
00:25:29
Loading summary
Transcript19 lines
- [00:00]
Greg Kilstrom
What if your customer experience and culture strategy didn't just create value but actually funded your growth? I'm inviting you to join me in a powerful mastermind group I'm co leading with Journey Spark Consulting. It's called the CX and Culture Connection, creating a self funding growth flywheel and it brings together CX and culture leaders who want to drive change, build stronger internal collaboration and actually reinvest efficiency savings into growth driving initiatives. This isn't just theory. You'll get monthly virtual sessions, one on one coaching, quarterly workshops and access to the value Accelerator tools focused on strategy, cultural alignment and voice of customer. Plus you'll connect with leaders across functions from marketing to product to ops who are facing the same challenges and pushing towards the same goals. Up to four team members can participate from your organization so you're building alignment while building momentum. Want in, learn more and sign up at journeysparkconsulting.com that's journeysparkconsulting.com mastermind let's turn your CX investment into a growth engine.
- [01:06]
Eitaken Tonk
The Agile Brand.
- [01:15]
Greg Kilstrom
Welcome to Season seven of the Agile Brand where we discuss the trends and topics marketing leaders need to know. Stay curious, stay agile and join the top enterprise brands and Martech platforms as we explore marketing technology, AI, e commerce and whatever's next for the Omnichannel customer experience. Together we'll discover what it takes to create an agile brand built for today and tomorrow and built for customers, employees and continued business growth. I'm your host Greg Kilstrom, advising Fortune 1000 brands on martech, AI and marketing operations. The Agile Brand podcast is brought to you by Tech Systems, an industry leader in full stack technology services, talent services and real world application. For more information, go to teksystems.com to make sure you always get the latest episodes, please hit subscribe on the app you listen to podcasts on and leave us a rating so others can find us as well. Now onto the show. Agentic AI has a number of potential applications, but what if it could not only save your customers time while solving their questions more easily, all the while elevating the roles of the humans on your team? To help me discuss this topic, I'd like to welcome Eitaken Tonk, Founder and CEO of Jotform and author of the book Automate youe Busywork. Aydaken welcome to the show.
- [02:35]
Eitaken Tonk
Hi Greg, thank you for having me on your show.
- [02:37]
Greg Kilstrom
Yeah, I'm looking forward to talking about all this with you. Why don't we start though with you giving a little background on Yourself and your role at Jotform.
- [02:45]
Eitaken Tonk
So yeah, I mean for me I was this developer, like a programmer two decades ago. Then I started, I bootstrapped my company and that I called it Jotform. And today Jotform is a pretty big company with over 700 employees and 30 million users. And yeah, so we have grown as a bootstrap startup and I learned a lot about business, marketing, customer service, all kinds of things. And happy to share my experiences.
- [03:24]
Greg Kilstrom
Yeah, yeah, looking forward to this. So we're going to talk about a few things today, but I wanted to start with relatively new development here. Jotform AI Agents. Certainly agentic AI, you know, top of many people's minds. If not, it feels like everybody. But I run in certain circles, so, you know, top of mind for a lot of people. Now can you talk a little bit? You know, what does agentic AI mean for a company like Jotform?
- [03:51]
Eitaken Tonk
Yeah, definitely. But let me first start with how we actually came up with this product. About two years ago we were doing this hackathon within the company and some of our people actually came up with this idea that because we are a form business and we serve over 50 million forms for our customers, what if we could actually fill out forms on the phone like over voice or by just talking to a chatbot? Because forms are boring. We all know that people would prefer having a conversation over like filling out a form. So we started working on this product and we released it as a beta to some of our users. And we discovered something. Most of the people were actually using this as a customer service product and only small number of users were actually like really using it for forms. And we thought that okay, maybe we need to pivot this as a customer support product. Customer service product. And in February we released JotForm AI agents to the world as a customer service agent. AI agent. So one of the things that we did was we started using it for our own support. So because Jotform is the large brand and we serve so many people, we already have this 200 strong support team. And most of our support is like ticket based, like emails or tickets. And we also have like phone or zoom support, but that's mostly like enterprise level, you know, customer success. But most of our business, most of our customers are like self service and they just use ticket system. So we thought like, you know, why don't we put our chatbots, why don't we offer our AI agents as a customer support option? And once we did that, like we Started getting like, you know, 2,000 conversations daily between AI and our support AI agent and our customers. And here's the thing, we actually switched 20 of our support people to become these like overseer of AI. So they became like, you know, they started reviewing AI. Yeah. And their job was instead of answering customers, they actually started reviewing AI conversations. And then between our customers and our AI and then grading them and finding the bugs and things like that. And we started with this low 25% resolution rate. AI was sometimes good, but most of the time we would cringe, just give wrong answers, hallucinations, just couldn't solve the problems, all kinds of things. And during this three months, every day, we actually reviewed every single conversations between AI agents and our customers. And like we took all those learnings from them and then we started implementing solutions. I think one of the biggest problems with our product in the beginning was, are like the Rack system, the knowledge base wasn't really being utilized correctly. So we actually made it much better and that made a big difference. And our users were also like getting all these benefits because we were improving the product. Our users who were using these AI agents were also getting the benefits of these improvements. And then the next thing was we discovered that most of the time tools were needed. The AI agent needed to do an API call to check something. If someone says hey, someone submitted my form and I didn't receive an email, then the AI agent would actually use the API tool to make a connection and just query that email address. Just make sure that email was actually found so that it can say hey, this email was actually returned to us because your inbox was full or you should have received this. Can you check your spam? Because our records show that this email was sent. So in the past our support team would do that manually. They would go to the tool and do that and the AI started doing that and then like all these things like just started making a difference. And that 25% rate gradually moved up to 75%. Right now we are over 75% resolution rate with our AI agent and we keep improving it but it's getting harder and harder to find all those like know patterns, all these different things that all the, you know, this, it's, it's not easy to find like this big, big wins, like all the low hanging fruits are already sold. So our expectation is that we are probably not going to be able to go farther than 80% but I think that's, that's good because one of the results of using AI agents for our support was for our customers, the support is very quick, like they don't have to. Our average response time before the AI was like one and a half hours. So people had to wait like one and a half hours on average to get an answer to their problem. And then if their problem is not resolved, they needed to wait again another hour. So with the AI, what's happening is that most of the people are actually solving their problem problems instantly by talking to the AI. And if they cannot solve it, they can actually connect to a live agent. Because we were able to reduce the support load on our system, our average response time was reduced to 45 minutes. And also we were able to put more people on the live chat support. So this meant that if you have a problem, you talk to the chatbot and if chatbot cannot solve the problem, it actually transfers you to the live human and then the live human solves your question and then you get much faster response. So I think this is a great example to how AI and human can come together to build a great solution. Automate many things, but also use that human touch to make sure that everything is taken care of. And so far it's going well. And with our users, so with our users, we are now reached to around like 5,000 conversations daily. Like our customers are using the chatbots on their websites, they are buying phone numbers from us so that they can actually provide phone support. You know, we provide all kinds of different ways, like WhatsApp or like messenger, different platforms to provide these AI support. And that's been growing as well. And today we are getting around 5,000 conversations on our user side and their ratings are also increasing. And since we cannot review those, we can only review our own support. We kind of rely on our kind of ratings and the AI based reviews on that support system. But we are also seeing increase on that end as well because we are improving the product and we are providing more tools. So there's a reason that we didn't call our product a chatbot or, you know, something like that. We use the name AI agents because we knew that the solution was actually like the AI is not just giving answers, just like LLMs actually use tools to be able to make decisions. And that's why we called it jotform. AI agents want to learn more and.
- [11:39]
Greg Kilstrom
Join the discussion about marketing and AI. Attend a premier conference dedicated to marketing and AI. That's Meacon, the Marketing Artificial Intelligence Conference. From October 14 through 16 in Cleveland, Ohio. Meakon brings together the brightest minds and leading voices in AI don't miss this opportunity to connect with a dynamic community of experts, visionaries and enthusiasts. The Agile brand is proud to be the lead media sponsor of this important event. Register today@marketing AI institute.com that's marketing AI institute.com and use the code AGILE150 for $150 off your registration fee. I can't wait to see you there. There's a couple of things I want to dive into a little bit more and the first is that really is my experience. Many people's experience with chatbots are that, I mean, you know, frankly, they're often dumb, for lack of a better word. You know, they can give simple responses, but they kind of end where, you know, someone hasn't already thought through the potential response or whatever. And I know, you know, there's some more like conversational AI that gets a little better and is trained, but I think, you know, the, the potential here with agents, it kind of goes back to what you were saying earlier, which is they're able to do multiple tasks, they're able to look beyond, you know, whatever database or whatever limited set of information that they're, they're able to access and they're actually able to look things up or perform actions. So I would imagine, and let me know what your thoughts are here. But like I would imagine that also speaks to the resolution rate is just the fact that they're able to do so much more and take additional, multiple additional steps means that things can actually get solved versus just maybe some answer is given or some link to some other website for more information. Is that accurate?
- [13:39]
Eitaken Tonk
Exactly, that's actually one of the socials we did. So one of the problems was our user guides and we made our AI just rely on our user guides and we discovered that many of the things that Jotform has like so many features, like it's a product that's like two decades old. Like we added like so many features and many of the things are actually our support team knows on their mind, but they are not actually visible on the user guides. So in the beginning what we did was we kind of improved our user guides. Like we kept adding more and more information to our user guides, like every time. So because our support team was reviewing every single conversation AI has and if AI doesn't give a good answer, then we would mark that and then we will improve the user guides to make sure that the next time if that question is asked, it can be answered correctly. So we started with improving the user guides, but that wasn't enough. We also Started implementing this feature so that the AI agent can start doing some reasoning as well. Like just, just like the reasoning AI LLMs agents like the models like O3, O4 mini, those models can actually do some reasoning and then they can do a web search. So we actually started latent AI to kind of go back to the old support tickets and read them. If it cannot find the answer, like, just like what a support agent would do, it would tell the user that like you know, doing some reasoning and then going to the ticket system, doing a search and trying to find a solution. And this, this also helped our users as well because we found that many users actually didn't take the time to like train their agents. But because we knew their websites, we could always go, go to their website and just find something on their website and crawl it and find the answers and sometimes just try to find the answer by using the LLMs can also do web searches. So that just using that you can just say, because we are using OpenAI and Gemini in the backend, these models can also do web searches. So we can actually say, okay, here's this website, try to find this information. So it's just like AI agent is kind of like a human. It's just going to different places and searching for stuff to learn stuff and all those knowledge, when it learns something, it can actually record that for feature usage. But also like we started also building all these additional tools. So one of the most commonly used tools is like appointment taking, right. It can integrate with your Google Calendar calendly so that you can actually like during the conversation you can take, you know, appointments. It can also fill out forms. That was actually the original idea, but that's also used a lot by our users. If someone needs to apply for something like you know, they can ask for questions, get their answers and then, and when they're ready, you know, we can just take those answers in a conversation on the phone or on audio on their computer.
- [16:59]
Greg Kilstrom
Yeah. So one other thing I want to talk about here as well is going back to what you were talking about earlier is the idea that. I know there's lots of talk about AI completely replacing humans in the workforce. I don't believe that. I don't buy into that. I think there are many things that AI does better than humans and vice versa. But you're talking about the AI being able to deliver at speed and at scale in my mind and based on your description, kind of elevates the role of the human to not just be one on one with an individual customer. But to be kind of looking meta overall and improving systems, processes, responses, all of those things overall, is that kind of the future of customer service and agentic even is this role of almost like human managers of, of AI, but again that elevation of humans in the role from less like order taker to manager, I guess.
- [18:04]
Eitaken Tonk
Jotform is a no code SaaS service product, so it's also free. So we have huge number of like small businesses and small businesses don't have the resources to actually hire people.
- [18:16]
Greg Kilstrom
Yeah.
- [18:17]
Eitaken Tonk
So the decision is not between like you know, firing someone or anything like that. It's usually like not providing that service, not providing that support 24, 7 or even like just putting a form on their website. But with AI agents now they are able to provide like 247 support. And if you have like in our case we were able to utilize our support team better. Instead of like answering the same question like I forgot my password, you know, I didn't receive my form submission, I lost my form. Instead of answering like these very simple questions that they are actually, they're able to spend more time on more difficult questions and give better answers and we are actually seeing the results of that. They are able to respond much faster and they're able to provide much better answers to our users. And with AI we are able to handle more of those easy tasks and then leave humans for, you know, more difficult things. It's also like when you're always dealing with all these like busy work, it's just like you forget to be nice to people. Right. You forget to smile or just, just you try to like rush to make sure that you give them an answer quickly. So it, it just made our human support people their job much easier. And we are able to utilize the live chat more. Like we couldn't put more people on the live chat, so. So there's always this problem with resources. You don't have enough resources to provide the kind of service you want to provide. But with AI, you can provide a great service, but also keep that human touch in the back end so that if you cannot resolve the problems quickly, humans can also take over and solve them. So I think this applies to anything I'm seeing like every part of our company. Marketing, growth, development, products. You know, wherever I look, people are using AI and they are much more productive. And instead of like doing all this busy work, instead of wasting their time doing these manual boring stuff, they're able to like give those stuff to AI and be able to spend more of their time in things that actually Makes a difference. Yeah. And the productivity skyrocketed since people started using AI more and more within our company.
- [20:45]
Greg Kilstrom
Yeah, no, I completely agree. Well, Itikin, thanks so much for joining today. One last question for you before we wrap up. I like to ask everybody here, what do you do to stay agile in your role and how do you find a way to do it consistently?
- [21:01]
Eitaken Tonk
I became this really this person that's constantly using LLMs AI to do deep research. Like just even like in the past, like when I had like some free time, I would actually, you know, launch my Kindle and start reading some book. Today I'm usually like, you know, I'm talking to a reasoning model or talking to a deep research and like just asking for something and then while I'm waiting for those deep research to end and I'm like just reading those stuff and I'm triggering this stuff. So I feel like I became much more productive because in the past, like when I needed to research about something, like about the market, for example, about different competitive landscape, about different options for solving a problem, in the past I had to do research, I had to go to G2 to read all these reviews. But today I can just ask AI to give me the answers and just find the right products, use the right products. And this is so much more fun because you get it, you get those answers quickly. You become more agile because you can make decisions much faster. Like whenever I need to make a decision, I can go to ChatGPT and start having a conversation with it. Even in the car, I can just start Talking on voice, ChatGPT voice mode and having this like, you know, ask my questions, get my answers, like think about all these different options and just become more agile that way. So I feel like this, you know, we live in a great times.
- [22:41]
Greg Kilstrom
Yeah. Love it. Well, again, thanks to Itakin Tonk, founder and CEO of JotForm for joining the show. You can learn more about it again and Jotform by following the links in the show notes. Thanks again for listening to the Agile brand brought to you by Tech Systems. If you enjoyed the show, please take a minute to subscribe and leave us a rating so that others can find the show as well. You can access more episodes of the show@theagile brand.com that's theagile brand.com and contact me. If you're interested in consulting or advisory services or are looking for a speaker for your next event, go to www.greggkilstrom.com that's G R E G K I H L S T R O M.com the Agile brand is produced by Missing Link, a Latina owned, strategy driven, creatively fueled production co op. From ideation to creation, they craft human connections through intelligent, engaging and informative content. Until next time, stay curious and stay agile.
- [23:48]
Eitaken Tonk
The Agile Brand.
- [23:53]
Greg Kilstrom
Before we continue, I wanted to share a key strategic resource that a majority of the Fortune 500 are already aware of. Finding the best technology, business and talent solutions is not easy. With business demands and competitive pressures mounting, you need to be able to design, deploy and optimize your technology to provide leading customer experience experiences while driving business growth. Those of you that have been listening to this show for a while know that this podcast is brought to you by TechSystems, a global provider of technology, business and talent solutions for more than 80% of the Fortune 500. TechSystems accelerates business transformation for their customers. Whether you're looking to maximize your technology roi, drive business growth, or elevate customer experiences, tech systems enables enterprises to capitalize on change. Learn more@techsystems.com that's teksystems.com now let's get back to the show.