#280 Aytekin Tank: How to Fully Automate Your Customer Service with AI Agents

This episode is brought to you by Extreme Networks, the company radically improving customer experiences with AI-powered automation for networking. Extreme is driving the convergence of AI, networking, and security to transform the way businesses connect and protect their networks, delivering faster performance, stronger security, and a seamless user experience. Visit https://www.extremenetworks.com/ to learn more. JotForm began nearly twenty years ago as a simple online form builder. Today, it has grown into a platform powering millions of users and has recently stepped into a new era of AI. In this episode, Aytekin Tank, founder and CEO of JotForm, shares the story of how the company pivoted from forms to launching AI Agents that are changing the way small and medium-sized businesses handle customer support and automation. Aytekin explains how JotForm's AI journey started as an internal experiment to fill out forms by voice, only to discover that most users wanted it for customer service. That insight sparked the creation of JotForm AI Agents, tools that don't just answer questions, but perform tasks, scale support teams, and transform the customer experience. Beyond the technology, Aytekin outlines a bold vision of the future where every business, no matter the size, will have its own AI-powered agent available around the clock to schedule appointments, answer questions, and even process transactions. This conversation explores how AI is reshaping customer service and why small businesses, nonprofits, and educators stand to benefit the most from this wave of innovation. Stay Updated: Craig Smith on X: https://x.com/craigss Eye on A. I. on X: https://x.com/EyeOn_AI (00:00) The Scale of JotForm Today and the AI Vision (01:32) Aytekin's Journey: From Developer to Bootstrap Founder (03:36) The Pivot: From Voice Forms to AI Customer Service (05:25) Building and Launching JotForm AI Agents (06:20) Using AI Internally: Cutting Tickets and Boosting Resolution Rates (08:13) Why AI Agents Go Beyond Chatbots (12:22) Demonstrating JotForm AI in Action (21:40) The Tech Stack Behind JotForm's AI Agents (24:59) Training and Teaching Your Own AI Agent (33:27) Competition, Differentiation, and Market Strategy (36:55) Why Small Businesses Need AI the Most (38:24) The Future: Every Business With Its Own AI Agent (40:19) Pricing, Growth, and Lessons in Scaling AI (42:26) Slow Growth vs Explosive Scale and Lessons Learned (43:55) Aytekin's Vision for AI in Business

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  Jotform is two decades old. We have 30 million users. We are hosting 50 million forms for people. So we are now this large company and we have this support team of 200 employees. And we also have like a large support load, like we receive over 2,000 tickets every day. So we decided to use our own product. And when we actually start using our own product, the resolution rate with AI was around 30%. The vision is having every organization, every small business have their own AI agents. So that when you want to talk to that business in the middle of the night, like 3 a.m., you can talk to that business, you can get information, buy stuff and, you know, set up appointments,

• 0:43Speaker 00

  whatever you need to do, like, you know, all the tools that you could make available to your AI agent could be done by the AI agent.

• 0:51Craig S. Smith?

  Our sponsor for this episode is Extreme Networks, the company radically improving customer experiences with AI-powered automation for networking. Extreme is driving the convergence of AI, networking and security to transform the way businesses connect and protect their networks to deliver faster performance, stronger security and a seamless user experience. Visit extreme networks dot com to learn more. That's extreme networks all run together dot com to learn more.

• 1:32Speaker 00

  So I started Jotform two decades ago. And the reason I started Jotform was before Jotform, I was working for a media company and we had like, like over 100 websites and like, and I was a developer and I was constantly like creating these online forms for our editors. And I, I was like, you know, creating all these different kinds of forms, like contact forms, surveys, questionnaires, payment forms. And then I decided that like, you know, when I start my company, forms could be the uh stuff I could do because I I couldn't find a product that could actually take away from me like just uh I had to like do it manually.

• 2:15Speaker 00

  And um and then I started just for my created this uh easy to use form builder two decades ago and um and we you know we we have been a bootstrap company uh since then, uh still we are bootstrap no outside investors. And um and we have been steadily growing over the last two decades. And um and how we, you know, got into AI, um I can talk a bit about that.

• 2:44Craig S. Smith?

  Yeah.

• 2:45Speaker 00

  Couple of years ago, like uh one of the things we do at just for is like we we have these um technical hack fix um during this hack fix like our, you know, product teams, they try to come up with uh different ideas, different product ideas, prototypes, or like different technologies. And um some of our teams like uh like two years ago, uh some of our teams actually built this uh this idea that like you can actually fill out forms by voice, right? Uh and just because like, you know, the you can use AI to understand and ask ask questions and and then get the answers and just uh even during like this hectic, they were able to actually implement prototypes.

• 3:31Speaker 00

  So we saw that this was possible, so we started working on this product. Uh but what happened was um whi- while we were building this product, like, you know, uh filling out forms with AI, uh with AI agents, that uh we discovered that like uh most of our beta users were actually using it for different purpose.

• 3:53Craig S. Smith?

  Hmm.

• 3:53Speaker 00

  They were actually using it for customer service. They were mostly like they were putting it on their website as like chatbots. Uh because they discovered that um you know, in addition to like form uh answering features, we all said that these like knowledge based features.

• 4:10Craig S. Smith?

  Mm hmm.

• 4:10Speaker 00

  Like rag and just, you know, uploading all this knowledge or giving your website URL so that, you know, it can learn from it. And they discovered that it's just like it can really give good answers, provide like customer service, answer their questions. And like just, you know, one of our beta users, they were doing this like huge, um, you know, uh, event like a festival. And like they were getting all kinds of questions and all those questions, the answers for those questions were actually available on the website. But people were actually asking to chat but because people don't read and they, you know, they're busy, they're, they don't wanna read stuff, they just wanna get answers quickly. And our, our agent was able to answer questions like, you know, what's the price of this concert, like,

• 4:54Speaker 00

  where can I park my car? You know, where can I put my tanks, things like that, like just all kinds of wild questions and he was answering them really well and, and because this was a festival in a short amount of time, they, they were receiving huge number of questions and like, in a normal settings, like even though this is like a, you know, large festivals, like they don't have a lot of employees. So if you think about it, like it can scale, the AI can scale, like you don't have to wait on the line for a long time. So they were able to get answers to it. So, once we actually looked at what's happening, uh, we pivoted our product and we started calling it, uh, a customer service product.

• 5:34Craig S. Smith?

  Hmm.

• 5:34Speaker 00

  And we, we are calling it Jotform AI Agents. Uh, we released it, uh, two months ago and, um, you know, people are using and really getting, um, you know, their questions answered. And, uh, actually, the biggest user in our platform is, is Jotform, like, so Jotform is like now two decades old, we have thirty million, uh, users, like we are hosting like, you know, fifty million forms for people. So we are now this large company and we have this support team of two hundred employees. And, um, so we, we also have like a large support load, like we receive, you know, over two thousand tickets every day, like support tickets every day.

• 6:19Speaker 00

  So we decided to use our own product and, um, And when we actually start using our own product, the resolution rate with AI was around thirty percent. So we are actually comparing the answers of our human team with the AI team. And we started thirty percent like we could. And basically, we actually, you know, got like twenty of our support employees started actually reviewing all these AI answers. And and then they they started kind of like, you know, flagging them result or not result. So we we decide if it's resolved by comparing to how a support employee would answer that question.

• 7:03Speaker 00

  And because of that, we are pretty sure that, you know, we are really measuring it correctly. And then in the early days, like essentially, we launched this like two months ago as well, like during the launch of Chatform AI agents. And during this two months, we started with thirty percent. And now we actually reached seventy percent resolution rate. And the the AI agent can actually answer those questions. And the way we were able to do that was basically like we had to rewrite our, you know, RAG embedding system or, you know, the database. Because we saw that it wasn't handling it right, like we moved to a product called Gina for our RAG embedding.

• 7:44Speaker 00

  And we basically, our product teams actually, you know, picked different topics, like, you know, someone is trying to change their email address. Someone is trying to, you know, create a form by asking to AI, you know, all different kind of questions. So basically, they implemented these solutions and then we connected them with with API to the AI agents. And basically, what makes this powerful is AI, what's what makes AI agents and like regular chat, L L M chatbots is what makes them different is.

• 8:17Speaker 00

  AI agents can actually do things as well. They they are not, they don't just answer questions, they can do things. And in just from AI, for example, you know, there are lots of tools that the AI agents can use. Like, you know, they can actually do appointments. They can connect to your Google Calendar, Calendly, to be able to, you know, take appointments, you know, by voice or on a chat bot. So they can make make calls for an API to be able to answer questions. In the case of like changing an email address. So they are actually, you know, they are making an API call to authenticate the user, to be able to make sure that this is the real user that the AI agent is talking to.

• 9:01Speaker 00

  And also like taking their new email address, maybe if there's a verification email sent to the old email address, then they have to enter some numbers as well. But this is all done conversationally and the AI agent can actually use these tools. So because because of all these tools, we were able to go from thirty percent to seventy percent. I would say half of the improvements came from improving our RAG system, improving our knowledge base. Like, we had to write tons of new user guides because basically the our AI agent is using our knowledge, our user guides. And we discovered that our user guide was missing like so much. And then when, once we added all these missing pieces,

• 9:44Speaker 00

  the AI agent was able to answer much better. But half of that came from that knowledge. But half of some actually came from like implementing all those tools. The AI agent uses to be able to like to resolve the problems of the user. And so that's really good, like get into seventy percent. You know, it wasn't easy, but it was possible. So we envisioned that, you know, we are going to, we are going to be able to go to eighty percent as well. Like, we can see that we will be, but after eighty percent, it's just going to get so much harder and harder. But it's okay, because it's always like. So one of the things that the AI agents can do is they can actually like, you know, transfer the the chat to an human agent.

• 10:29Speaker 00

  And if we can handle eighty percent and if we have to transfer twenty percent, you know, we are fine with that, that's still better. And and another thing we discovered was like, so we implemented this system separate from our support ticket system.

• 10:46Speaker 00

  And our support ticket system was all, we were getting like over two thousand tickets every day. And that actually went down to like, went down like twenty five percent to fifteen hundred a day. And it's even going further down now. Um, but it didn't like, you know, it didn't completely replace that. So people, you know, what we discovered was some people, you know, you know, because we are also showing our average response time on our support system that says like, you know, our average response time was one and a half hours when we started this, but because now the AI agents are helping, uh, our actual average response time went down to thirty minutes now, like just uh, much faster. But even though it's thirty minutes, like many people, they don't wanna wait thirty minutes for an answer.

• 11:31Speaker 00

  They either wanna get like an instant answer, or they don't wanna just bother with support at all. So, so that's, that's another thing, like, we are actually supporting more people now. Uh, on average, like we are supporting every day. In the past, we were responding to like four, two thousand um, uh, queries, two thousand support tickets. But now we are having like uh, thirty five hundred uh, you know, the uh, support, you know, uh, threads. And, and you know, two thou- about two thousand is actually coming from the AI and then uh, fifteen hundred is uh, on, on our human support. So we actually, you know, providing more support to people.

• 12:11Speaker 00

  Uh, but uh, you know, we, we are able to provide a better support thanks to the AI agents.

• 12:17Craig S. Smith?

  Yeah.

• 12:17Speaker 00

  And it's just so much fun to be able to use our own product. Uh, I haven't said it.

• 12:21Craig S. Smith?

  Yeah, yeah. Uh, let me go back to the original AI application was in uh, using voice to fill out forms?

• 12:32Speaker 00

  Yeah, that's actually how we started this project two years ago. You know, uh, we talked that, hey, um, you know, forms are boring. Uh, maybe you wanna be able to fill out a form on the phone. Uh, or over voice on a website. Or maybe you just wanna talk to a chatbot as opposed to like just filling out this boring form. And that's how we started. And then when we released that, uh, only like ten percent of the beta users were actually using it for that. And ninety percent of our beta users were actually using it for completely different purpose, like we never thought that, you know, that, that would be the main use case uh, for this product. Yeah.

• 13:13Craig S. Smith?

  But how does that work? I mean, does the AI ask like, what's your first name? What's your last name? Does it? Does it track the form and ask questions? Or are you looking at the form as the user and telling it? First name Craig, last name Smith, and it finds the box? I mean, how much instruction do you need to give the AI?

• 13:48Speaker 00

  Here, this is the landing page, and I can create an agent here.

• 13:55Speaker 00

  And I will create this voice agent.

• 13:59Speaker 00

  And then I want to create, let's say, I want to create an appointment form, appointment agent for a personal trainer, personal trainer.

• 14:17Speaker 00

  And it's going to create the agent in the backend. Basically, it's doing some research to be able to create a good one so that you don't have to like play with it a lot to be able to have something useful.

• 14:40Speaker 00

  So now we can see that I have this Alex appointment scheduled. So this is the AI agent builder. This is kind of a place where I can actually play with settings like, you know, the voices.

• 14:52Speaker 02

  Jotform AI agents are powerful tools that handle.

• 14:55Speaker 00

  Jotform AI. Yep. Okay, good. So I can, I can pick a voice here.

• 15:02Speaker 02

  Jotform AI agents are powerful tools that handle. Jotform AI agents are powerful tools that handle tasks like answering questions.

• 15:11Speaker 00

  You know, I can play with it, but so I can actually train my knowledge base, add information here. And it actually also created a form, probably created a form for me as well. So here we are, we can actually see the form here, and I can take a look at the form. I'll have to share this tab. Um, so it actually created this form because I said personal training appointment form, uh, appointment agent, it created this form on just forms form builder. This is just from form builder we are on now. And it created this form and I can take a look at and you know, I could change things on this, but I, I feel like this is pretty good.

• 15:52Craig S. Smith?

  Yep.

• 15:52Speaker 00

  So I will keep this form. So now, um, I'm going back to the age agent builder. And, um, just, you know, I'm ready to test it actually.

• 16:04Craig S. Smith?

  Yep.

• 16:04Speaker 00

  Uh, we could just test it here or I could just open it by itself. Um, here, okay. Let's try it.

• 16:16Speaker 02

  Hello, I am Alex, your AI agent for scheduling personal training sessions. How may I assist you today?

• 16:23Speaker 00

  Hi Alex, I would like to set up an appointment.

• 16:30Speaker 02

  That sounds like a great plan. I'm here with you and ready to help get everything set up smoothly. To proceed with scheduling your appointment, may I have your full name, please?

• 16:39Speaker 00

  Aytekin Tank.

• 16:43Speaker 02

  Thanks for sharing your name, Aytekin. We're making good progress here, just a bit more to go. To continue, could you please share your email address by spelling it out letter by letter to ensure accuracy?

• 16:54Speaker 00

  A Y T E K I N at J O T F O R M dot C O M.

• 17:07Speaker 02

  Got it. Thanks for sharing your email so quickly, Aytekin. We're almost there, just a little more to finalize your appointment. To make sure I have it right, could you please spell out your email address letter by letter? For example, is it a E E T E K I N at J zero T F zero R M dot com?

• 17:27Speaker 00

  Uh, it's A Y T E K I N at Jotform.com.

• 17:38Speaker 02

  Thanks for confirming that so clearly, Aytekin. We're getting everything lined up nicely for you.

• 17:47Speaker 02

  To ensure I have it right, could you please spell out your email address for me, letter by letter?

• 17:53Speaker 00

  A Y T E K I N at J O T F O R M dot com.

• 18:02Speaker 02

  Dot form dot com, is that correct?

• 18:05Speaker 00

  Yes.

• 18:11Speaker 02

  Got it, Aytekin? Thanks for sharing that detail so clearly. Everything's shaping up well for your appointment. To proceed, could you please share your phone number?

• 18:20Speaker 00

  It's four one five, five five five, five five five five.

• 18:32Speaker 02

  Thanks for sharing your phone number so smoothly, Aytekin. We're getting everything lined up nicely for you. To confirm, is your phone number four fifteen, fee fee fee five five five five?

• 18:43Speaker 00

  Yes.

• 18:47Speaker 02

  Perfect, thanks for confirming that for us, Aytekin. We're almost ready to move forward with your appointment. To proceed, could you please let me know a convenient date and time for your appointment?

• 18:57Speaker 00

  Tomorrow, nine AM.

• 19:02Speaker 02

  Got it, tomorrow at nine AM sounds like a great time. We're getting everything set up just right for you. To confirm, the appointment is set for April twenty-third at nine o'clock in the morning, Eastern daylight time. Do you confirm this?

• 19:16Speaker 00

  Correct.

• 19:19Speaker 02

  And, your options are: yes, I confirm.

• 19:23Speaker 00

  Yes, I confirm.

• 19:39Speaker 00

  And yeah, so on, I think it's

• 19:43Craig S. Smith?

  Yeah.

• 19:43Speaker 00

  My gut. Yeah, sometimes it gets stacks.

• 19:49Speaker 00

  But let me, let me also show you something else.

• 19:56Craig S. Smith?

  Did it fill in the form then?

• 19:59Speaker 00

  Um, take a look, let's take a look. Yes, it did. Let's take a look at the conversation here. Um, so we can see here, it actually got my name correctly, which was actually hard, like we had to kind of train my name on it.

• 20:18Craig S. Smith?

  Yeah.

• 20:18Speaker 00

  Just because it's nothing, uh, American name. But, uh, you know, it got my name, uh, my email address, uh, phone number and the appointment correctly. And, um, let's see, and the conversation was actually here.

• 20:35Craig S. Smith?

  Right.

• 20:37Speaker 00

  Yeah. And, uh, so one of the things that, like, I was talking about the tools. So basically, what I could do is like, if I'm a personal trainer and if I'm using, uh, like, Google Calendar to keep track of my, um, you know, uh, my appointments, I could actually connect to this and I'm all, I'm ready, I'm ready to use this product. Uh, right away and just, you know, I could use, um, you know, one of the options, um, you know, like, I could get a phone number, buy a phone number, uh, like it's just $10 a month. I could, uh, if I have a website, I could put a chatbot on my website.

• 21:21Speaker 00

  Um, you know, I could use text, I, you know, just get a texting number or like a WhatsApp, connect to my WhatsApp account, um, and, um, be, be able to, you know, take appointments.

• 21:35Craig S. Smith?

  Is this, uh, what models are you building this on?

• 21:41Speaker 00

  Yeah, I mean, um, so that's, that's the great thing about, uh, building, building products on top of AI.

• 21:48Craig S. Smith?

  Yeah.

• 21:48Speaker 00

  Because, um, like usually like, uh, and I've been building a product for like, you know, two decades.

• 21:56Speaker 00

  And there's, there's this, there's this thing called software entropy. And it's just like, you know, software actually gets worse over time, like just stuff stops working like, you know, you get stuff broken, like just out of blue for some reason, like it shouldn't happen, right? But with AI, it's actually like the, the grant below you is actually getting better. Um, so like we have been, you know, working on this for two years now. And in addition to all the like stuff we have done, um, like one of the biggest improvements actually came from the, um, the L L M's, the,

• 22:41Speaker 00

  you know, uh, chatbots we are using. Um, so right now, uh, we just switched from like four O to, four O to four point one, and even that made a big difference. Even our support agent had like a, like a couple of point improvements, just by moving to a new agent. And so we are basically using two, two different, APIs. Uh, one of them is, uh, OpenAI four point one and the other one is Gemini two point zero. And we are also moving to two point five on, um, on Gemini as well. Uh, like we are just testing that out and we see that that's also going to improve things.

• 23:26Speaker 00

  So we, we also give people options to use, like, you know, which model to use, but we actually prefer to use multiple models. Uh, because this actually gives us an ability, like we have dashboards, like we can actually compare the performance of different models to each other. So that like, you know, we will usually round like fifty percent of the conversations, like we have about ten thousand uh, users right now, but our biggest user is ourselves, like, uh, Jotform's own support. And, uh, so, you know, uh, we are actually comparing the performance of different models. That's why we always run like two of them together. And if, if we see, and sometimes it's just our prompts, like we have to write different prompts for different models, like you, we cannot just

• 24:11Speaker 00

  write prompts for ChatGPT, like OpenAI four point one, and then use that for Gemini, like we, Gemini, we have to change things.

• 24:21Speaker 00

  Sometimes like we have to improve our process. Sometimes the the fault is not with the LLM. It's it's actually our fault. But because we are always like comparing them to each other, like we can discover that one model is actually performing much better. And then if some if if you see something like that, we will actually move most of the conversations, like ninety percent of conversations to a some model. But right now, the like the performance of Gemini and the OpenAI ChatGPT is very similar. They are very good compared to the older versions. So there we are. We are keeping it fifty percent. Yeah.

• 24:58Craig S. Smith?

  Yeah. Another question.

• 25:03Craig S. Smith?

  Excuse me, the knowledge base that you're attached to. Can you do set up your knowledge base on the platform?

• 25:14Craig S. Smith?

  Or is that something you need to contract JotForm to do in a in a bespoke way? I mean, if if if I wanted to build a chatbot for my podcast and load all my podcast episodes into it, is that something I could do on JotForm? Or would I have to contact you guys? And, you know, have you build a RAG system?

• 25:46Speaker 00

  It's all self-service. You just go there, you add your own knowledge base. You can you can upload your files. You can link to, you know, your website. You can link to your YouTube channel to be able to, you know, get the transcripts and use that as a as a knowledge base for for your from your podcast. So basically, it's just it's it's all self-service. It's all free in, you know, up to up to hundred conversations per month. It's it's all free. And then if you get over hundred conversation per month, it's still like, you know, very affordable prices.

• 26:28Speaker 00

  But it's also self-service, so you can build your knowledge base. And one of the things we did was actually, we actually made it so that you can actually talk to your own AI agent to be able to train it. So if you think about it, let's say you hire the support employee. Would you just dump all the documents on the support employee and say, okay, learn this and go talk to the customers? Or would you actually talk to that person, like, train that person, talk about your, you know, culture? How you talk to customers, like, what's important for you, things like that? You know, you would probably talk to this person, right? Uh, the same way, you can actually teach your agent. So we have a section called teacher agent.

• 27:11Speaker 00

  You can have conversations with your agents, you can train your agent. In the train section, I was talking about the knowledge base, uh, you can upload files here. Uh, you can enter knowledge, you can link to your website, upload files.

• 27:24Craig S. Smith?

  Yeah, on uploading, on uploading, uh, to the knowledge base, is there a, a limit as to how much you can upload? How many pages or?

• 27:35Speaker 00

  Um, there's actually, uh, there's a limit, um, it's, it's available in our pricing. I, I don't remember the exact number, but we give all of, uh, all of space, uh, for our users. And what happens is that, um, if you, if you get, if you don't, if you add like, um, small amount of, uh, knowledge base, um, you actually don't need to use the RAG, right? Um, so I believe it's, it's about, um, you know, I can't remember the exact number, so I don't want to give the wrong number, but it's until some point, like, you know, your knowledge is actually basically giving me the prompt. But if you actually load lots of, uh, data, lots of knowledge to your, uh, knowledge base,

• 28:19Speaker 00

  then what's going to happen is that, uh, we are going to, you know, automatically actually switch your data to the, uh, RAG, uh, the vector database to be able to query, you know, uh, the questions correctly. Because, you know, the context, uh, window for the, uh, uh, L L M's, like the A I, A I is limited. So we, you know, we don't want to give everything about like, with, with Jotone support, we have hundreds and hundreds of, uh, you know, uh, user guides. And you don't want to look up everything when you're asking a question about like some specific question. So, um, yeah, and, um, and it's all happening, uh, you, you just enter your data and then we take care of all those things, uh, for you.

• 29:02Craig S. Smith?

  Yeah, great.

• 29:04Speaker 00

  And yeah, this is the teacher agent, uh, section. Basically, good evening. It's wonderful to connect with you. Imagine if you're sharing a cozy cup of tea as we chat. Feel free to tell me what's going on your mind. So I could teach my agent, can you change your name?

• 29:24Speaker 00

  Your name to John. So its name is Alex.

• 29:36Speaker 00

  Okay, so basically it's.

• 29:41Speaker 00

  It says that it can change its name. And right now it's Alex. So updated name is John. Yes, proceed.

• 29:52Speaker 00

  My name has been changed to John. So as we can see, it's actually here. We can see that the name of the agent is now different. If you go here, it's now John. And so I can, you know.

• 30:09Speaker 00

  I can, you know, train it as well. Like, you know, I don't work on Tuesdays. Because it's taking appointment, I don't work on Tuesdays.

• 30:21Speaker 00

  Let's see if it will be able to take care of that.

• 30:26Speaker 00

  Scheduling purposes. Hopefully, it's not going to do that. Obviously, I will have to test it to see if it really did that change. But we could also check our knowledge base to see if added this to our knowledge base somewhere. Maybe I don't see it here, but maybe I will just test it here.

• 30:49Speaker 00

  Let's see.

• 30:54Speaker 02

  Welcome back, Itech and Tank. I am John, your AI agent. How may I assist you with scheduling your personal training session today?

• 31:02Speaker 00

  Do you work on Tuesdays?

• 31:06Craig S. Smith?

  Tuesdays can definitely be a great day to focus on your training.

• 31:11Speaker 00

  It looks like it doesn't get that. So, you know, it's sometimes just

• 31:20Speaker 00

  some of the things may not be, but one of the things I could do is I could just go to a tool. Let's see, appointments are here. It's actually using form, probably why it's not doing it. I'm not using the appointment tool. And if I was using the appointment tool, I could actually, you know, edit the time here.

• 31:41Craig S. Smith?

  Yeah.

• 31:42Speaker 00

  And say, you know, not work on Tuesdays. I could, I could handle that.

• 31:47Craig S. Smith?

  Right. Say on the training, can, can, can, this is a problem I've had. I've played around with building chatbots and depending on the model on the back end, there's a lot of extra empty language that the chatbot adds, you know? How are you today? You know, is there something I can help you with? And, you know, I don't want all of that. I just want will the training, how responsive is the training? If you tell it, you know, don't ask a follow-up question. Don't just get to the point.

• 32:28Craig S. Smith?

  Is the training responsive in that way?

• 32:32Speaker 00

  So I think that's more of what we call the personal AI persona. So basically, we actually have like a specific section for that, like on the chat. It's actually set to minimal, like just talk very short. On the voice, actually I'm on the voice. It's actually set to minimal. So I can actually make it longer. If I want to, but it's already there. But I can also like these guidelines kind of like prompts. Then I can just say things like, you know,

• 33:09Speaker 00

  you know, keep your answers short.

• 33:11Craig S. Smith?

  Right. And this is all going into the system prompt or something?

• 33:18Speaker 00

  Yeah, these are actually going to the prompt. So it's actually adding these guidelines to the prompt when it's answering the questions. Yeah.

• 33:29Craig S. Smith?

  Yeah, that's fascinating. I mean, one thing I want to ask.

• 33:37Craig S. Smith?

  How are you dealing with the competition? Because this sort of thing is happening all over the place. Everyone's building agents or a lot of SaaS platforms out there.

• 33:50Craig S. Smith?

  Frankly, not many as straightforward as yours. But to build agents, how do you see the market developing?

• 34:03Speaker 00

  I think it's a good thing that there are many products. This actually helps, you know, helps everybody because even though we don't really follow competitors, competition very closely, what usually happens is like, you know, some people will try all the products and they will come and tell us, hey, you are, you know, you are slow. You know, they will tell us, you know, yeah, your answers are bad or something like that. Like just, they will start asking for features because they saw in other products. So as a result, everyone improves. So I think it's a good thing that there are a lot of competition. But I think the biggest difference between Jotform and the main competitors like, you know, there are products like

• 34:46Speaker 00

  Agent Force by Salesforce. There is like Sierra from Brett Taylor, right? There's just, you know, all these big names. But the great thing is like they're all focused on enterprise. And we are more focused like as a company, as a product, we are always like, we have 30 million users and most of them are actually using the free product. And we also have an enterprise product and we have, we also have like big name companies using our product. But our main user base is actually like all those like small businesses, small businesses, small healthcare, small, you know, nonprofits, uh, small education.

• 35:28Speaker 00

  And we prefer like, instead of like, you know, uh, building a product for enterprise, we prefer to build this like SaaS service, easy to use product for everyone. And I think that's the main difference between Jotform's product. It's easy to use, it's available to everyone, it's free, uh, easy to start compared to those, you know, big shops, big names. Like just, uh, you know, we are more grounded. And I feel like that's a better approach because like, you know, we get more chance to be able to improve our product, to make it great. Because like we are talking to like so many customers and uh, yeah, uh, that's our approach, but uh, it's only been like two months since we released the product, so time will tell us.

• 36:11Craig S. Smith?

  Yeah, no, it's, it's wonderful, it's fascinating. And uh, to your point about small and medium sized businesses. You know, I talk a lot to the big consulting firms. You know, their minimum bill is a mi- million or something.

• 36:31Craig S. Smith?

  And uh, but there are so many companies out there that want to implement something uh, on their website or uh, or internally. And uh, they can't afford these consulting firms to come in and and tell them how to do it. But this is self-service, this is something that people can do.

• 36:58Speaker 00

  They need this product the most because they're like, if you're a small business, if you are this, you know, education administrator. If you're nonprofit, you know, healthcare, like you're just so busy, you don't have time.

• 37:11Craig S. Smith?

  Yeah.

• 37:11Speaker 00

  To like constantly answer the phones, constantly. And that's why, actually, there are really big form users as well. That's like why we know these. You know, this segments really well because they have been using just from forms for two decades. And and we understand these industries, like the small businesses and. And they actually need AI the most because the AI can actually, you know, work twenty four seven. It's very cheap. It can, you know, it it doesn't have to be perfect, you know. It doesn't have to be perfect. If it can just answer the questions of your, like, let's say you're a restaurant, someone calls.

• 37:53Speaker 00

  All of business hours, you know, if you can just answer their questions, like where are you, what's what time are you going to be open, things like that. It's okay, like, it's not, you know, it's not a deal breaker if, you know, AI helps us once in a while. And just, you know, it it can do so much for these small businesses. And I feel like the people are also ready for AI. So like people are used to using ChatGPT and it's like, it's now a consumer product. Even my kids, like my kids are in elementary school, like my two boys are in elementary school. And and when they are doing their homework, they are actually talking to ChatGPT on the voice mode.

• 38:37Craig S. Smith?

  Yeah.

• 38:38Speaker 00

  On on the, you know, on the phone. And it's just very easy for them to like research things, ask about things. Even they can really use the AI. And now people are used to talking to AI.

• 38:51Craig S. Smith?

  Yeah.

• 38:52Speaker 00

  But but with L L M, they can only get in general information, right?

• 38:56Craig S. Smith?

  Yeah.

• 38:57Speaker 00

  What if like every every business, every organization in the world had their own like ChatGPT?

• 39:03Craig S. Smith?

  Yeah.

• 39:03Speaker 00

  That instead of going to chatgpt.com, you could go to their website and talk to them.

• 39:10Craig S. Smith?

  Yeah.

• 39:10Speaker 00

  By using their AI. You know, with their knowledge base in the backend, with their information, with all the forms they have. You know, if you need to buy something, you can just talk to the AI and buy it. Uh, if you want to learn about something, you can learn it. And if every business actually had this, like, it will be, you know, it will be also great for the consumer. You know, it would make things so much easier for everyone.

• 39:32Craig S. Smith?

  Yeah, yeah, and, and, you know, uh, I think everyone should have a personal website. I mean, why not? Uh, and you could, everybody could have their own chatbot on the, or agent on the website where, you know, instead of hunting for their phone number or email, you could talk to their agent and if you wanted to, you know,

• 40:03Craig S. Smith?

  set up an appointment or something.

• 40:06Speaker 00

  Uh, it's instant, you get that, so you can call, you go to their website, get instant, instant as first, um, you know, it's great.

• 40:15Craig S. Smith?

  Yeah, so.

• 40:17Speaker 00

  Yeah.

• 40:18Craig S. Smith?

  So you see, uh, and, and, uh, and you, once you, um, are beyond a hundred calls or a certain amount of, uh, data in the knowledge base, you start charging. And is that a flat subscription fee, is it a usage meter, or how do you charge?

• 40:40Speaker 00

  It's actually, um, uh, like, the same with JotForm, um, we have all these like bronze, silver, gold, uh, plans. Uh, it's like monthly subscription fee, uh, uh, and like starts at like twenty nine dollars a month. Or, you know, you can buy a year, get fifty percent discount as well, but basically it's just, um, you can increase your limits by upgrading to a higher plan. So it's not like credits, it's just, uh, you know, subscription plans, monthly subscription plans and. You know, if you want to use it for two months and then stop using, you you know, you are free to do that as well. I just um. And um and most people actually start with the free plan, uh see how it works and.

• 41:25Speaker 00

  You know, once they get like, you know, so much, uh so many people using their forms or their AI agents, then they upgrade to the paid plans.

• 41:33Craig S. Smith?

  Yeah, yeah. Uh, yeah, this is wonderful and and uh and and you're growing uh. Quickly or how is the the demand, the uptake? I mean, is the fact that. The market's crowded uh making it difficult, or is the market so large that. That there's plenty of room for growth for everybody?

• 42:00Speaker 00

  So uh in two months, we were able to grow to like uh ten thousand uh users. And um so it's uh it's actually grow growing in a steady pace. Uh it's not a huge like explosion. It's. Uh growing and and most of the users actually came from existing Jotform users, uh people who already they have forms. Uh and uh but. And I'm actually uh okay with not growing too fast, uh because. Um because we are, we are actually improving the product every day, like just uh we are. And the and because we are using our own product with our own support. That every day we are uh improving the product. So you know.

• 42:43Speaker 00

  Uh I actually prefer to go in a steady uh growth path as opposed to like sudden explosion of, you know, millions of users. And and uh and because the quality is improving every day. And I feel like, you know, uh I actually liken um. Liking AI to Dragon, like uh if you think about like Game of Thrones.

• 43:08Craig S. Smith?

  Yeah.

• 43:08Speaker 00

  Uh like, it's like a, you know, Dragon of Calithia and uh.

• 43:16Speaker 00

  It has like so much power, but it can also burn you.

• 43:22Craig S. Smith?

  Yeah.

• 43:23Speaker 00

  So you wanna be careful about like just, you know, it takes a while to train, train, you know, a product, you know, improve a product. So I'm actually okay with taking the time, like just improve it every day. And as the, I'm sorry, also improving like the quality of the product is also improving. So I'm actually very happy with how things are going so far.

• 43:49Craig S. Smith?

  Yeah. Is there anything that we haven't covered that you would like listeners to know?

• 43:55Speaker 00

  Nothing. I feel like, you know, that's the vision. The vision is, you know, having every organization, every small business, you know, every business in the world have their own AI agents. So that, you know, when you, when you wanna talk to that business in the middle of the night, like 3am, you can talk to that business. You can, you know, get information, you know, buy stuff and, you know, set up appointments, whatever you need to do, like, you know, all the tools that you could make available to your AI agents could be done by the AI agent. So I feel like that's a great vision. And it's

• 44:37Speaker 00

  gonna be a great feature for everyone. And that's the promise of AI, and I really believe it.

• 44:44Craig S. Smith?

  Our sponsor for this episode is Extreme Networks, the company radically improving customer experiences with AI-powered automation for networking. Extreme is driving the convergence of AI, networking, and security to transform the way businesses connect and protect their networks to deliver faster performance, stronger security, and a seamless user experience. Visit extremenetworks.com to learn more. That's extremenetworksallruntogether.com to learn more.