Name: Protecting Revenue and Loyalty with AI Voice Agents for E-Commerce
Uploaded: 2026-10-22T15:01:35.797Z
Duration: 45 min 56 s
Description: Protecting Revenue and Loyalty with AI Voice Agents for E-Commerce

Transcript for "Protecting Revenue and Loyalty with AI Voice Agents for E-Commerce": Hello. Welcome, everyone. Come on in and get settled. We're just gonna give people a few minutes to trickle in, and I'll just share this live in the in the meantime. K. We'll just give it thirty seconds or so, and then we'll get started. Okay. Hi, Gonzalo. So we're just gonna give people a few moments to to join. But while we wait, maybe instead of the usual where are you tuning in from, we're all here to talk about ecommerce today. So I just want you to open the chat and drop an emoji that perfectly encapsulates how you or your team maybe how maybe you feel during peak season, during Black Friday, or the holidays. So just drop a relevant emoji into the chat. I feel like that GIF could communicate so many different things. Okay. A lot of a lot of happy emojis. I feel like people... You must be feeling ready. Ready to crush it, John. No. Matt. Melissa, you're looking a little bit worried. Everything's gonna be okay, guys. You are in the right place. Nice. Okay. Cool. I think all of our speakers are here now, so let's get started. So before we get into the webinar itself, hello. My name is Kate Gallogly. I'm on AirCall's product marketing team. I'll be moderating the webinar for today. We'll just get the housekeeping stuff out of the way quickly. So you may have noticed already that you were in listen only mode, so your mics and cameras are automatically turned off. But we absolutely do still wanna hear from you, so please feel free to get into the live chat as you have already already been doing. Feel free to drop any questions that you might have in the q and a tab. So you'll notice that we have three tabs on the right hand side of your screen. We have the chat tab, and we also have the q and a tab as well. So feel free to use the chat tab as much as you want to leave any comments that you have. But if you have a specific question that you really want an answer to, please enter your question in the q and a tab. We will have some dedicated time for questions at the end of the session. But just in case time is not on our side, because you never know, we will also be answering questions throughout the session backstage in the q and a tab. And, yes, also, we are recording this. A recording will be shared after the event as well. So to kick things off, let's just talk quickly about taking some of the pressure off your team into which I can... If I'm understanding these emojis correctly, I think some of you are feeling. So you are gonna see in a few moments a poll that's gonna pop up in your screen right about now. It should be the bottom left of the screen, or you'll also see the little poll tab open on the top right of the screen as well. So it just asks, what would you want an AI voice agent to handle for your business? You'll see a few different scenarios on there, whether you use an air voice agent for routine questions, like where's my order, whether you're more interested in outbound calling to chase up abandoned carts or share updates. Just go ahead and lock in your votes. Okay. The results are moving pretty quickly. Okay. Yes. No surprise here. I'll see if I can share them so you can see the results too. Okay. So with my order, it's definitely taking a massive lead. Absolutely gonna be the bane of everyone's existence around this time of year. But I'm also seeing some votes for outbound calling, so calling customers with delivery updates, chasing up abandoned costs. That is really interesting because it tells me that you are not just looking to save time and support. You're also looking to use AI to generate revenue, which is a perfect segue into what are we gonna cover today. So let's get into the agenda. Okay. So firstly, we have Hami on our partnerships team, our director of global tech partnerships and new beds. He is gonna break down the shift into this new era of voice in ecommerce. Then we have Matembe, our senior product manager, who will take into... Take you into our product road map, so what have we released lately and what is coming up next. And after that, we will have all one of our four deployed engineers who will share a demo of agent creation and Shopify AI actions as well. And then finally, we will round it all off with a live q and a with all of our speakers. So let's get right into it with Hami who is going to kick us off. There I am. Thank you for the intro, Kate. Hello, everyone. My name is Hami, as Kate mentioned, a leader of global technology partnerships team. And and we've been really working for many years now, with ecommerce merge... Ecommerce merchants. And we know there's a growing challenge. Customer expectation keep rising. Support teams can really scale at the same pace. And then during peak periods like Black Friday, the call volume goes up. Teams get flooded, calls, you know, go unanswered after hours. And, especially for lead merchants, we understand that there is quite a frustration, from from customer that comes with it. And And so that's what's really changing now. VoiceAI makes it possible to handle more of these conversations, capture more of the demand, and then also turn, these interactions into business intelligence without really growing headcount at the same rate. So before we kind of, talk about what was changing, I wanted to step back for a second and also speak about why voice as a customer engagement channel has always been quite critical. So we've worked with merchants globally, and we see really three main reasons why they invested in voice. So the first one is, you know, across generations, customer really turn to voice when they need a fast answer to a problem that they can't really easily solve. So imagine you've bought an expensive wedding gift. It's Thursday. It hasn't left the warehouse yet, and you need to be... You need to have it in your hand really delivered by tomorrow. So you wanna kinda speak to someone who cannot only solve your problem, but also can tell you in a very reassuring voice that it will arrive on time, which is really difficult to mimic, in asynchronous channels. The second one is voice has an impact on sales conversation. So especially for, like, high value purchases where personal advice can actually make a difference between browsing and buying. So imagine you're you're about to spend $2,000 on an espresso machine, but you're unsure whether it fits in your kitchen. You have four different vendor tabs open, and and a conversation, that addresses your specific needs can, in that moment, really give you the confidence to actually complete the purchase. And the third one is voice has really a positive impact for businesses who rely on recurring revenue. So to stay kind of like with this coffee analogy, think of a coffee bean suppression business. A customer wants to cancel because they're unhappy with the quality. But during that conversation, the agent actually discovers that they've been using an espresso roast in a French press. And so with the right recommendation, they can not only solve that problem, but also potentially save that subscription. So voice has always been critical for ecommerce businesses, but the challenge has been the cost of delivering and scaling it. So you can only have a kind of, like, one on one conversation at a time. It requires trained human personnel and also the cost of maintaining that infrastructure. That's why we, you know, and and many merchants have historically prioritized asynchronous channels like email or chat. They're cost effective. They're easy to scale, and AI has been improving them already for years. For voice AI, that's a little bit different. So, you know, the technology advancements are are changing rapidly. Just a few months ago, voices, Sonnet, still very robotic. Agents kind of struggle to recognize emotional cues and also respond in a natural manner. But today, we're seeing remarkable progress. And so with voice AI, becoming more natural, more capable, what we also see is that the cost of serving these customers is coming down, while the value that you can unlock from these conversations is going up. Let me unpack a little bit what I mean by that. So voice agents can already resolve most of the complex queries with human intervention. Now it's not a one on one conversation anymore, but it's a one to many. It's serving customers around the clock, including weekends and public holidays. The second one is voice is a channel that generates incredible rich customer data in a very short amount of time. So think about the spoken word. It's so much faster than what you're able to... Or even willing to type. And now imagine you you take a chat GBT or a cloud, and you connect that to Aircall and cross reference that with thousands of conversations and then with data also from Shopify and your other business tools. Now all of a sudden, you are able to identify the most common customer pain points, understand what drives these conversations, and also rank these opportunities by the potential revenue impact. And that really can help you make incredibly critical business decisions. And the third one is voice will become one of the best selling channels. Email campaigns are becoming increasingly crowded. It's increasingly, difficult to, get mindshare in a in a very crowded inbox. Voice is a powerful new way to engage customers proactively through one off or evergreen outbound campaigns, while also turning everyday conversations into personalized upsell opportunities. But we know that voice agents alone won't solve every merchant's challenges. So what merchants really need is a platform that supports the entire customer engagement from one side to the the end, basically. We believe that voice agents and messaging agents need to work seamlessly alongside human teams. We need to give merchants also the tools to improve pickup rates and increase conversion, such as the ability to, have your logo on an incoming call. And so we we can't be... We couldn't be more excited about what this future will unlock for our customers. I will wrap here. Just... You know, Aircall, is already the partner of choice for many merchants around the world. But I do wanna call out that, our team is a very curious bunch. So we love learnings from from customers. We love learning from partners to understand really their business challenges and find also new ways to help you grow. So please don't, hesitate to reach out to me or the team, whether it's on LinkedIn or or through your account managers. We'd love to hear from you. And and, you know, with that, I will kind of wrap it up here and hand it over to Matembi to walk us through our product insights. Thank you so much, Amy. Hi, everyone. My name is Metembe. I am the senior product manager leading our AI voice agent team here at Aircall, and super excited to be talking to you all today. So what I'm going to do today is, walk you through what we've just released, the exciting things specifically for, like, ecommerce to help, ecommerce, people working in the e commerce industry like yourself deliver more value to their own customers and make sure that you handle the volume, the higher volume that's coming up, especially in the next coming weeks. So we have a really exciting road map and very excited to show share with you all what we've done and what's also coming up. And we also have a a a surprise for you all, because you're here in this in this chat today. We would love your feedback on a specific feature, to make sure that it better serves your needs. Okay. So first, as mentioned, I'm going to share what is new. And so we've released a few exciting things, a lot of exciting things in the past, couple of weeks. And on a high level, we really want to make it easy, for you to create agents. We want to make sure we give your agents, the capabilities that they need to bring more value to your customers and to you. And then we also want to make it easy for your agents to to handle large volumes and save time for you and your team. And then while doing all this at the same time, giving control... Giving you even more control over your customer experience. At the end of the day, we do acknowledge and we do know that, you're trusting these AI voice agents to, talk with your customers, to interface with your customers, and you need the control, and you need to know what exactly is happening. So we want to make sure, as Air Corp, we want to make sure that we give you this control and we give you the assurance that you need. And so I'll be diving in deeper into the specific things that we've just, released. And let me know in the chat if you have any questions. The whole whole team is here, so happy to answer. We'll go through all your questions today. I'll start with the top. As mentioned, we want to make it easy for you to create agents, and so we we released an order support, agent, which we call also the Wismore agent, that you can... In a in a... With the tap of a a few buttons, you can literally, like, launch a live, Waze My Order agent that is able to get your your your customer's queries, ask... Answer your questions, answer your customer's questions as they call you and share details about where the... Where their orders are. I'll... And exciting thing is Paul, who's our port... Port deployed engineer, will be sharing with us later on how to set that up and how it... How, actually, that works in a live demo. And then the second thing is, the the the Shopify AI actions. So we're going beyond just reaching back, or, like, reading back an order for calling customers. But what we're actually doing now is helping things helping helping things get done for you. And this means going beyond just one specific, like, order to checking returns, checking refunds, putting shipments on hold, putting, a delivery address, updating delivery addresses, and sending, link updates. All this can be handled by your AI voice agent. And so that's the AI, the Shopify AI actions that we we released as well. And then thirdly, as mentioned, we want to to make sure that you're hand... You're able to handle outbound, high outbound volume. And therefore, we have what we call the AI voice agent outbound campaigns, which is AI voice agents that are able to get a really long list and make outbound calls all while while you're relaxing, all while you're, like, making sure that you follow-up with the customers that you need after they make, key purchases or follow-up with VIP customers, or maybe even check up... Check-in with customers that, did something specifically in your ecosystem that you want to follow-up with, in an outbound campaign. We've seen really interesting use cases here for these, like, AI voice agent outbound campaigns. And I'm really excited, to hear your feedback on how exactly, maybe for your company, how you'd want to use them in your use case. And then lastly, as mentioned, we launched, we want to make sure that at the end of the day, you have control over the workflows, for your for your agents. You have control over how exactly your agents, talk to your customers because we understand that is very important. And so we launched what we call, AI voice agent simulations, where you can actually simulate the different types of customers that your AI voice agents... We expect your AI voice agents to talk to. You know your customers best, and so you can actually create these simulations as maybe that, interesting customer that your team spoke to last week or the other interesting customer that your team spoke to customer that your team spoke to two years ago and actually see and control how you want your AI voice agent to interact with the customer. I believe that we might also like showing that, in a live demo later today. Cool. And then, of course, I will also share what's what's coming next. The team is hard at work, and we have a lot of exciting things coming up. And, the first thing that I'm personally very much excited about is the abundant cut template. So as we as we've spoken to a lot of, ecommerce merchants like yourselves, we do know that one of the biggest, like, one of the biggest things that, people really want to solve for is reaching out to customers when they maybe come to your to your website, add something to their cart, and then abandon the cart. Maybe it's been two days, and you want to maybe reach out. And as Hany mentioned, e... Like, email is getting very clouded, and now is a really good time for us to use, voice agents to reach out to those customers. And so we're working on a solution, to better enable, like, through Shopify, enable, like, abandoned cuts, like, campaigns, so that we can automatically reach out to to to the customers that edit stuff on their cart and abandon their cart. Secondly, we're adding even more voices and voice controls. There's been so much improvements over the past, specifically the past three, four months here, and we're continuing to even improve the more voices and add more controls so that you have control over, your brand and how your agents sound. Thirdly, I'll be showing this in a different slide, but, we are adding, the AirCall MCP and expanding, to Ask AirCall, making it easy for you to even create agents from your Claude, from your, from your chat GPT, and, which has even more context about what... Where your, echo... How your echo live live agents are interacting, and where maybe your team... Your your team of live agents is struggling today. And then lastly, we want to give you even more control, and therefore, we'll be redesigning and improving our procedures, to make sure that you understand how the workflows, for your agents... For your for your for your AI voice agents and to even understand, like, better... Have better control and better manage those different workflows and procedures for your AI voice agent. So cool. So that's basically what we've we've did, what we've just released and what's coming up next. And in the next slide, we'll be diving a bit deeper. Actually, before we dive a bit deeper, as I mentioned, we have a lot of use cases. We've seen a lot of use cases for outbound campaigns, and we just wanted to pause here and see, and get a feedback from you here live in the call. Like, how specifically would would you use outbound campaigns with Shopify? So in the next thirty seconds, we'll pause and then just point your phone, to to the link, on the... To the to the QR code and go through the questionnaire and share share some feedback with us. Yeah. It's a two minute survey. Would just love to hear from you, on how exactly, you would use outbound campaigns with Shopify. Alrighty. Seeing that, like, feedback is starting to come in. We'll try to also share a link in the in the chat. But... Cool. Thank you all so much for the feedback, and, do let us know if you have any, like, other other questions in the q and a section. But cool. I'll go back to dive a bit deeper into some of the the launches that I mentioned. Do let me know. We also have the team on this, in this in this presentation. So happy to answer questions about any of the of the of the of the features or any of the workflows or new things that you're seeing within the the the echo ecosystem. So in the next... And in this slide, I want to just show you what the the order support AI voice agent template looks like. So, again, we want to make it as easy as can be, for people with the e... With with the Shopify integration to launch, test, and release their AI voice agents. And we've done so, for the past few weeks by this, Shopify integrate... Shopify ways my order template. And so we also call it the order support template. So all you need to have is a Shopify, yeah, the Shopify integration. If you have an inter... A Shopify integration, go on, like, the Aircall dashboard. And once you open the Aircall dashboard, you will see the the list of available templates there. It's listed in the available templates. Just, you can either test the agent, see how the agent, the typical agent, would would would, behave, and then, try it out in a in a in a click of a few buttons. You will be able... Your customers will be able to, to talk to the AI voice agent, which is integrated to Shopify and able to pull up their order their order details. Cool. And, again, Paul, in a few minutes, will be walking us through, this template. And then, also, I did mention the Shopify AI actions, which is... We're just going beyond giving order support, which is, just where is my order? But we actually want to deliver even more, value to you. So we'll be sharing details about, like, updating addresses. We'll be thinking, we'll now have AI actions about, shipments on hold. We have AI actions about, sending link link, payment link sending, and generally, like, or also orders and refunds. And generally, the TLDR here is we want to help you get things done. And I'm really excited about this. And this opens up so many more possibilities with our AI voice agent. And then lastly, as I mentioned, now in beta... And please reach out to your account manager or, like, to your... The Forteployed engineer that you're working with. And, we have now, like, our AirCall MCP, which is able to do even more things and connect it with your cloud, connect it with your open AI to have more context about your own setup. Ezyby also is able to deploy AI agents, with the context of how you are set up and what your customers or your calls... The calls that are coming into your live live agents are are being... How they are being handled. Super excited about this, and the the the possibilities here is, the possibilities here are endless. So, reach out to your account manager to get into the beta, and, happy to support you here, and as you find really interesting use cases for your team and for your customers. I think that is all in terms of our our road map. What we've what we've released and what's coming up. Please do reach out, if you have any questions. Happy to answer them. And then next, we'll have Paul who'll be, doing a live demo. Thank you, Matame. So, yeah, I I guess the demo is worth a a thousand words. What I'm gonna do is I'm just gonna walk you through maybe, how I could create an agent that's connected to Shopify using our templates, that Matame talked about and then comparing it to maybe a more of a of a manual setup just to see, how you could, use both. So, then I'll just do I'll just do a live call and show you, how it sounds. Okay. Let me try to share my screen now. Okay. So there's essentially two ways to get started and connect your AI voice agent to your Shopify account. The first one, and it's probably the easiest way to go about this, is using one of our AI agent template. So if you go to your dashboard, click on AI voice agents, and select AI agent template, you will see this new order support, template that we're gonna use. And here, you need to select your integrated app. We have quite a bit in our instance, but you would only... Typically only see one. I will click on save changes. And here, I'm being brought to this, this setup flow where I'm gonna have the opportunity to customize a few things. On the first page, you see, you know, the role of the agent and the different tasks that it can do. We see that it can look up an order, status and tracking from Shopify, answer product and store questions, with its knowledge base, explain the return and refund policy, depending on on the status of the order that was returned, and also take messages for your team, if they're not available. So I'll continue on the setup. Here, I can review the voice and speech settings. So I can change my agent name. I can also change the the greeting message, which is the first sentence the AI says when it's picking up the phone call. And I can also change the farewell message, which is the last sentence if the AI, ends the call. That's for how it's gonna introduce itself and and end the call, but I can also change the voice. We have a few different options, 38 different languages and different accents for each. So if I wanted a British accent, I can choose, one of those and select the tone of voice to be either friendly or professional. Next, you can, start creating the knowledge base of this agent so that it can answer some questions. So you can add, some knowledge sources either with a block of content, which is just plain text. Here you can copy and paste q and a's, for example. Let's say I will add an FAQ section with what is the return policy, And my answer for that will be we offer a thirty day return for every item. So now the AI will be able to answer that question. I can also add directly from a website. We are able to crawl, your your website if you have information that is relevant there, and we can also use third party sources, like Zendesk articles, Salesforce articles, or even HubSpot articles. Next section is the Shopify intake questions where we can review, what all EVI is gonna ask. For example, we see here it's gonna ask the order number, and this is to collect order name. This is part of the standard procedure for this action. We automatically detect the phone number from the, you know, the phone number that's calling, and the email, we all also automatically detect it. And once we're done, we see a conversation overview, which will tell us pretty much how the call is gonna flow. So here, the customer calls the agent. Olivia is gonna say, hi. I'm Olivia. That's the greeting message that we that we defined. If the customer asks where their order is, Olivia will say, sure. I can check that for you. Can I just add your order number? If the customer gives their order number, it will, be able to give out information and end the call there. So if it looks good to me, I can click on looks good and continue and select the number that I wanna add it to. Just choose this one, for example. And then, you have the ability to directly click on this button to edit the call flow and add your agent. There you go. So once you've finished the template flow, you get this, fully built config of the agent. You can see all... Everything that we put has been entered in the different sections, like our greeting message, our farewell message. The knowledge base that we've created is is here, and you can obviously tweak it as you go, but you have this version of the agent that is already ready to test. So now that you've seen how you can create an agent with the templates, we can also create the agent ourselves manually by selecting build an AI voice agent and clicking on inbound. So that brings us to, you know, a plain agent that doesn't have any settings yet. It has a placeholder for the name, the standard voice, and it doesn't even have a call script yet. All of that, you should be familiar already, from what you've put in the template, but the most important part is how you can add this this Shopify connection. So you can go to the AI actions tab and select add an action. Here, you would select Shopify and select the right environment. Again, you probably only have one on your account. And you see here, these are all available AI actions for this, Shopify integration that I have. I have quite a few, but if we're only interested in in Wisma, where's my order, we probably want to select get order details and add that action. Now you see this message saying missing required actions. Some, AI actions have dependencies, from one another. Get order details is taking the internal order ID from Shopify to retrieve, specific information about the order, like the request address change, the check return eligibility. So for that, you need to first get the order by order number. I will go back to this integration and select get order by order number, and I will add this action. You see this flag is gone now, and my voice agent has the ability to call these two actions in the middle of the call. But for now, it doesn't really know when to do that, so I can add a bit of instructions in here in the conversation guidelines for these actions. We have an example here. If the caller wants to update their contact information, first use and then the name of the action. So here's an agent, for example, that I've built manually. I gave it a name. The company is gonna be Acme Store. I changed the voice to, Emily, which is an American English accent. In my call script, I said, hi. Thanks for calling Acme. I'm Maya, Acme's virtual assistant. How can I help today? And in the conversation guidelines, I gave, some details as to how the call should, should flow. For example, I have a procedure for finding an order where I I tell the AI to always ask the order number if they haven't given it, and, then ask the email address associated with the order. And then it will call the AI action to retrieve my order by order number. You can see this action takes, can take multiple parameters like order name, which is the order ID, and the email address. So we even have the ability to transfer to different teams, if people are not happy with the resolution of the AI, and we can route based on, you know, what they're experiencing. If it's a delivery problem, if it's a a return or a refund, or if they just simply ask for a person. We have these fallback options, just in case. So we can just give it a test and, do a real call now. Hi. Thanks for calling Acme. I'm Maya, Acme's virtual assistant. How can I help today? Hi, Maya. I wanna check the status of my order. Sure. I can help with that. What's the order number? It's 11352. And what's the email address on the order? It's jeff.smith@gmail.com. Okay. Let me see. Your order is shipped with UPS, and it's on its way to you. Was there anything you needed to know for this order? Yeah. I wanted to know what I ordered on that one because, I I have different orders with your business. For this order, I can see three items, air travel survival guide, Baba lot air pure, and Harry Potter full edition. Anything else you need for this order? Yeah. And what's the tracking number? The tracking number is, 1Z206574. Anything else you need for this order? Yeah. So I looked at the tracking page, and there doesn't seem to be any movement. So I don't know if it was picked up by UPS. Can you, cancel that order? Let me transfer you to our team. One second. Yep. Thank you. I'm putting you through to our delivery team now. They'll have a note of what we discussed. Okay. I'll just, hang up. So there you go. You can see the voice agent passed the call to the delivery team because it's not supposed to handle, cancellations, and it provided a bit of context as to what happened during the the voice agent part of the conversation. Okay. So, hopefully, that gives you a a good idea of how you can build these agents and and how you can test them, how they sound. I think the next part is just opening up to you guys. If you have any questions, you should have a section in your UI to ask to ask your question, and we're gonna answer it live. I don't think we can hear you, Kate. Can you hear me now? Yeah. Alright. Okay. Sorry about that. Yes. So I was just saying thank you so much, Paul. Like he said, please feel free to go into the q and a tab and ask any questions that you might have. I see there's also just a a comment from Gonzalo. Is this going to be uploaded? Yes. Everyone who has registered for the webinar will get a recording in their inbox, after the webinar is over. And we also have one from Nicholas. So how does the AI voice agent pass notes to the delivery team? Is this via insight cards? I assume that means, to the... To your human teams. But does anyone wanna take this? Yeah. Maybe Nitembe will... I I can take it. Basically, the voice agent, when it transfers the call, it, it creates an escalation reason, which is just a one sentence, a little paragraph that explains, what happened, what information was collected, and and just giving it a bit giving a bit of context, for your team. So when the call is transferred, it rings on the workspace, of your team, and they see... So they see the call coming in and they see also this message, popping up. Nice. Okay. Great. We. also have a question, from... Oh, go ahead, sir. yeah, I... Just to add on what, Paul just said, so each time that, their... The AI voice agent escalates to or transfers the call to a live agent on your team or yourself on your team, you will be able to see that escalation reason why that person wanted the escalation. And so that way, we want... We make sure that, live agents also get some context and don't answer, don't answer a a a call with no context. Nice. Okay. Great. And, also, I think one for for you, Matembe, from Julian about the Aircall MCP. Oh, yeah. So, Julian, so it it is live, but it's on beta right now. So we would have to add you to the beta, but it will be rolling out in the coming weeks. So just reach out to your account manager or reach out to any of us, and we can add you if you're interested in starting to test and playing around with it, and we can add you to the to the to the beta group. Okay. Question from Nicole. What if, the call is outside of business hours? Can we offer callback requests? And where is the escalation info kept in that case? So, Nicole, remember when I shared, my screen, I showed you the the call transfer rules. For an after hours agent, this is typically turned off. So we don't generate an escalation reason because we don't, we don't escalate the call, but we can also take down the message and log it in the notes of the call with, intake questions. We could still log a summary of what happened, and we change a bit the script of the agent to, offer a callback and say someone will follow-up shortly. Okay. We also have a question in the q and a tab from Jessica. Is a voice agents, are they charged per minute? And the answer is yes. But the per minute rate does depend on your overall usage. So, we can start with a pay as you go rate, which the the rate really does depend on your specific currency. So I would encourage you to check out the the pricing page, but, our pay as you go rate is really designed for testing. It might be, a little bit higher. I believe it's, 49¢ for up to twenty five hundred minutes and 39¢ thereafter. However, we can... There... There's a lot of flexibility around pricing, so we can actually go, as low as, 19¢ a minute or lower in certain very high volume use cases as well. So I would encourage you to reach out to our sales team and and really, discuss, pricing for your specific use case with them. Question, for... I I can take that one because I had. the same issue with a client, the other day. It... This might happen when you've integrated HubSpot to air call a while ago. It... The integration that you have might be missing some permissions to retrieve articles. So my recommendation is go to your Aircall dashboard, click on integrations, find HubSpot in the list of apps, and try to reinstall, like, a second version that you can call AI actions and see if that one is able to retrieve your articles. Nice. Question from Chris. We don't use Shopify, so what options will be available for non Shopify users? I don't know. Maybe, Harmar, you wanna take us through some of the, the integrations that we work with, potentially? Yeah. I'm, happy to, like, you know, receive any type of requests for additional integrations. We, developed our... Developed a platform in such a way that it's actually quite easy to build new agentic integrations. So if there is specific integrations that are not covered today, feel free to reach out to me, and we'll we'll gonna look into it, with our team. Nice. Yeah. And, also, please please share in the chat which integration you use if you... If it's not Shopify. We're always open to expanding, and, we we can find a way to to integrate with you. I. think that's true, Chris. Yes. And we have a related question from, Nicholas if, someone wants to take that as well about custom API actions. Yeah. I can I can, take that? So, Nicholas, I appreciate this maybe for, different seminar. We primarily use a custom platform database, Snowflake, and CRM HubSpot. Can we create custom API actions that can replicate the functionality of these native Shopify actions? Yeah. Totally. The the native Shopify actions offer you, like, a quick way of syncing your accounts, with with template actions, but we also have the ability to create custom API actions, for different systems as long as they have a... You know, API endpoints that we can communicate with. We're able to create these, API actions and then orchestrate them in the script of the agent. Awesome. Okay. We have one minute left if there are any further questions. If not... Oh, here we go. Is it possible to connect AirCall to an ERP like Business Central? We have all of our business data in a specific server. Who wants to take that one? I can take I can take it. It depends what type of data you wanna retrieve. If it's for the knowledge base, we... It it would be a bit different than doing custom API actions like I was talking about. But as long as you have an API, generally, yes, we can connect with your tool. Awesome. Okay. Thank you so much, everyone, for the questions. Thank you to our speakers. And, yeah, we will share a recording of this session after, after session via email. And, yeah, thank you so much for joining us. And, yeah, see you at the next... At the next one. Thanks, everyone. Thank you. Bye. Thank you. Thank you.