Name: Aircall AI Agent x Salesforce Integration
Uploaded: 2026-08-13T01:01:07.965Z
Duration: 44 min 20 s
Description: Aircall AI Agent x Salesforce Integration

Transcript for "Aircall AI Agent x Salesforce Integration": Okay. Good morning, and and welcome to everyone on the webinar. I believe today we have some partners. We have customers, potential customers, and, just some real AI agent enthusiasts on the call as well. My name is Frank Eagleton. I haven't had the pleasure of meeting you before. I'm the managing director for Aircall APAC. I'm based here in our Sydney office, doing the call from our Sydney office today. Today, we're going to continue our webinar series on AI voice agents with the focus being on how we connect with Salesforce and the benefits for you, your business, and, ultimately, your customers. I guess before we start, just Aircall today as a as a business here in Sydney, I'd say we have about an average 120 conversations or meetings or both in a day. Today, more than 80% of these conversations include AI voice agents. The market. People are are really curious and want to understand how to bring these agents into their practice to ultimately deliver better customer experience, increase revenue, and ultimately focus human resources on the tasks or conversations where it matters most for the business. This is what we're gonna dig into today. On the call today and leading the webinar, we're gonna have Trent Wall. He's our Aircall APAC forward deploy engineer. He literally lives and breathes AI voice agents. He's ex Salesforce, so, he's really passionate about the webinar today. But before we start, just a little bit of housekeeping that I want to talk through. So the webinar today is being recorded. It will be shared with all registrants after the meeting, so there's absolutely no need for for any note taking. Book a meet I encourage you to really book a meeting with your account manager at any time via the book a meeting button on the screen. I encourage you to do that. Have a conversation ultimately leads to you talking to Trent at the end of the day as well, getting some expert opinions as well so it starts from booking a meeting. And then we'll close the webinar with a q and a. We will have Will Cody, who's our customer success leader, monitoring the q and a at the end, but encourage you to put your questions on the right hand side as well. We'll answer them, but box them for the end as well. So I'd encourage you to stay to the end. There's normally some good discussions at the end as well. That's it for me. I'm gonna pass you over to Brent. Brent, over to you. Alright. Thank you so much, Frank. Awesome. Well, very nice to meet you all, on today's webinar. Thank you so much for taking the time to attend today. I'm really excited, to present today. Like Fran said, I think there are some people on the call that might recognize me from my Salesforce days. Previous to my role here at Aircore, I spent about ten years in the Salesforce ecosystem both here in Australia and in Europe. So really excited to present today. Yeah. Not with a mustache. I'll try my best. But, no. It's it's really exciting today to present some of this. I think that ultimately, the people that know me know that I'm passionate about the Salesforce platform. So when I talk through some of the examples and things today, I wanna make it really clear that, where I see some of the amazing capabilities coming forward from an AI perspective and from a, our core platform perspective here at Aircall, you know, are are a shared future between Salesforce as a platform and Aircall as a platform. So, obviously, some of you might already be customers that are on the call, and some of you might be considering Aircall as your core platform from the telephony side. But, ultimately, when I was a consultant, I used to recommend the platform time and time again to enhance what is essentially, an all in one system. So for a lot of people that are looking to maximize their investment in Salesforce itself, there's no better way than integrating, I guess, your other core systems, whatever they might be, but, specifically, Aircall in that regard. Having things automatically logged, I'm sure there's numerous sales teams that complain about logging things and various other people in your organizations that would like things to be more intuitive, and that's what drew me to Aircall as a platform originally. Since then, we've seen huge investments from Aircall into the AI space and specifically, becoming the market leaders in AI around voice and messaging across, numbers that we support. So in that regard, we we're endeavoring to become a leader in the space and endeavoring to fulfill all of our needs when it comes to what they desire, in the in in the medium and the channels that we support. So that's where, I guess, I joined, Aircall, and that's been, I think, almost four months now. It feels like a lifetime because we are releasing things all the time. The last even few weeks has been, hard to keep up with in terms of the number of features that we're shipping, and some really exciting ones are actually, available as of today. So we'll dive into a few of those in relation to the capabilities of our voice agents, how they interact with, the Salesforce platform, and where that I see from, I guess, my time as a Salesforce consultant, where I think it might show a lot of promise for some of the things that you're looking to do in bridging the gap between a awesome system of record and your customers in the real world. So just to, jump into this slide a little bit, over the last six months, there's been, you know, I would say that in the last few months alone, there's been tens of releases around our support for Salesforce. That involves our core platform for things like, you know, showing insight cards to your reps through the CTI and for the, inevitable transition to Sales Cloud Voice. But also from our our AI voice and messaging side, connectors into things like the Salesforce knowledge base to make it intuitive so that your agents can read the records and the system and the knowledge based articles that you've spent a long time maybe building. Additionally, being able to have intuitive actions inside your CRM, whether or not it's creating cases, looking at records and things like that, which we'll get into later today. But when it comes to, I guess, what's coming next, you know, there are lots of things on the road map still to come. We we are not stopping and definitely not slowing down. So everything from more customized voices right through to preset actions that your agents can take within your tools so that you're not constantly having to think of, I guess, every little bit. We we try and bridge the gap to help you deploy fast because agents are filling the gap for a lot of organizations that are essentially facing maybe economic headwinds and filling staff or or, I guess, in terms of, numerous examples of their being able to fulfill actions with inside of their own information capturing or in or enrichment inside their CRM with automation specifically through AI. So one of the big things that's literally landing today that has been coming for a long time is that our agents can now natively take actions with inside of Salesforce with what we call AI actions. And we'll explain and show some examples of what that that looks like today, but that means that it's only a one one click away to getting your voice agents who are talking to customers in real time, being able to retrieve records, create cases, or do various things inside of, inside of your Salesforce instance, without complicated setup. So when we actually what that actually will look like is a relatively simple process of if you are of, looking into voice agents to actually simply come into the interface inside of, Aircall and add the action from a drop down list that should support not just actions, with inside of Salesforce, but many more as well. So we'll dive a little bit more into this today. It's just a sneak peek, and we'll take this through this a little bit more, more methodically, but I'm I'm excited to announce that that is available and landing today for all of our air air core customers that are integrated with Salesforce. Additionally, we have opened up the ability to for you to connect your knowledge articles directly with agents as well. So no longer do you have to, I guess, in this case, point it at maybe live website pages or replicate content into AirCall's AI agents, but you now can collect connect them directly to knowledge base articles inside your Salesforce instance. So some of the applications that I've seen across our customer base, mean that you can share, maybe non publicly facing knowledge based articles with your agents that might support conversations that your agents are having with partners rather than outward facing customers, and it gives the grounding to the agent that, is unique and tailored to what you've already created inside your Salesforce instance. So these are just a couple of extensions of our product that have, become live in the last month or so, and we're looking to honestly get as much feedback from our customer base as possible to continue to improve these, throughout our platform. We always are looking for feedback, please. I'm I'm dead serious about that. Most of my time is spent with our engineering team overseas to to feed that knowledge and information back, And some of these exact product releases were the requests of some of our biggest customers in this region. So, yeah, really looking to hopefully work more to expand more and more into the platform. Outside of that, when it comes to building it ground up, I do wanna stop here for a second and talk a little bit about just the overall integration between AirCore and Salesforce. So we obviously have a foundational connection, which you might be used to from our core product for many, many years, which means that our platform's numbers are connected to your Salesforce instance to do all the things that you would expect, like logging calls or connecting the ability to make calls in and out of Salesforce, send messages, and all that stuff that you, is should be low hanging fruit for any calling platform. But what it does mean is that we have the capability of automating things like, case creation or lead creation from inbound, outbounds, voice mails, or various other calls. And the reason why that's important and often why I cover this with a lot of our existing customers is that the foundational integration is there and operating and our agents can act alongside it. So if every call to your support line is already creating a ticket, then your agent can reach in and now read that ticket and update it throughout the call. In or I mean, alternatively, a lot of organizations choose to let the agent create tickets at its discretion. But either way, what we're looking at doing is, basically working hand in hand with what we already have as a foundational integration to being able to store information inside Salesforce. And by the way, as much as Salesforce is the topic of today's webinar, our integration goes far beyond that. So if you have peripheral tools, like, for example, external databases, other external tools, we support a wide range of integrations with other platforms if you want to create actions in two platforms even on the same call. So plenty of our customers that have peripheral applications like different marketing automation tools, different, meeting tools like Calendly or things like that can do this, can do multiple things over the course of a call with their agents. So look up a lead inside Salesforce, go and book a meeting inside Calendly. As much or as little as you like to integrate with inside the tools that you have in Salesforce as well as your others, we can support those as well. Now AI actions, you can see here is the second tier up from the bottom on the screen. That's what I was talking about a little bit earlier, which allows you to pivot into the Salesforce platform and, at the click of a button, give your agent the superpowers of doing things inside Salesforce. Now some of you on the call might go, but we've spent a lot of time building customizations into our Salesforce instance. I know I did for a lot of customers in my time. So in that case, we have the next level that you see here, the API actions, which gives you the capability to integrate with custom objects, custom fields, and maybe even do things like integrate your data into data cloud, data lake objects, to bring that into data cloud's data model for analysis, segment creation, and inevitably, a stronger data environment for you to use for whatever purpose that you're looking to drive towards. And lastly, on the top here, I have some use cases. You know, I've seen a lot from the Salesforce side and the air core side. So today, at the end of towards the end of the presentation, I'd love to elaborate to give you some inspiration on what I've seen other customers doing that really shifts the needle to make an AI agent more than just an answering answering bot, but the one that works alongside your human agents or your teams to drive the outcomes that you need. So let's actually just pivot across into the Aircore platform very quickly to give you an idea of some of the just a bit of a walk through of the agent builder itself. So inside of Aircall, if you haven't been in here before, on the left hand side, you can see our numerous capabilities across the platform. Those involve number management, user and team management, conversational center monitoring. We even have a new tool called Ask Aircall, which is your own personal AI assistant in the platform to derive insights from wherever they exist. These tools aren't gonna be covered in-depth today, but, again, if you have questions about these and our ability to integrate, drive analytics, or any of the other tools in here, please book a meeting with your account manager or executive. They would, more than happy to cover these with you. But for day today, we're gonna be looking at our voice agent suite. Again, we do support voice and messaging, so messaging over WhatsApp and SMS. And both of those setups look relatively similar in the way that they're set up but do slightly vary. So for example, this is an inbound agent. The capabilities we have is also the same in our outbound agent. So I'll show you a flow today as well that actually gives you the example of triggering an outbound call with Salesforce because that's when it gets really exciting. So in here and our Salesforce demo, agent, on the left hand side, we have our configuration pane where we have getting started, teaching the agent what to know, controlling if it steps if humans step in, and then deciding what it should do. Now each of these steps itself, gives you the ability to configure the agent to become familiar with what that you are doing or producing this agent for. Our knowledge base can extend, like I said, into all the way into Salesforce's integration with your knowledge base articles as well as, extended retrieval augmented, retrieval mentioned so that you can obtain records at vast scales if you want to go beyond these character limits. The other thing as well is that there is an intuitive interface to allow to define to the agent when it should actually hand off. And one thing that I would like to start with here is that everything involved in building an agent is in plain English. So what that means is that when you are configuring it, there's no, extension or technical need when you are describing to it what it should do inside of its call script, as you can see here, or describing when it should hand off to people. All of it is written as a prompt. Now prompt is an interesting way to put this. I just really defer it as imagine what you would write in a procedural manual for the humans in your business that are doing maybe a similar task. That can be used here to describe to the agent exactly what it needs to do throughout the course of the call and even extending here if you've got a keen eye to creating those actions inside of your Salesforce instance. So here's an example of create underscore k Salesforce, and you can see just like the rest of the scripting here, it's written in in line and in sentences to make this as intuitive and easy as possible to tailor this agent to exactly what you need it to do. So it is a blank slate. And when you build it from the ground up, we have numerous examples that our account management team can pass across to you in different thing in different scenarios. So whether or not that's an out of hours agent, one that collects CSAT surveys, one that qualifies inbound leads, they can all be tailored with simply writing your conversational guidelines into the call scripts. And then from there, like we mentioned before, down the functional abilities is where that we tap into Salesforce, like I showed previously, with adding our integrations into things like, finding customers, maybe up, inserting or updating records inside Salesforce, creating cases, and a full list of these can be provided after today as well. As I mentioned before, the last step is our custom capabilities, where if you really wanna go one step further, you have the opportunity to even dive in and trigger things with custom, with custom actions to tap into those custom objects and fields in your instance, with ease with these API actions of which Salesforce in the has an incredibly broad set of APIs to drive from. So if you have somebody who's tech technical and part of your team, basically, the opportunity is limitless with being able to retrieve, edit, change, or look up records inside of your instance. And then lastly, we'll dive in a little bit more deeper today to talk about what did the agents provide back to you. So if you're wondering how can we improve these or when will I be alerted, well, the agent will tell you when it has gaps in its knowledge, if it's being asked questions it can't answer, or if your configuration is not providing the outcome the callers want. That's fed back to you in real time and gives you suggestions as tooltips over the course of your use as well as providing all of that information in Ask Aircall. So if you wanted to know why are most people, being transferred or maybe why, which particular phonetics need to be changed, Ask Aircall is the next step to answering those questions alongside the individual agent performance. So that's just a bit of an introduction, but let's just step back here to to kind of follow on. Now one thing that I always like to mention here as well is that numbers themselves can have individual configurations. A lot of people that I've worked with in the Salesforce world would have the mentality of a system is integrated to a system. Well, inside of Aircall, it's a number that is integrated in your and each number in your organization can be integrated in different ways. So for example, your support number can be used to create tickets on every inbound call, but for your sales line, maybe that's leads instead. So expand your thinking in the way that the foundational integration between AirCall and Salesforce can be tailored per number rather than just a straight platform to platform integration. And when it comes to the agent lay it on top of that integration, it can then enhance either the lead that was created or the case or various other things alongside of that. So when it comes to, a lot of the things that we see with some of these, integrations between our platform from an AI sense, a lot of the time, what that companies are driving towards is autonomous, I guess, completion of calls or autonomous, qualification in various sense. And often what we're seeing is that there is the ability for agents to step in to really kind of clean out the rats and mice of maybe low hanging fruit tier one support questions. So the little things that an agent can do before it transfers to a person. Some of our customers in region have the ability to really do their best to, give the agents actions to to kind of assist, not overtake their internal staff. So one of the reference examples I can give as well is that one of, a really successful deployment in region, they have an agent in front of their support team that handles 20 to 25% of their inbound support questions, which means that they don't miss any calls from their support team anymore. And every call gets a case created. If the agent resolves it, it updates it itself as resolved so that all of the amazing reporting that you have under the hood inside Salesforce can be enriched by the actions that the agent is taking as well. So when it comes to the actual conversation itself, it can use all of those handy tools from the, from the API world of getting, posting, putting, patching, and deleting. And when it comes to the conversation, they're triggered off in line. So the I did see a question earlier on, in the call here, like, what if it's, different platforms or external ones? As you can kinda see here, as the agent is having the conversation with somebody, what it's doing is being able to call each of those individual actions that it has access to and completing them throughout the call itself. So it is sequential, and it gives each agent the ability to have its own unique set of tools for its own unique task. So lastly, what's kind of, what's kind of working and and proving really successful? So I would like to give you some examples of what I've seen across the Salesforce ecosystem. So when it comes to things with inside of custom objects, the API actions are proving really valuable in having that ability to dive deep into any object. And then as cross into some of the AI, AI actions that we're releasing, the beta customers that were part of that initial beta release are finding it very easy to deploy quickly. So without having to do, too much to configure those agents, their the speed to being able to deploy agents is now turning into days rather than months or, weeks or months. So when it comes to when when it comes to deploying agents in plain text with actions, it really does give you the opportunity to start addressing maybe some areas in the business across that, maybe really low hanging fruit. So I really wanna give some examples here and show some additional stuff inside of Salesforce, to give you maybe some inspiration as to where that this might land. So one of the most, used areas of our agents is for most businesses, after hours conversations. Most organizations will implement something like a voice mail, maybe out of hours. That's a perfect opportunity to surprise to surprise and delight customers. In that sense, having that after hours agent to create a ticket or a lead or maybe send a text message to an external team member that is not part of the head office as the head office is closed. You know, all of those are great examples of where an after hours agent can perform tasks without maybe sending someone straight to a voice mail, reducing those repeat calls because people think that they might not be getting through and whatnot. Another really common one that we're seeing, especially with Salesforce commerce customers is qualifying leads. Now I know that there's a lot of hot topic around the SDR agents and AI and whatnot, But, ultimately, some of the biggest success that we've seen are organizations prioritizing the qualification with agents in lower value leads. I probably spent a year of my life as an SDR at Salesforce. And during that time, there are there's people that wanna be contacted and have a demo and very high, like, very high intent. And then there are those people that filled in a PDF on a Facebook advertising campaign. So maybe you're if you have a long lead queue or you're struggling to define what should be prioritized, leaving the lower value stuff or the lower conversion things and giving an attempt at least within a AI agent first, gives at least the opportunity for you to prioritize the important things with the people on the floor and the human agents in your team. And then everything one of the last things as well is following up calls and then closing the loop with things like CSAT surveys is another one that we're seeing, I guess, quite commonly rollout. Agents performing that action based upon something that fires off inside of Salesforce to make that outbound call to understand, you know, whether or not and collect the feedback in an interesting way. So one of the powers, I guess, I've seen from a large language, a large language models that underlie our technology is that we might all be used to CSAT surveys that go, give us a rating of one to 10 or give us a rating of one to five. Ultimately, we as, I guess, businesses have defined that as it has to be between one to five to make it measurable for us. But if the agent can listen, understand, digest, and then give you the sentiment back, that becomes a much more powerful conversation and survey. And instead of just a one to five metric, you might get qualitative feedback from an element that can understand a human pride feedback in terms of sentiment and things like that so that you can evolve the way that you're kinda asking people for feedback rather than tying them down to a number between one and ten. So all of those are interesting examples. And just to demonstrate them a little bit as well, I do wanna step into, one flow that I have here. What we have I guess, what we can have the capability of doing is triggering off some of these things to happen right within your phone environment. If anybody's created an outbound action inside of flow, the, the the advantage that we have as well is that any of your information from within inside Salesforce can be passed across in that payload. So if you're triggering a call from an agent to call out for, say, a survey, you can include things whether or not they're one of your top tier customers, what they purchase, maybe maybe what service that they were provided from you so that the agent can have a warmer conversation with them than just a generic survey. So when you're passing things from, say, a lead, it could include all the fields in lead, some of them, anything that you would like to include from inside Salesforce or retrieve inside of a flow. Initiating that call with context gives the agent a lot more of an ability to to bring that up and work with that on customers. The other thing as well is that bringing that information back in can be brought in and added to your records, through a flow as well. So as much as the agents can take actions on their own, you may also not want them to. That's completely fine. And what you can then do is actually digest the information that comes back from the agent and process it down a deterministic automation to place the data exactly where you want or change it or, I guess, adulterated in a way that would make sense so that your, your core data remains safe behind a the due diligence of something like an automated flow. So, hopefully, this has kinda give you giving you some examples of where that, where that we've seen Aircall agents slot into businesses really quickly, the out of hours or the overflow agent to take phone calls ahead of agents to reduce, you know, the simple questions or the easy qualifications, book meetings on behalf of, of customers looking to find time with your your reps, being able to bring in and then go out and retrieve information from customers whether or not a lot of service based industry might be confirming appointments on-site so that people that are going out to to maybe on-site calls and things like that don't waste their time if people are looking to book or change. All of those things that I've spent a lot of time in my past programming through a lot of, you know, deterministic workflows through Flow or Marketing Cloud, all those kinds of things become a bit more intuitive now because there's an element behind the scenes digesting the information, understanding the customer, and deciding when to input back information or helping them assist with changing things inside of your system of record. So, yeah, I mean, sometimes this can be scary. I know that it's obviously a scenario where it's like, I don't wanna do those things. The good thing is is that the agents under the hood will only perform the actions and stick to the guidelines that you give it. If there's any times where it's missing information or it's missing configuration, the good news is that it provides that back to you in metrics, and it gives you the opportunity to know to spot the gaps and know where to change things so that you can do that and pivot on the fly. Most of the times when I've worked with customers to deploy agents as well, it normally takes maybe one to three iterations, and then it often becomes a set and forget, especially if it is an after hours agent or receptionist that is handling calls. And then, ultimately, as the platform evolves, giving it the capabilities that might give it more responsibility, but also know how it can reliably manage and, and provide service to your customers on the front line. Outside of that, hopefully, that's given you some insights, but it is often a situation now where, having the ability to ask questions and to dive into your unique examples and use cases, can be more effective than just, I guess, what we're going through today. So I do wanna open it up the opportunity for Will to take questions and to ask about some things that we've touched on today or reiterate and go through examples that, you know, that you might be looking feedback for. You're on mute, Will. Great start. Thanks. Trent. We actually have a few questions here that cover a few different topics. So I'll start from the top. Do AI actions fire mid call or post call? And does the caller hear any lag while it happens? That's a good question. So, depending upon what that you're looking to go. So let me ask in the first one first. They can happen in both. So we have the opportunity for them to fire as the call is picked up so the agent has context as it's greeting you behind the scenes. We have the opportunity for the actions to be called midway through the conversation by asking the agent to call them when the person is interacting with them. And then lastly, if you would like to finish the call with actions that's capable as well, you know, the agent can fire off calls as it's hanging out in the call. There's toggles inside of the configuration to allow you to program in things what we call, sorry. It's good. The name will come to me, but, essentially, the ability for the agent to say, let me look that up for you if it's complex action. So for example, there are a few customers that have, maybe field service or that have complicated assignment rules or, case assignment. If that the agent assigned something and it needs to wait a moment or two, it can use what's called filler words to define. Let me check that for you or whatnot. But ultimately, we're moving to an environment now and seeing huge improvements from our engineering team to perform these actions very quickly. Most of the time, it's actually Salesforce is the one that we have to make keep an eye on for latency. So in that regard, it can be a bit of a case by case basis, but most actions like updating, looking up records, and things like that happen very, very quickly around that three to four hundred millisecond mark, which means that the conversation can flow as per normal. Okay. Great. Steven has a question around outbound. How does it work? Is it the same as discussed here? Yeah. So the outbound configuration is basically identical to the inbound information. The only difference is how that you initiate the outbound call. So the outbound agent is initiated by, in Salesforce's case, a HTTP post, which is a standard tool inside of your flow or Apex. So in that regard, we actually have a full written guide on how to initiate those. If you'd love to get in contact with me after today's call, I can forward that guide onto you as well as anybody else who would like it. Please indicate what you would like, and we can forward the PDF on how exactly to configure that and what options you have available with inside of Flow. And then from Gemma, in this scenario, there are no agents available to take the call, and the AI agent is transferring, will it go to the voice mail, or will there be a case lodged in Salesforce for a callback? Completely up to you. Our SmartFlow editor has the ability to put agents ahead of your human agents or agents after your human agents or maybe even both. In that regard, if it transfers to your agents, which are humans from an AI agent, the agents themselves, if they miss the call, it can then be transferred back to an AI agent. So this is one thing as well with inside of the Echo platform, and I know for most Salesforce customers, this is not the usual. But we try and make it really simple when it comes to buying our agents from a commercial lens. It doesn't matter if you have 50 agents making one minute of calls or one agent making fifty minutes of calls. All that we care about is how many minutes your agents have spoken for. So you can have them placed as many where in many places throughout your call flows as you'd like. We just simply bill on minutes here. No charge for the underlying model, no charge for API calls themselves. It's just minutes. So if you would like one ahead of your agents and one as an overflow, if they missed the calls, absolutely possible. And the great news is, is behind the scenes inside of those workflows, agents talk to each other. So if an agent transfers to another agent, it actually behind the scenes goes, Hey, this person was asking about this or the person wanted that. And in that case, the second agent can pick up the conversation and go, Oh, hey, I know that you're asking about this. I know that we couldn't get you on to a person, but can I take your information to log in as a ticket? And I'd say, trend, we're working with a lot of customers now who started with the one agent for those more basic use cases. And as that's developed, they've started to add more and more agents, into their workflow. Yeah. It's a really good point. You know? Getting, getting off the the monkey off the shoulder per se and and getting your your feet in finding your way through the tool is always, you know, the best place to start. You know, once the what we saw as the number one use case here was the out of hours agent. Most organizations, if you're missing calls out of hours, then it's pretty low risk to kinda get an AI agent in there at least to try and get it to give an answer to somebody or log things in the way that are gonna be way more helpful for your team inside of Salesforce. That's a great place to start and a great place to build confidence with what is a pretty new technology. So I can completely understand that, if you're dipping your toe in here, start there and then move into maybe other use cases as you gain confidence, or maybe you have the right integration set up with Salesforce to stage the data that maybe the agent is collecting. A good follow-up question we have here, which I'm happy to answer, is what's a realistic first go live look like, and how long does it take? What would be ideal from here is if you did book a meeting with the account management team. From there, we would do a preliminary discovery to understand what the use case is, what the requirements of these agents are, and then you would work with us in getting that sent set up, and then working with Trent as that, developed, from a from an agent perspective. But, typically, we can move as fast as you can move. It can take a week. It can take days depending on what your use case is. So it's quite an easy turnaround depending on what you what what your requirements are. Most of the most of the engagements that I'm dealing with with customers that have integrations into Salesforce normally start with a discovery workshop of exactly what your use case is, defining what success looks like and the aims of the agent, as well as defining which knowledge that the agent should hold and how it should conduct calls. Most of the time, it takes about an hour or two to workshop with yourselves, and the initial deployments can be rolled out within the first week. Most of that next stage is iterating because nobody ever gets it right first the first time around. I know that we're all probably still working on Salesforce subjects as we speak. So in that regard, you know, doing that iteration process and reiterating the agent once or twice, most engagements with with myself, I'd say, would last about anywhere from two to four weeks. Are there AI actions that can reach into Shopify and also Salesforce in a single call, such as get. the order from Shopify and then create a Salesforce case. Yeah. That is absolutely possible. We actually support an out of the box AI action for Shopify to do that where is my order, check as well, or getting the order information. So we're we're enhancing that, that investment into connecting more natively with Shopify as well. In that regard, we also have the capability of using those API actions for Shopify. There's been a few retail customers that have asked me to create things like, for example, checking stock levels so I can provide information back to customers and various, checking stock at locations. That's been some fun things that we've done. But the absolutely thing is, yes, it can go and check an order once that it's, picks up the call with a person by looking up the corresponding phone number or going through the process of asking what their order number might be. And then from there, if it's, if the agent deterministically can see that the order is maybe far ordered in the past or the person would like a follow-up, it can then pivot and make a case in Salesforce. Absolutely possible. Great. And then a follow-up to that is, does the agent have the ability to send a follow-up text or email? So this is an interesting one. So the answer is yes. It can, and it can do it mid call as well. So we actually have a custom a couple of customers doing this right now where that when a conversation goes in a particular direction, the agent can text them something. So for example, there's one customer right now that has a, a well built out knowledge base, but a lot of their knowledge base articles don't have individual steps. They actually have videos on them. So when the agent retrieves that knowledge base, article and then says, oh, I have some help with you in this regard, but it's in the format of a video. Would you like me to text it to the phone number you're calling in right now? It can text that link to that knowledge base video live in that exact moment. Alternatively, you can, of course, on the outcome of the call, either tag it or do something like that, and your flow can then follow-up by triggering a SMS out of the air call platform following the call as well. So there's definitely possible. Your agent can initiate it, or you can do it downstream with a flow, back through the air call platform. So it's to the same person. So this actually raises a good point, and I don't wanna elaborate too much on this. But if you have, a good understanding of some of the automation capabilities of Salesforce, they are incredible. Flow is an amazing tool, and Apex is, is is such a diverse set of tools as well, as well as what you have inside of data cloud or even some of the agent for suite. In that regard, empowering your AI agent inside of Aircall with downstream automations, having them work together is where we see Salesforce customers make an enormous amount of success from their deployments. It means that you have the ability to, to kind of have one system dependably follow process and the other one patch the gaps through a nondeterministic flow. So extending Salesforce to have a voice is kind of what we're talking about, and that can happen outbound and inbound to, power, I guess, what that you're looking to from an automation perspective. Great. Do you have live use case examples that are able to be called to experience and understand? We do. We have a few different numbers that you can call to go through that experience based on your specific use case. Derek, we can follow-up with you, from account management with those examples as well. I know Derek quite well. I'll I'll make sure to reach out to you, Derek. If do do you share with customers what little updates are being rolled out by your end team, so we can stay abreast and make updates to our calls script accordingly? Again, I'll take this one. That's a great point. We do. I think, the velocity in which these enhancements are coming out are quite quickly, are coming out quite quick. So, the account management team, will be sharing those on a more, rolling basis so you can understand what's being updated and how can you can adapt your existing agents based on those, amendments. Oh, this is an interesting one, Trent. Could we make a call to a customer with the name of a business instead of a phone number? So the answer is yes. I would love if that person asked a question to maybe clarify a little bit more exactly what you might intend to do in the chat so that Will can maybe retrieve it. But, if you are looking to make an outbound call, when that the outbound call is initiated, you can send the agent any information that you have inside of Salesforce, basically. So, for example, if the call is initiated and you've stored in a company name, inside of your call script, a couple of curly brackets, essentially inputs that variable into the call script. So the agent, when it, it commences its conversation, it says, hey. I'm just calling I'm just trying to get on to x company. I just wanted to confirm this is the right number for it. We do have a couple of customers that perform some data cleansing tasks with, with their flows. So that calls out and confirms contacts and businesses, and it does that by asking if they're the IT contact and things like that. So, yeah, I would encourage, you know, due diligence with use cases like that. But at the same time, any any field from any object inside of Salesforce can be added to your call scripts, to to empower the conversation a bit more relevantly. So the answer is yes. I hope that answers your question. Whoever asked that. And then in the last couple of minutes, this has been asked a few times. It's about the, the voices, the customization of them, and how we deal with certain accents, be it Australian, New Zealand, non native English speakers. Yeah. So we have a bunch of out of the box accents available and, in there including Australian English, of course. We are extending that now, so I hope to very much soon have the ability of a few, New Zealand accents out of the box in our platform. However, we do have the ability to customize the voice with custom voices. So we have partnered with a number of, of leaders in the space from text to speech. We have some customers that have custom recorded their own voices and hosted them on those platforms so that we can add them to the agents. So it's actually quite an effective tool to be able to do that because some of the customers that we work with have very specific, industry terms or if they're or hospitality terms and things like that, for their ingredients, their services, or the products that they have, and their customer recordings pronounce them perfectly. So in that regard, we can absolutely support that. Yeah. Just please let me know who whoever, asked that question, if you wanna get in touch with your account manager, I can we can definitely walk through the process of adding that voice if you'd like to or that accent. Yeah. Thanks, Trent. We have a few more questions that we won't be able to get to today, but what the what we will do after this is the account management team will reach out to you with those answers. Again, it would be great if you can book a meeting with the account management team so we can continue the momentum of discovering what can be done with these agents, and getting them to be deployed and go live, based on your requirements. But, again, thank you everyone for attending. These are webinars that we're we're looking to host every month based on specific integrations, verticals of industries. And, Trent, thank you again for presenting. No worries. Have a good afternoon, everyone. Thank you so much for all your questions, and looking forward to connecting with some peeps and friendly faces in the chat. Thank you so much. Thank you.