Lierin: (00:03)
Good morning and afternoon everybody, and welcome to our webinar. today Cornerstone is hosting SkyLink. so this is actually going to be really exciting. SkyLink is an artificial intelligence and automation, company. And, we’re, we’ve got a lot of exciting things to show you with, with what SkyLink is doing and how we’re integrating with Cornerstone technology.
So this webinar is going to be recorded. I’m recording it right now, and, everybody who is here will be getting the recording after this show. but we are going to give it a little bit of time. I see that we still have a lot of people coming in. We’re climbing at a pretty good rate. so Mat and Atyab, if we can hold tight for another minute, I think we should be ready to go.

Mat: (00:53)
Absolutely. No problem at all.

Lierin: (00:57)
Excellent. For everybody in right now too, I’ll let that there is a chat feature at the bottom of your options for the webinar here. You should be able to go into the chat box and any question you type in there will go straight to me and I’ll make sure it gets to our panelists throughout the webinar. So some of the questions we’ll be holding to the end, but if there’s something that you have, really pertinent to what’s being discussed at the moment, feel free to just shoot me the question over there and we’ll get it answered for you. We’re still climbing here, but it looks like we might be topping out.

Mat: (01:44)
Yeah, it’s good, good turnout today,

Lierin: (01:47)
Definitely. I’m always excited to see a good amount of people come out to see what we’ve got new going on.

Mat: (01:54)
Especially for the lazy days of summertime and where people are running off getting off the beach and cooling off in their room listening to a webinar from ours.

Lierin: (02:06)
Well, I’ll tell you what, it’s raining where I am, so I don’t know. This webinar might be a bit of a productive use of this time for right now, in this weather anyway. what, it looks like we’re pretty stable, so I’m going to say we can get going. For everybody that’s joined recently, let me go ahead and reiterate the webinar is being recorded. So you will get the recording of this afterwards. And there’s a chat box in your options over there. If you have any questions at any point during the webinar, go ahead and shoot me the question. it will probably be held to the end unless, you’ve got something really pertinent to what’s being discussed at which point I’ll bring it up to the panelists at that point and we’ll go ahead and answer it then.
So yeah, let me go ahead and introduce: we’ve got Cornerstone and SkyLink. Cornerstone, CEO Mat Orrego here, and SkyLink CEO Atyab BhAtyab here. And we’re going to be discussing a lot of really exciting artificial intelligence automations that SkyLink is going to be bringing to the travel industry. Mat, I’ll let you go ahead and take it away.

Mat: (03:18)
Alrighty. Thank you. Lierin. So, hey, welcome everyone. Welcome, Cornerstone customers and friends and colleagues from, all across the travel industry to really our, our kind of ongoing series now of webinars, that are introducing, some really important partnerships for Cornerstone and collaborations with some really exciting companies that I think are are wonderful opportunities to help our customers not only leverage Cornerstone technology, but , innovative tools that are emerging, to help, both travel operations and, travel procurement programs. So today we’ve got SkyLink with us and their co-founder and, and CEO, Atyab. Our conversation today is how we intend on really increasing agency productivity and the productivity across the selling platform of agencies and how we, at the same time, we can increase customer satisfaction through AI and automation. So welcome Atyab, good to have you with us today.

Atyab: (04:30)
Thank you, Mat. I appreciate you. And folks, it’s great to great to see everyone who’s attending and thanks for taking the time out of your day. As a brief introduction to myself and the team: My background primarily comes from being a traveler myself as a consultant at McKinsey, where, where I focused a lot of my time on business digitization, op models, and corporate finance, particularly with the travel industry, across suppliers, travel management organizations, and then also across the infrastructure players. And SkyLink is really nothing without its team. so I’d like to also give mention to Kush our, co-founder and CTO of SkyLink. We’re a part of Y Combinator, the technology accelerator that created organizations like Airbnb, Coinbase, Stripe, Reddit, et cetera. And then we also have on our team, Doug Anderson, who is a senior advisor to the organization, former CEO of MSGBT and CWT.

Atyab: (05:25)
But without further ado, I’d love to jump right into it. So one of the things that, as Mat mentioned, it was it SkyLink and Cornerstone kind of hit two premises of the travel management industry, where Cornerstone has done a phenomenal job with mid-office automation, making sure your quality control is, their travel policies are enforced, they’ve effectively turbocharged the travel agent from these bid office and back office tasks. And then there’s that natural, kind of inclination of how can we drive a similar level of automation on the front office. And, historically it’s, it’s been quite challenging to do this largely driven by the prevailing technology that’s existed. AI and machine learning have been buzzwords for the past 10 years, but as I’m sure everyone here has seen, Chat GPT has made more than a splash when it comes to actually redefining what artificial intelligence is and leveraging that kind of technology, you can effectively start to drive optimizations that we could never see before is particularly across the front office, automating things like your booking workflows and your non booking workflows, but doing it in a way that is actually meeting the travelers where they operate today.

Atyab: (06:36)
And what we mean by that is targeting the areas of like, where these travelers communicate both in their enterprise chat channels like Slack and Teams, but then really also in in their email as well. And we’re doing this in a way that’s incredibly comprehensive, focusing on things like security, making sure we have SOC two compliance, which is the US enterprise grade standard. And then on top of that, managing data privacy, making sure everything meets the standard enterprise requirements so that we can serve your kinds of organizations in a way that is not only compliant, but to distinctive in the sense that it allows you to deliver best in class experiences to your travelers and to your agents. So, broadly speaking, one of the things that really drove the genesis of skylink was covid. we, while, while, while we saw corporate travel pre 2019 be this $1.4 trillion industry, it effectively got turned on its head overnight, a 71% market contraction driven by the pandemic.

Atyab: (07:40)
But what did that mean for organizations like TMCs and corporate travelers? Well, really, it was simple. TMC agents were forced to actually comply with policies oftentimes not being direct, reflected in the GDS while still managing the standard processes of getting travelers from their origin to their destination and back. But what was interesting to us really started what was the rebound on corporate travel. And what we really started to see there, we spoke with several large travel management organizations, leaderships. And what we found was that in the post covid environment, the rate of acceleration of the return of corporate travel has been faster than anything we anticipated. But with that, there’s also been talent shortages. Agents are not a commodity. They’re actually quite difficult to get. The average agent is 52 years old, oftentimes, many of them have actually retired.

Atyab: (08:35)
Several of them switched jobs during the covid environment when furloughs were happening. And as a result of this, these shortages have spread talent thin. So how do you actually combat this? Well, we took a look into the numbers because the data speaks the truth. What we found was that across the average TMC, the average agent is doing roughly 1.2 bookings per hour. Now, when you comp this back with their fully loaded cost, you’re looking at anywhere from $20 to $30 per hour, an hourly rate. Now, these agents are doing a fantastic job delivering the value to your corporate traveler. But the challenge faced is that their attention gets spread thin as a result of these shortages. And as a result of the increased rebound on corporate travel, enabling your talent to actually maintain that distinctive human touch that really drives value to your end corporate traveler, but still allowing them to be more productive and supported is becoming increasingly more critical.

Atyab: (09:36)
So how did we actually address this problem? Well, we defined it as twofold. One was that we recognized that the inbound workflow for a traveler comes from two primary venues. The first is actually email. Now, every traveler loves to send an email to their EA, and the EA might oftentimes pinging the travel agent to actually execute the transaction. And on the other side, occasionally the traveler will directly email the TMC. Now, while OBT adoption rates have been anywhere between 40 to 70%, depending on the industry and organization that that inbound email flow consistently shows itself as a lever that travelers prefer. But what’s interesting, beyond that, the coordination of actually sending that email is happening on the chat side. folks are coordinating with their colleagues, pinging them, letting them know, I need to be in location A and location B, and by intercepting the workflows right where they’re happening, we can drive actual, we can, we can naturally intercept that, that conversation and help drive down email volume as well.

Atyab: (10:40)
So taking this bifurcated approach of enabling the digital chat agent inside the Teams and Slack environment while supporting the email functionality provides this increased productivity when it comes to actually servicing your clients in a way that you could never before. And that’s largely enabling higher availability, lower wait times, and faster resolution. So what does this look like? Well, on the first hand, we have SkyLink Chat. SkyLink Chat can be used by agents or travelers to actually do rapid research within your systems. You can talk with SkyLink, just like you would with a travel agent giving complex inputs of, I need to go from these three cities and come back at any given point. But what’s interesting with this is that it can respond back in a matter of seconds. Tasks that might take anywhere from 15 minutes for a simple trip to 30 minutes for a more complex one can now be handled in under two minutes, total end to end.

Atyab: (11:35)
Now, what’s powerful about this as well is the way that Skyline can be configured. It can serve up suggested options that are aligned to your business priorities. So what does that mean? Well, that means that if you have a preferred supplier or your, your, your customers are looking to drive cost savings and lower cost options, you can start solving for these particular scenarios and serving up the results to the traveler directly in a programmatic fashion. And I encourage you to actually check the, check the OBT data of your clients and compare it to the agent book data of your clients. And what you’ll find is that by actually providing curated options and giving visibility to the end traveler, you actually nudge their behavior towards lower cost options and eventually also green options as well. But as mentioned, chat isn’t the only resolution, really.

Atyab: (12:25)
It’s about showing what’s the power of email as well to be able to flexibly understand complex inputs and create options rapidly in a way that we can never do before, can drive, can, it can effectively eliminate the research component an agent has to do when actually doing a booking for a traveler. And as a result of this, now, you can programmatically detect details when they’re missing. You have higher quality control given the fact that you have a machine really driving consistent behavior across the organization. And on top of this, you can customize it on a per client basis, but I think it’s one thing to talk about it. I’d love to give everyone a little bit of a demo here. So if we pull up this slack right here, this is a regular Slack channel, and we’re going to talk with SkyLink. We’re going to say a relatively complex trip.

Atyab: (13:12)
We’ll do, Hey, I need a flight from London to New York on August 1st, then NYC to SFO. So we’ll switch from City Id to airport Id now on August 5th, and then San Francisco to Los Angeles on August 15th. So keep in mind, we’re intentionally making errors here. We wanna have spelling mistakes, different types of way of conversing, just to show that flexible human level understanding is incredibly important when folks are communicating. And what’s happening on the backend here is SkyLink is actually understanding your request and presenting you with options. And so if you look here, we sent this at 9:14, we got a response back within, I believe roughly 12 seconds now, London to New York. And these are real flights. You can actually click and expand and look this up as well. But what’s interesting about this is that these flights are roughly $16 to $1,700.

Atyab: (14:17)
But look at what’s right here, this behavioral nudging now, right now what we’re doing is we’re calling out, it’s a cheaper flight, but we can customize this to actually just say, here are your three options without insinuating it’s cheaper or it’s a green flight or the preferred supplier. But what’s powerful about this is that if you show these three curated options, you can fundamentally drive behavior in a way that you wouldn’t be able to be before. And so if we go here and we click accept on this JetBlue flight, what’s going to happen is SkyLink will then revert this back very similar to like an OBT workflow and allow the traveler to book it based on their profiles and their credentials as well. Now, what you’ll find is that it jumps to that next leg of the journey that’s at New York to SFO flight. Here we have a JetBlue and, options, two JetBlue options available. But say you don’t fly JetBlue. Say you want to fly United and you’re an Alliance member, well, you can converse with SkyLink just like you would with the agent and say, do you have anything from United?

Atyab: (15:17)
And what’s really exciting about this is you can change the airline the time you’re departing the dates these, the city, the particular airport, and have this very natural conversation across the different things you might want in your flight path, just like you would with a travel agent, right, directly in your current chat channels. And so you’ll see here we have the New York to SFO flight. We have two United options, and again, that third cheaper option, at 4:30 PM And so if we expand this out, it’s an economy cabin. You can change this to business if you like. and what’s interesting is that once you accept this, it, it interprets what you’re asking and, and goes to the next leg. The great thing about SkyLink as well is that it can programmatically enforce your policy. What that means is that if something is out, out of policy, Skylin can present the particular option to you, but then also flag that, hey, you can’t actually book this without approvals.

Atyab: (16:13)
And alternatively can be customized to simply not present out of policy options as well. And so if you look here, we have that SFO flight, and one of the things I wanna call out is how we had egregiously, misspelled and miscommunicated what we were actually asking, which was the San Francisco, Los Angeles, like making all these errors, right inside the chat channel. But SkyLink is able to understand it dynamically and provide you with the options here. And again, you can ask for a later time, change the airport, you can switch at request a business classify, and if that flight is at a policy, have it get flagged as well. And just like that, you’ve been able to book a multi-city trip from London to New York, to l to SFO, to LA all within a matter of roughly, if we look at the timestamp here, four minutes.

Atyab: (17:01)
Now, we know that TMCs don’t just need chat. They get a tremendous amount of email volume as well. But what happens when you can programmatically sort through your emails and for resolutions that Skyline can actually take care of, handle them automatically, and the ones that the skyline can’t handle, push them to the agent. Well, that fundamentally changes the way that you operate. It allows you to become increasingly more productive. And so what you’ll see here is this relatively complex flight. Now, depending on the tenure of the agent that the capacity they might have, this kind of request typically takes anywhere from 25 to 35 minutes. Because it’s multi-city. Now, if this was simpler, it would be roughly 10 to five minutes. Now, we’ve actually gone ahead and sat with your, with agents across travel management agencies and found that roughly 65 to 75% of the time these agents are spending is actually on the research component of figuring out what options to actually present the traveler with.

Atyab: (18:01)
And that that drives majority of the time that they’re actually spending. And the actual booking component is, is quite quick for these organizations. And so right now, we sent this out at 9:17, and if we open up the response 9:18, we have all the options listed out in the email right here. Now just like chat, you can actually respond back and customize what what you’re looking for. And again, what we’re showing you here is all the airlines available, but we can filter, we can make it so that, not only are we showing you the airlines, but we actually request whether or not you need hotels, in, in your itinerary as well. And we’re detecting different aspects of your trip. Like one, your preferences, you’re not asking for hotels, but you’re staying multiple days asking for these particular questions preemptively to drive streamlined operations.

Atyab: (18:56)
So now what we say is I’ll take option A and option E, do you have anything via anything leaving around 4:00 PM for flight three? So we’ll send this out and again, you’ll be able to have this end-to-end email conversation with SkyLink directly. What’s powerful about this is that SkyLink doesn’t even need to communicate directly with the end traveler. It can actually intercept the request and communicate for the agent as well, where the agent can just chat with SkyLink either in the email function or in the chat function and say, here are the three places I need to go. Can you list out all the options? And then by simply just copy pasting all of a sudden that 35 minute workflow is taking two to three minutes to get done. And that is incredibly powerful. And so we come here, there we go. We ask for that 4:00 PM flight. We got a 4:00 PM, 4:10, and then obviously the low cost option being shown here, which is more than, 50% less. So we’ll go and say, I’ll take Option I.

Atyab: (20:09)
And so once we kick this out, styling just like this will run a standard backend OBT like flow that allows the actual booking to get executed as well, or leave it up to the agent to actually execute this in the event. There are certain externalities that we need to take into account because what we do recognize fully is that the travel agent can never be replaced because they know things about the corporate traveler that may not be documented or something that a computer cannot interpret. And as a result of this, the traveler can take over, actually execute the end booking. But all the homework that needed to go into this is being taken care of by Skyline right here. And so there we go, record locator travel booking confirmation. Now, one of the things that we pay close attention to is we can customize the way that this information and the UX is laid out, to your particular organization. So that all that is all the workflows that SkyLink is integrating with are very representative of how the agency does them today. And so now jumping back to the presentation here, and if there are any, go ahead.

Lierin: (21:16)
Just one question, if you don’t mind, came in asking: If this would work with Microsoft Teams as well as Slack?

Atyab: (21:23)
Yes. So the chat functionality can sit inside any of the major chat enterprise systems, so Slack and Teams can easily be supported. And we have Slack today and we also have Teams in the pipeline. The, the representation of the visuals look a little bit different ’cause Teams may Teams as you, you like interface looks different than Slack, but it, the, the end-to-end workflow looks very similar.

Lierin: (21:49)
Thank you.

Atyab: (21:50)
Hopefully that answers questions, right? But one of the great things that Skyline can do just beyond these kinds of workflows is it’s non booking workflows. If given access to your backend systems like Cornerstone System, we can pull certain information as it’s mapped to the travelers. Things like info on your existing bookings, itineraries, invoices, e receipts, things that your agents might get, message for today. we can actually pull that information directly from your backend system, either serve it up to your agent or to the traveler themselves in a way that they wouldn’t have, in a way that wouldn’t require human intervention. And that is, that is the power of when you sync up your front end operations with your back and mid-office and back office operation is you get this full end-to-end integrated workflow. Now, it is important to recognize that while Skyline can do quite a bit, the agent is always in the loop.

Atyab: (22:47)
And so now it’s, you can think of it as an analog, an analogous to getting your iPhone the first time. people, always had access to books, they had access to their laptops, but when they, when they got the iPhone, they had access twenty four seven even while they’re walking on the sidewalk to the global breadth of information. Now, what that means is that at the end of the day, people are still doing a substantial portion of the work, but solutions like SkyLink and, and generally tech, these iterative technologies that have come out just make these people, these agents far more productive than ever before. And to sum it up simply, I think really what this drives is three main things. It’s improved client satisfaction. You can detect things that are missing. You can serve them much faster.

Atyab: (23:35)
Instead of having that two hour back and forth between the agent or, a 30 minute call that’s taking the traveler’s time away, they can, they can actually just communicate directly with the SkyLinks digital agent right inside their pocket. On top of this, this drives agent productivity by obfuscating away these relatively complex, relatively simple or straightforward workflows that take quite a bit of time, agents can focus on the things that Mater most. An example I think that’s great of this is, what a calling the airlines is something that only an agent can do to, to kind of handle some of these more complex requests. And on top of that, the financial benefit, now all of a sudden you can reduce your cost basis over time and incrementally drive revenue, by serving more and more of your agents via chat.

Atyab: (24:25)
We have a case study that we haven’t released yet, but the insight that we saw from it was a very large travel management organization opened up chat for an enterprise with more than 50,000 full-time employees. And what they saw by opening up the chat channel, albeit at this point it was backed by agents, was the drawdown in email volume. All of a sudden now all these travelers are hopping onto Slack and Teams conversing with the agents and found no need to actually email their particular clients. This particular TMC lowered their emails for this organization and eventually shut off email altogether. And as a result of this chat became the core, primary venue that they were actually using to communicate, and that led for massive productivity gains. But the difference here was that you can’t scale your agents like you can technology. And that’s where organizations like Skylink can play a powerful role in serving your customers in a way that they couldn’t have been done before. I’ll pause there and open it up for questions.

Lierin: (25:32)
Yeah, we’ve got a couple. First question: can SkyLink book cars and hotels and can it process exchanges?

Atyab: (25:41)
So we’re in the process of getting those integrations done today. We’ve been fortunate enough to work with a fantastic vendor that helps us aggregate air content, and our systems are fully capable of booking hotels and cars as well, but we’re building out those integrations, especially during these like beta customer phases so that we can do a full service suite of all the different aspects of the tech stack.

Lierin: (26:04)
Cool. The second question: Can it read screenshots if somebody sends in a flight option they want?

Atyab: (26:11)
So we have this capability, we just have not deployed it yet. we’ve been primarily focused on some of these like, natural language, based bookings, but we know that this has been requested and it’s in our roadmap and we have it available. We just haven’t put it into, the production environment yet.

Lierin: (26:26)
Excellent. next question: Does this service work with Sabre, Travelport and Amadeus? And other than a Cornerstone integration, are there any other web services or APIs needed for it to work?

Atyab: (26:38)
Yes. So you, we can integrate with any API to actually drive the automation layer. we work with a fantastic aggregator that allows us to access the content across all the major GDSs, including, getting marine fairs, travel port, even travel port subsidiaries that are in the process of getting deprecated like Apollo, Sabre, and, Amadeus as well for the air content. Now, as we look to expand into hotels and, and some of these other, content venues as well, we’ll be importing their APIs as well to increase this functionality.

Lierin: (27:16)
Another question is: does it actually build a PNR or just relay the info to an agent?

Atyab: (27:23)
It can do both. It really depends on the internal systems of the TMC. And so in the event that the TMC wants the agent in the loop, we can relay the info back to the, to the TMC agent themselves. Now, in the event that you want to actually have the p and r built, we have to look at your systems to see how you’re actually running your processes today. But SkyLink does have that capability because we actually work with vendors that have decades of experience in actually, enabling bookings usually through a traditional like, traditional like desktop application. Their systems allow you, us to execute the actual p and r on the backend. And so SkyLink is operating as a middleware layer that is making their systems execute the backend just like you would with an OBT. but we just, we, we can do that. We just make sure that the agent is in the loop too, just so they have context on the conversation.

Lierin: (28:17)
Cool. Does it support any other languages?

Atyab: (28:22)
Yes. So Skyline can support several languages. it can do Spanish, Punjabi, Chinese not as great as Punjabi, but like all the major languages, German, Spanish, Chinese, we can actually do an example of it right now if that would be helpful, if folks are interested. But yes, it can support all the major global languages, with relative fidelity. it’s best with English, but it is in parody or slightly better than what a Google Translate might give you if you were to, move between English and, and Google Translate

Lierin: (28:52)
A follow up question is asking specifically about Japanese

Atyab: (28:58)
Specifically about Japanese, we, I would have to look back on that, but I do believe we can support Japanese, in terms of switching, being able to interpret what’s being asked and responding. but let me get back to you on that. I know that we can do Mandarin and we can do Spanish and major European languages, and I believe Japanese is in there as well.

Lierin: (29:19)
Excellent. Are there use cases in operation today with the iQCX integrations?

Atyab: (29:26)
Not at the moment. That’s something that Mat and I have been discussing right now. Really what we’re at is looking to co-create this journey with your organizations. And so what we want to do is hear what you have to say and what you’re looking for and then, work together to see how we can get there eventually.

Lierin: (29:48)
Excellent. And can it connect to workflow management tools like Salesforce or front to process emails?

Atyab: (29:56)
So, yes. So we can do those integrations. Typically, we ask that the TMC integrate with our systems to actually do that. Alternatively, we can help support on that implementation effort. It really, like, there’s a bit of a co-design effort where we look at your enterprise architecture and manage, understanding how your systems are tacking talking to each other so we can plug in skyline.

Lierin: (30:18)
Cool. those were the questions that I’ve had so far. I’ll let if I get any more in.

Atyab: (30:23)
Great.

Mat: (30:25)
Wonderful. Well, hey, I’ve got some questions. So, when, I look at the interactions, and I know we’re early days in looking in our collaborations, and Cornerstone’s been really involved, in building out it’s workflow automation, it’s mid-office to be able to be integrated into, third party, kind of interactions. And so how that happens, I mean, is really key. I mean, when we look at, automation and AI coming together, there tends to be a conversation that takes place in various aspects of, especially post ticketing issues, such as schedule changes such as, managing refunds and exchanges and handling all of that process. And, and that’s where some of the complexity of, of travel comes in.

Mat: (31:19)
One of the challenges that, many of our customers are having is, is understanding the value, for example, and the residual value of an unused ticket. And so, looking at all the rules that airlines have put out, as to how that happens, can this ticket be exchanged? What’s the parity of that, compared to that ticket, how do you see your platform or just AI in general being able now to kind of interpret this complexity, in the industry that, is, like you said, a lot of agents spending time researching and, and figuring out if I can do something. What do you say to that?

Atyab: (31:54)
No, I, I I think that’s a great question. And, and I think beyond just SkyLink, but the, the, the AI advent that’s been happening over the past two years, AI can rapidly understand deeply complex concepts and its ability to reason is improving substantially. , these AI models that came out two years ago, or let’s say start with 10 years ago, like an Amazon Alexa or a Siri were relatively simplistic. They wouldn’t be able to understand the nuances of, here are the 20 different rules we have to address. Here’s how we, how we actually go ahead and, and resolve ’em, right? But then when these chat G P T models came out, these new G P T models had the ability to reason the initial models were a bit like a high schooler. they, they, they could understand what you’re asking.

Atyab: (32:41)
They could write a response, but they weren’t as astute as say, like a seasoned professional. But when you take these models and you train them in a way that is incredibly robust, where they have specific use case duties that they’re designed to address, then they can start to programmatically understand the, the different requirements you’re solving for. So in the example that you just gave, which was the exchange of a particular ticket, right? You can train these models and the this task is relatively complex to be able to create an architecture where these models are doing it with a very, very high quality, to be able to understand here are the 50 different rules for this particular airline, and there are 50 different airlines we have to address. Now all of a sudden, we have 250 different scenarios on a per airline basis that we’re looking on a total basis that we’re looking at now. And AI can programmatically understand these and then also be able to adapt and see what the potential options are and serve that back up to the agent in an incredibly fast and efficient fashion. Right? You can generally expect over half the time to go down, and then a portion of that tends to get handled by the agent as well.

Mat: (33:51)
Right. And so, I, because I see that, that agents are having to go outside of the, the booking environment, whether it’s a G D Ss or otherwise to get this information, they’re having to go onto the airlines website, they’re having to look up these kind of things. So do you see that, that the SkyLink platform is almost also going to incorporate that knowledge management? Yeah. Now that, that the industry so much needs, I mean, because as we talked about, post pandemic, I mean, it’s a, it’s an industry that’s losing that knowledge rapidly. Yeah. It’s documented there, but it’s hard to find and it’s hard to curate.

Atyab: (34:29)
Yeah. And, and, and I, I think that’s exactly right. Effectively creating a very nuanced search engine one that actually finds specific information that doesn’t require the, it’s no longer about Google searching. It’s effectively talking to Skyline to actually, instead of a, to directly answer the question you’re looking for, right? We believe and we tracking towards getting there over time. it’s, it’s documenting and finding where all this information exists is the first part of the problem, right? Like, you need to know here are the, 500 different places all this information is laid out in then pulling that information, making sure it’s up to date constantly and, and making sure that the systems can talk to it, that’s something that’s our bread and butter, right? And, and then enabling the agent to be able to have those conversations is, is incredibly powerful too.

Atyab: (35:16)
And so, just like you mentioned Mat, like, it, it, it is about getting there eventually and, and understanding what are the particular things that are taking up the agent’s time the most. And so that research component, that r and d that comes today, that that is exactly what we’re targeting and, and we’re starting with these like primary workflows that we like to call, like the bookings and the r and d and the itinerary development and the p and r, and then, programmatically moving from there all the way down to, making sure it’s, hitting all the right spots in terms of, getting all the research and, and the, the details around an exchange or the policy of a particular client so that they don’t have to go ahead and read it. Or a new agent comes in and doesn’t have to ask another agent, but instead can just have this whole end-to-end conversation, with Skyling to figure out what are, how to answer all of their questions directly.

Mat: (36:06)
Yeah, exactly. Because I, it’s, it’s, for me, it’s, I love those automations that we’re going to put together initially. I mean, those are ones that, you and I hear, from, our customers that, are, are really important because they just, waste time. Can you send me a copy of itinerary? Can you do this? Can you, those are, yeah, those are the obvious ones. But even in a chat, in a channel Slack channel, you can say, Hey, can I exchange this reservation for, for this unused, with this unused ticket? Exactly. And just bring that back to them because that, that’s where the complexity hits. That’s where, if they could say yes, and the answer is yes, you can do this. You want me to do it for you, bam, it’s done. Or, you can’t, or, , you, you might, if you can get a waiver on this, that kind of thing. Yeah, exactly. ’cause that’s just a much faster interaction.

Atyab: (36:57)
Yeah. And, and, and that’s, that’s exactly how we’re thinking about it. Like the, the way that we operate in the operating models of travel management agencies will fundamentally change now. Yeah. And it’s not just travel management agencies. This is like a broad base shift on, on the advent of like how our economy works and how these industries work, across banking, manufacturing, the, the use of AI to actually drive these operational efficiencies will be very similar to how telephone made communications faster. The internet made information accessible. This will be the, the, the next iteration of that kind of, evolution of technology.

Mat: (37:36)
Yep. Yep. Alright, we’ve got about, seven minutes left here. Let’s, let’s talk about our next steps.

Atyab: (37:43)
So next steps wise, folks, we are looking for beta customers that are willing to design and co-create and actually specify, what are the specific things you might be looking for, across, the capabilities, of both Cornerstone and Skyline together, and really just provide, working with us to provide feedback to, to iterate and co-design what great really looks like. one of the things that, both our organizations really value is kind of working from the user’s perspective because there’s no better input than, than the end user who kind of tells you what the pain points are and where, where they want to be able to see like, , kind of this new, efficiency gain. And so we’re looking to solve for beta customers upfront. We have a couple of spots available, but they’re closing up quite fast. and so if folks are interested, please reach out to either , myself or Mat and, and, and express your interest, in, in, in a collaboration here.

Mat: (38:42)
Wonderful. Mat. I’ll hand it off. And that’s what we’re really, looking for now. I mean, it’s, like you said, it’s early days in all of this. I mean, these, this, we, we have, we have proof here that, that, that this works, that it can work. and, now we need, , if we need transactions, we need those integrations to come through. We need to train the models in how to do this well. and , and, and learn from that experience because it’s a, it’s a bit of a reset in our industry and, and it’s not a reset for, for the worse. It’s a reset for the better, I think.

Atyab: (39:17)
Definitely. And, and I, and I think one of the powerful things here is that the travel industry uniquely gets the opportunity to leapfrog. I, I think it’s no, it’s no secret that sometimes technology can be a little bit of a laggard in certain industries. And, as a result of this, we, we have a lot of legacy systems we might be using, but the benefit now is that you can, you can rapidly jump over that and, and, and experience the next iteration without having to have done all the legwork that came over the past 10 years. And so it is really just like you mentioned, the early days of like co-creating, bringing like good to great and really executing here.

Mat: (39:57)
Absolutely. Absolutely. So before we close out, Rin, do we have any other, questions

Lierin: (40:03)
We do, in fact, so a couple questions here for the end. at, you mentioned marine fairs earlier. Can Skyland interpret an email from a client with say, five travelers all coming from different places who all need to get to one destination at roughly the same time, for example, to board a vessel?

Atyab: (40:22)
That, that’s a great, that’s a great question, and it’s a use case we want to be able to address. we’ve, we, we would need to see what this workflow looks like to be able to train Skyline to actually handle it. The, the, the benefit of SkyLink is that as it understands more and more complex scenarios, it’s able to adapt to them. And so d as long as we can access the content and we know what this workflow looks like and it’s a meaningful workflow, and that it’s happening frequently enough, we can adapt to it and, and, and, and resolve it as well. I don’t wanna commit to, to saying yes for these five folks, we can get them onto a vessel, but until we actually see what it looks like and ha have the details there, but it is definitely within the realm of like, doable as long as we make sure we have access to the content and to the profiles associated with those folks.

Lierin: (41:10)
Cool. can SkyLink use the G d s or umbrella traveler profiles to make bookings?

Atyab: (41:16)
Yes. Yes, it can. So it can, it can pull the profiles from these systems as well. We’ve been speaking with a couple of folks, across those kinds of organizations to make sure that, we can work to get those integrations live in a way that’s very high fidelity and, and has everything mapped internally.

Lierin: (41:33)
Cool. next question is, does SkyLink manage the look to book ratios of any customer chat so that the airline sources being pulled upon aren’t getting a large volume of hits and charges with every, which somebody needs to pay?

Atyab: (41:48)
Yeah, so we do, we do the best to drive those optimizations. What we do like to do is work with the TMCs agents to see how they run that process flow, today. And then what we do is we try to make sure we’re in parity on that look to book ratio so that we can recreate the agent flow as closely as possible.

Lierin: (42:05)
Cool. next question is, does or can the system give travel advisements depending on the location the traveler is going to?

Atyab: (42:13)
We don’t have that feature yet. that is another feature that’s been asked about, and we plan on bringing it, I believe quarter two or quarter three of next year. we can pull it up if there’s a lot of demand for it, but not today. It’s, it’s, it is a functionality we can add in, and we have the capability to do it. We just need to understand how those systems look on the backend to be able to give a concrete answer.

Lierin: (42:35)
Awesome. And this is the last question that I have here, written up, is how does the system advise tickets are not guaranteed until ticketed?

Atyab: (42:45)
How does say, sorry, could you say that again?

Lierin: (42:48)
d how does the system advise that tickets are not guaranteed until ticketed?

Atyab: (42:56)
Oh, I see. I th Well, it depends. Like if you, you could customize the workflow with SkyLink, I think I showed this on chat, on how we can change the way that responses are laid out. And so in the event that you wanna say, Hey, this issue is ticketed, we will run the booking confirmation, over and, and send you the kind of the standard of post ticketing or post booking workflow where it goes from booking to ticketing. we’ll, we’ll send the email out just the way that you normally do it to confirm that you actually have that particular flight or hotel. Awesome. and so it’ll effectively reunify your traditional workflow. And you can think of Skyline kind of running that same workflow in like an O B T like fashion, except instead of an O B T, you’re chAtyabng with it.

Lierin: (43:37)
Cool. So what I’m going to say here, we don’t have any other questions, but, when I send out the recording for the webinar, everybody is more than welcome to respond to that email with any other questions that you have, and we’ll make sure we get them to and Mat so that they can be answered for you.

Mat: (43:54)
Wonderful. And I’ll add one more thing there, is that we will, as part, of that email, we’ll send out a short survey, with that, that also, allow you to express your interest in, in joining one of our betas. and as I said, this is, this is an evolution and a revolution of sorts. And so, I really very willing, to, to meet first obviously with those folks that are interested and, and sort of lay out a plan on, on how that would work. We certainly have our ideas of where the integrations between the SkyLink, platform and the Cornerstone Automation platform will come together. And we presented some of those, integrations here, but really wanna listen to others as well too, because we’re, we’re building this from the ground up and we wanna make sure we’re working with folks that,

Atyab: (44:44)
Yeah, I, I just wanna echo Mat’s point, like, AI is, we, there’s a lot of hype around it, and there’s a lot of like,

Atyab: (44:52)
Very loud voices in the room when it comes to ai, to actually drive like, value and, and, and efficiency. It’s, it’s really about taking it from the perspective of the, the agents and the travelers and, and co-creating it like that. There, there’s a lot of noise that we have to like, kind of work through so that we can really make sure that this is addressing the core problems that, that, this industry faces and, and which ones they want to resolve. And so, kind of working with folks like yourself and making sure that, that journey is very smooth and, and it is truly a journey of, of all three parties here, to, to make sure that it’s like very cohesive and productive for everyone. But, thank you folks for your time and thank you, Mat, for having me.

Mat: (45:35)
Thank you, AAB, good to have you with us and, good to be working with you. Alright everyone, thanks and have a wonderful day and and a wonderful week. Take care.

Atyab: (45:45)
Thank you folks. Bye.