Data, AI and the new TMC: Conversations with leaders driving change

Claudia Lewis

00:04 - 01:33

Welcome , everyone , and thank you for joining us this morning. I'm Claudia , and I'm I lead , Domo's travel sector here in EMEA.

And a big part of my role is talking to TMCs every single day. And over the past twelve months , a few things keep coming consistently in those conversations.

People are asking us for more rights , get them those faster , better visibility into their data , and , of course , AI is is on everyone's radar. But there's a lot of uncertainty about it , where to actually start , what it looks like in practice , and then , of course , how do we actually bring our teams along with us on that journey.

And I think , additionally , there's a lot of pressure within this industry specifically to demonstrate the value of AI , but it's an industry that's built on people and their interaction. So today will be a practical discussion about what it really takes to put data and AI to work within that travel management context , and I'm really looking forward to the conversation.

And and just as a quick reminder that we will have a QA at the end as well. So as we go through the conversation today , please throw your questions into the chat as they come to mind , and we'll answer those at the end.

So with that , I'd like to welcome our speakers today and get them to introduce themselves. So , Kelly , why don't you yourself first and and introduce yourself ?

Kelly Luney

01:33 - 02:20

Hi , everyone. I'm Kelly Luney.

I'm the travel technology director at the thegoodtravelcollective. co.

uk. I've been in the travel industry now for over thirty years.

So So giving me a unique perspective on how technology has evolved and transformed in our sector over that time. Today , I lead our technology and strategy innovation agenda focused on data , digital transformation , and more recently , obviously , the opportunities that AI presents to improve customer experiences , streamline operations , and drive smart business decisions.

So I'm very passionate about leveraging technology to create meaningful impact , building high performing teams , and supporting other women as well in our leadership role. So I'm really excited to be here today and looking forward to the conversations.

Claudia Lewis

02:20 - 02:37

Thank you , Kelly. Oh , can I can I just quick one while we're going through the question ? Can you pop yourself on mute ? So we were getting a little bit of background noise as well.

So just as we go through , just pop that on mute. Next , I'd like to thank Helen for joining us.

So quick intro introduction from Helen as well.

Helen Mairs

02:37 - 03:33

Hi. Thanks for inviting me today , actually.

It's really good to be here. So , my name is Helen Mairs.

I am head of data at Take2Eton. I've been in the role for about three years , and I look after the data team here working within the technology function.

So we are really looking about how we can progress data and then deliver new ideas , products , etcetera , for our customers. I've worked in data quality before that , so I get a really good understanding of data quality issues in the travel industry.

And I've been in the industry for over thirty years , so I I know it really well. I think , really , what we're looking at now is is how we can use data to really sort of change things moving forward.

I think data has been quite static in the in the travel industry. And , actually , now with new tools and new new ideas , we're really getting the opportunity to sort of change things moving forward , and I'm really excited to be chatting about this today.

Claudia Lewis

03:33 - 04:24

Perfect. Thank you.

So , Helen , Kelly , really really excited to have you both here and that you're both operating in a market that that's changing so so quickly at the moment. And the way that your clients access and use their data has shifted significantly , not over not just over the your guys' careers , but just in terms of the last twelve months.

And that's because AI is creating real opportunities for the way that TMC work and giving you opportunities to work in ways that just weren't possible before. So that's the context I wanted to to build today's conversation on.

So , why don't we start with that that perspective on , you know , what's changed in the TMC landscape over the last twelve months that makes data and AI feel so urgent , you know , now more than , say ,.

Kelly Luney

04:24 - 04:24

Yeah.

Claudia Lewis

04:24 - 04:29

two , three years ago. Helen , why don't we we kick off with you on that one ?

Helen Mairs

04:29 - 05:36

I think global events really are shaping what we what we need to do with data at the moment. People wanna know more about what's going on and and what's happening , and and data is obviously gonna move forward with that conversation.

We had you know , those of us who've been in the travel industry for a while live through the ash cloud , but it was sort of a big one off event. But what we're finding now is events are sort of happening and then happening and happening , and it's almost continuous.

We are really good at reacting to the the data that we have. So we have it looking for you , it's changed , how do we react to that ? But actually , what how can we use data now moving forward to predict what we've you know , what's happening ? So , you know , this flight is getting canceled week on week out.

We know that you normally book it. We're advising you to take a different time from so I think we need to be more reactive with that sort of predictive element , and data is obviously gonna be for you know , in that.

So I think we used to have a more reactive position , and now , really , we need to be sort of predicting and people want insights into what's happening moving forward. So that to me has been a massive shift.

Kelly Luney

05:36 - 05:38

What that can happen. pretty good.

Claudia Lewis

05:38 - 05:43

Kelly , what what's your your view on this ?

Kelly Luney

05:43 - 07:05

Because I think for you , what's really changed over the last twelve months is the pace , and accessibility of AI. Two or three years ago , AI was just something that many organizations were really exploring , whether through pilots and proof of concepts.

I think today , it's become much more tangible and practical applications that we can deliver , you you know , deliver value really , really quickly. And at the same time , within the TMCs , we're dealing with , you know , increasing complexity , whether it's around content sources , high customer expectations , tighter margins , you know , growing demands for , like , real time reporting to make , you know , better decisions.

We're looking for sustainability data , traveler insights. So I don't think the challenge is , though , is not so long.

I don't know. I don't think the challenge is no longer about the data , about the lack of data.

It's how we turn that into sort of like an actionable , intelligence. So AI is obviously helping bridge that gap , enables us to automate a lot of routine tasks , you know , uncover insights a lot faster , personalize our traveler experiences , and empower consultants and customers to make better decisions.

So for me , I think the urgency comes from the fact that organizations that learn how to effectively combine quality data and AI now have a significant , competitive advantage in the years ahead.

Claudia Lewis

07:05 - 07:42

Yeah. That makes sense.

And I think I think , yeah , what we're seeing really is this market that's just moving so so quickly. And a big part of that as well with that , you know , fast paced change , you do need real leadership to navigate that change.

Right ? And you're both senior women in those positions of transformation within your business. So I think it'll be really great to just understand a little bit more about what you've learned in in that journey and and what's helped you along the way.

And if there's any advice you would give to people in in a similar position to you guys. So , Kelly , why don't you you kick us off on that one first ?

Kelly Luney

07:42 - 10:02

Yes. Yeah.

So , I mean , I've spent my entire career in travel technology and operations. So I started out at the GDS Sabre.

I gained a strong foundation there in how technology underpins the whole of the travel industry. So early on , I was really hands on , field based , working closely with customers , understanding those day to day challenges that , you know , all travel businesses face.

That experience has obviously stayed without me throughout my entire career , and it's taught me that , you know , technology creates values when it only when , you know , when it solves real real business problems. You know , my most rewarding parts of the journey is was obviously helping shape , a TMC and scale that I was working through with through a significant period of growth.

So when I joined that business , they were around 20,000,000 in turnover. And over the time , I was part of that leadership team.

We helped grow it over to over just over 200,000,000. So , you know , that journey taught me that transformation isn't just about technology.

It is about the people , processes , cultures , you know , and having a clear vision of where we're trying to go. So , you know , alongside that whole career , also become a mom.

So , you know , balancing that leadership responsibility with family life , obviously , probably has taught me , just as much as my roles that I've held as well. You know , it's taught me about resilience , prioritization , and , you know , you don't have to do everything perfectly to be successful.

You know , there have been times when work needed me , times when my family needed me more , and learning to navigate that balance is really , really important in any leadership , journey. And I think , again , as a woman , leading technology operational change , I've learned the importance of just being curious , being adaptable , you know , not being afraid to take on any challenges outside of your comfort zone.

You know , and the advice I would give to anyone on a similar path is to build credibility through experiences. Just stay focused on those business outcomes.

Don't wait until you feel a 100% ready before stepping into that next opportunity. You know ? Some of my biggest some of my biggest breakthroughs have come before , you know , before , you know , saying yes before I had all the answers.

Claudia Lewis

10:02 - 10:03

Love that , Kelly.

Kelly Luney

10:03 - 10:04

Love that.

Claudia Lewis

10:04 - 10:20

Just taking that leap of faith , doing what feels right at the moment. You know , we're not all experts , you know , on day one of the job.

I think that's a great piece of advice and definitely something that , you know , people should take into this AI journey as well. Kelly , again , just a quick one.

We're still getting feedback. So in between speaking , can you just pop yourself on mute ,.

Kelly Luney

10:20 - 10:22

Yes.

Claudia Lewis

10:22 - 10:32

on the microphone ? That'd be great. Thank you , Kelly.

And , Helen , yeah , same question to you. Bit of background on on yourself.

and and any advice you've got.

Helen Mairs

10:32 - 12:28

Yeah. So I started in in travel , again , like , over thirty years ago.

So back in the day when we had to manually write you know , we could manually write tickets , and rail tickets were all manual as well , and even before low cost carriers , so that's how long I've been in travel for. I had a brief stint in academia.

I was doing , operational roles for a bit , and then I had a brief stint in academia and then went back into travel as those of us who tried to get out end up normally doing. I worked in implementations and I worked in the data quality role , so I did that for about eight years before moving into a head of data.

I think the big thing for me has been , first of all , her roles have changed in data. So first firstly , I think it's it's okay to be a data manager or a data governance manager or a data quality manager.

It's okay to have those roles that are less technical maybe. So if you're not a programmer or you're not a developer , it used to be sniffed on or whatever.

But but actually now it's it's much better. You know , I think those roles are more acceptable.

I think for me , the biggest things that I have learned is I have gone into any major projects or new roles with a Take2Eton we very much have a data strategy in place and that's what we work towards. So really for me , I think the biggest advice is yes , absolutely know your industry , you know , know what the processes are , know how you can support , you know , your your end users with data.

But actually , make sure that you have a longer term strategy because I think especially at the moment with AI and new fancy tools , etcetera , it's very easy to go off and get , you know , wowed by the new shiny thing when actually , you know , it then sends you off into a different direction. So build your strategy , build that with business values in mind and the business strategy in mind , and then move it towards that strategy as you're starting to deliver.

Claudia Lewis

12:28 - 13:22

Love that. Kelly , if you just hit the little mute button at the bottom of your page , you should see , like , a little mic button that will that will help us manage that feedback.

Perfect. So you both talked about what it takes to lead change and and transformation , not just in the current day , but , you know , as it's been throughout your careers.

And I think the real proof of that is in what it's delivering or what it will deliver for your business or your clients. So wanna move in onto that now and just understand a little bit more about what it's like to to actually , you know , implement that change and the impact it has.

So , Helen , again , let let's start with you. Where were things when you first started at Take2Eton , and where are you now ? So what has moving to this new approach given you , but also the the customers that feel the impact of it on the other end ?

Helen Mairs

13:22 - 16:57

Yes. When I joined , we had a analytics manager who was , you know , had had been running a product for a while , and it worked it worked pretty well.

But then we very quickly got on board some of the off the shelf travel systems for data and reporting. And almost immediately , we realized that actually there was a number of limitations on those.

So we ended up having to run different systems for our internal and our external reporting. So we were running a very well known , very generic , reporting system for our internal reporting.

And we were hitting a brick wall with them in terms of we had done a bunch of training , but when there were issues , we couldn't get past those issues , their support system was go on YouTube and find some videos , and that should support you rather than getting any support from the company themselves. And then on the off the shelf travel ones , we couldn't for our own reports , any report that we want to deliver.

We , we ended up having to just from essentially it cost us money for every report and every just report that our customers want to. So we ended up with order status , which was no fun for anyone who really wants to work in data.

You know , you want to be problem solving , and you want to be delivering , you know , change. So like I mentioned before , we have a strategy.

I put that in place and then we went off to Big Data London , spoke to you guys at Big Data London , and then in with Domo. And what that's done is we've been able to put both our internal next to our reporting on one system.

We have control of our data. So our we own our own data pipelines.

We own our ETLs. We own our ports.

We're not on the behest of third parties whereby , you know , if somebody comes to us with a report , then we build it rather than then sending it on to somebody else. We can , we can then do more with it.

So as the most product road map increases and there's AI products and there's , you know , the road map sort of moving forward and more products come out , we can then adjust and see what we deliver to customers. So there's been a couple of things.

First of all , we now have our customer reporting on that , and we have customers using our front end , reporting tool lens. And then we have an internal reporting portal.

So we now have a bunch of internal teams that are using our reporting as well , and they can build their own reports and build their own dashboards , but , really , they now have access to data. What that has meant is the mentality around data has changed as well.

So we're now getting people coming to us going , well , what can we do with this , and what can we do with that ? And my customers asked a question about the data , and we like saying to people , challenge us. We used to have big companies come to us and say , oh , can you do this or can you do that or can we integrate systems ? And the answer was , well , maybe , but it's gonna be expensive and it will take a while.

Whereas now the answer is , yes. Absolutely.

We can do that. And almost we can answer yes without going into the technical detail of it because we have now a data platform that's really connective.

And so it's just it's so much better having the confidence in those conversations to be able to say to customers , yeah. Actually , the chances are we can do that for you.

Yeah. There may be a cost or , you know , may take a bit of time to deliver or whatever we need to read through requirements.

But , yes , actually , we can do that rather than being sort of cautious or not sure and and let me see. So we have a much happier data team that are now delivering a bunch of products , that that our end users are really enjoying using , and it it's a it's a much better place to be.

Claudia Lewis

16:57 - 17:41

That shift , Helen , that you've described from fragmented systems to to being able to respond to your clients' requests so much quicker is is , like , the real proof in the pudding when it comes to transformation. Right ? And I think it's often underestimated just how much organizational change it takes to to to to deliver that for your clients.

Right ? And a big part of that journey , which is the hard part of getting those systems talking to each other and centralized , is often forgotten. So , that work that you've done to to centralize everything puts you well ahead of the curve when it comes to successful AI implementations moving forward because you really have established a really strong data foundation.

Helen Mairs

17:41 - 18:02

Yeah. Yeah.

Absolutely. And now the questions that we're getting are more around the , you know , what is my cost you know , it's less of the what is my customer spending , how many invoices , what have they done.

It's more right. My customers come back to me with an idea around this.

How can we potentially deliver it ? And so the conversation's changing. It's a much it's a much more fun conversation to have.

Claudia Lewis

18:02 - 18:03

Fun and and strategic.

Helen Mairs

18:03 - 18:03

data.

Claudia Lewis

18:03 - 18:05

Right ? It totally. changes the tone of it.

Helen Mairs

18:05 - 18:07

Yeah. Absolutely.

Claudia Lewis

18:07 - 18:37

Kelly , wanna come to you on exactly the same thing , but , you know , you're in a different part of the journey at at Good Travel , and , you know , you are shifting to using data and AI a lot more. But , you know , getting to that point of making decision we're gonna do this requires a lot of internal change , not just the the technology side of things.

So as you're going through this shift , how are you bringing the organization along with you on that journey , and what are some of the challenges you you've had to overcome during that ?

Kelly Luney

18:37 - 22:10

Yeah. Absolutely.

So our goal isn't just to implement , like , new technologies or AI. It's about creating a like , a single trusted view of the of the business that helps us make better decisions , operate more efficiently , and ultimately deliver a better experience to our customers.

So the challenge that we have or we had , should we say , was bringing two businesses together. Obviously , we acquired someone.

We had to bring them together. So each of each of us each of the companies have their own systems , their own processes , and own way own ways of working.

Naturally , both of the organizations believe that their approach was the right one. So before we could even talk about , you know , AI , what we have to do is address the much bigger challenges.

How do we create that one version of the truth ? And so one of the first discussions that we had to have was whether we should rip everything off and start again , find a completely new pod partner , or build on what we already had. So what we did , we took a step back and went back to basics.

So we knew we needed to just simplify the landscape. So we went back to having one GDS , one back office , and a clear strategy for bringing data together across those businesses.

And that's what ultimately led us to , obviously , coming to Domo because we needed a platform that that would help us connect everything , can give us , like , a holistic view for the organization. I think the biggest lesson for me was that transformation is far more than just about the technology.

I'd only been at John Good Group for around three months when when we started to introduce some of these changes , and alongside the technology shift , became significant process changes. So rather than leading with technology , we focused on , like , trust and engagement.

We spent a lot of it time explaining , you know , why and answering the questions that everybody was asking , which was , you know , what is in it for me ? And I think one thing that worked particularly well for us was involving the teams , you know , really , really early on. So , you know , one of the things we did was we brought our sales teams into conversations with yourselves right from the outset.

You know ? And probably at the time , it really felt overwhelming at first with , you know , all of the things that we were talking about. You know ? And , you know , everything didn't make , like , immediate sense , but at least they had visibility.

They felt included , and they could see that , you know , that the journey was unfolding , rather than having , like , changed imposed upon them. And I think now they're starting to see the values for themselves to the insight and visibility they're getting.

You know ? And if there's one thing that I'd stress , is , you know , none of this is gonna work you know , about the AI. None of this is gonna work without clean and trusting data.

You know ? Everybody wants to talk about AI , but the real work starts with getting , you know , getting the basics right , getting the foundations. Right ? You know , if your data isn't consistent and connected , you know , you're not gonna get any value anywhere near what you expect from AI.

So , you know , sometimes going right back to basics , it might feel like a step backwards , and it can feel really uncomfortable when you're having to have those conversations. You know ? But , again , if you're not failing occasionally , you're not probably pushing your boundaries hard enough.

You know ? In innovation , it does require willingness to change , you know , and challenge the status quo. Now learn quickly and bring people on that journey with you.

Claudia Lewis

22:10 - 22:21

I think that's , yeah , really a really important thing to bear in mind there is , yeah , back to basics and and just being transparent with with the business about what's happening , especially with something like AI. Helen ,.

Kelly Luney

22:21 - 22:21

Yeah.

Claudia Lewis

22:21 - 22:32

from from your experience of change management , again , later on in the journey than the guys at John Good Group , but what would you do differently if you were starting again now ?

Helen Mairs

22:32 - 24:42

I think there's a couple of things. So first of all , I think I I would focus a little bit more on the end product.

I think the the thing is in changing the data platform , it's it was very easy to get , carry away with the potential of it. Like , oh , look what we can do now that we couldn't do before and and use you know , as much as I said said don't get well by the new shiny thing , we kinda got well by the new shiny thing.

So I think having , you know , clearer , ideas about what you want to deliver and then work backwards in terms of how you build the data architecture in the system , how you put the data together , what you want to connect to , and be clear on those outputs. And , also , I think for me to have a clearer road map , so a clip from a product perspective.

So we had an idea about you know , we wanted to deliver customer reporting , and we wanted to deliver internal reporting. But I think in terms of engaging with end users about what that potentially looked like , I think we probably missed a bit of a trick there.

So I think that would certainly be something , that I that I would do. The challenges , of course , is that you are learning a new system as well as delivering new products and delivering , you know , a a new way of working essentially around data.

So maybe that period of , right , let's just do some learning. Let's just understand.

Let's build a test case before we sort of just go out and go , look , we can do everything and then build it. So I think maybe starting with some smaller examples and and building , you know , a sort of prototype tool , if you like , for for reporting probably would have been a good idea as far as we were concerned as well.

So I think having those elements in place. I mean , Kelly's right in terms of the people , you know , it's managing , you know , change as well.

And people are always really scared , I think , of a new product or a new system or a new , you know , being cautious on that journey. They they want something new and they want us to do something different , but actually , it's , you know , that there's always nervousness around it.

So I think probably more engagement for our side as well would have would have supported. But But , certainly , the product road map , I think , would have been a good thing , and then those sort of test cases would have I think if I'd done it again , I think those would have been the key examples.

Claudia Lewis

24:42 - 26:05

Yeah. I think we always underestimate the power of of of incremental change and and starting small , right , especially with something like like AI.

But , you know , what what I'm hearing from both of you is there's there's three things that we need to to keep in mind as we start these journeys , which is get your data in order , bring your teams along , and try to show people results before asking them to trust the entire process and everything that you've done. And that rings true across every industry.

It's not just travel. Right ? That's that goes for everyone , and that's why we see this this really scary statistic at the moment going around , which is is ninety 5% of Gen AI pilots failed to deliver any real measuring measurable p and l impact , which is is a crazy statistic considering how much money businesses and and and globally we're spending on AI at the moment.

So the next step is where I really wanna take the the conversation now , and that's around how do we use AI and and successfully apply it within a TMC context. So there's a lot of excitement around this , a lot of excitement around agents and and how we can use those within within travel , but there are some real challenges to that as well.

So , Helen , you've already got some experience in this area , and it'd be great if you could just talk through , you know , what AI agents are doing within the context of the TMC and what problems you've seen them solve that perhaps you couldn't solve before.

Helen Mairs

26:05 - 29:19

Yeah. So , actually , this is one of those cases where we've kind of started off small , and used an example.

What we didn't wanna do is go all out with AI because we are dealing with people's personal information. So we've got passport , you know , data and visa data and , you know , people's travel details and which seat they're sitting on on flights and hotels and all that.

So we didn't just wanna go knee deep into to AI and and , you know , run the risk of exposing the data. So what we have done is we've started off with a case study around master data management.

We AI is normally really associated with , like , you know , the automation around things like chatbots and , you know , those conversations , but we are a high service company. And so when people phone us , they phone us because they want to speak to a human being.

They don't wanna speak to a chatbot , or message a chatbot. So we have an online booking tool that , obviously , people use.

That's fairly ubiquitous in the industry. But when people speak to us , they want to speak to somebody and use their , you know , thirty years operational expertise to sort out a trip or or whatever.

So chatbots was never gonna work for us. So what problems do we need to solve ? Right.

Well , hotels is a big one for us. So when we book a hotel for somebody , we look at lots and lots of different booking channels so we can get the best rate for people.

So when we look at different booking channels , that means we get lots of different versions of the same hotel name. So we may get , you know , the the Macquarie Manchester or the Macquarie Manchester city centre or the city centre Macquarie and various iterations of the one hotel.

And , obviously , when that shows in the data , because it's different names , even if there's a space different , it shows us different hotels when we group them together. So if a traveler or a a travel manager is looking at either their top 10 hotels or looking where people have stayed so they can negotiate rates , you have different iterations and one of those those versions may be missed out of the conversation because of that data quality issue.

So that's a big one for us in terms of how do we manage data. So what we looked at is we have a list of hotels that we got from the our GDS , our our global distribution booking tool.

And when we bring in our transactional data , we use an AI agent to look at the hotel details in that transaction and match it against that master list. And we have criteria like phone number and , you know , fax , believe it or not , because we still have faxes , you know , address , etcetera.

And we match against a number of different criteria , and then we give it a score. The AI agent gives it a score.

And if it's over a particular percentage , then we update transaction data with the updated version of the hotel name. And if it's below that particular score , then we push it out to a team who can verify that that hotel is the same hotel.

And then it updates the transaction data , and so we get a better version of the data. And it works really well.

So that's the first AI agent that we've looked at. It's now , you know , in place.

It's it's updating our data. It's making us have better data , which means that we are trusted you know , it's trusted data , which is a big , big thing for us , and it it's more reliable.

So that's our first one that we've developed , and , obviously , we're gonna use that case study to to look at AI AI agents moving forward and how we can use them on on automated workflows.

Claudia Lewis

29:19 - 29:33

Kelly , where where are you at this journey ? You know , where are you on the journey ? And for anyone listening who's who's thinking about AI and using agents in particular , what advice would would you give them when they when they look to start ?

Kelly Luney

29:33 - 32:34

Yeah. So we're still at the beginning of the AI journey here.

And I think it's important to say that because I think there's so much pressure on all of us to feel like , you know , we've got everything already figured out. You know ? The reality is that we're currently mapping out where AI agents can generally add value to business or to our customers.

And I think , again , you know , one thing I've learned is that AI and AI agents will not solve every single problem. You know , they certainly won't replace the need for people , certainly not in my lifetime anyway , but , you know , especially in travel , you know , around disruption , uncertainty , or , you know , where traveler needs a reassurance , you know , people still want to speak to people.

Now they want that human expertise. You know ? They want that empathy.

Yeah. They're stuck at an airport because something's happened , but they need to get home.

You know ? And I think judgment remains incredibly important for those things. My advice , you know , for anyone starting out , you know , don't begin with the technology.

You know , start with what is the problem that you're trying to solve right now. You know , look for those repetitive tasks.

You know , what take up the most amount of time ? You know , what are the processes that involve multiple systems , you know , multiple contents , multiple touches ? You know ? And where you know ? And what are the areas where our teams are spending so much time and effort on what we call , like , you know , low value activities where we're not learning , you know , where we're not earning a lot of money , but they just take so much time out of your day. You know ? And then look at that end to end process , you know , and ask , you know , what are the costs here ? What are the time involved ? You know ? And what's the return if we can improve these ? You know ? I mean , right now , for us , what we're doing is we're focusing our attentions on you know , we're looking for the opportunities where AI can support our people , you know , remove the fix remove the friction , you know , and help them spend more time on activities and tasks where , you know , they're actually creating value for our clients.

You know ? And for me , it's not just about saying , where can I reduce headcount ? You know ? It's about asking how can I make my team more effective and give them better tools to do their job ? The other thing I'd say is , you know , be really clear about what good looks like. You know , define your goal , what are your gold standards before you start.

You know , if you're using AI and customer , like , interactions. Sorry.

For example , you know , it still needs to feel human. You know ? Customers don't want it to feel like they're talking to a robot.

You know ? They want fast and they want accurate , but they do want that personalized support. And , again , again , we've we've talked about this a lot of times , and we've touched about data.

You know ? Don't underestimate the point yeah. The importance of your data.

Before you think about , you know , the AI agent , make sure that you've got trusted , clean , and connected data. You know , the organizations that are gonna get the most value from AI won't necessarily be the ones with the most advanced technology.

It'll be the ones with the strongest foundations.

Claudia Lewis

32:34 - 33:15

Yeah. Kelly , I really like that you touched on , you know , already know where your your ROI is gonna be when you do this.

I think a big part of this that people tend to forget is it's really easy to to go , we can replace this phone process , speed that process up. You need to balance it with with the cost of AI and and , you know , none of us can.

pretend that , you know , there's LLMs out there that we're leveraging for this. They're they're not cheap to run , particularly if you want a high performance model to replace a , you know , a complex process.

So making sure that you've mapped out the cost of actually using AI and the model that you want to use is also a really important part of this journey that I think gets forgotten and the excitement around of what it can do and what it can replace.

Kelly Luney

33:15 - 33:45

Yeah. I think people get you know , you we all get giddy.

Come on. We all get giddy about , you know , this is what we're gonna be able to do.

We're gonna be able to replace an agent , you know , or with an AI and stuff like that. But , you know , the reality is there is a cost to it.

So , you know , look at those repetitive tasks where there is , you know , low , like , low low value effort cost. You know ? So it's those are the ones you wanna be replacing , not , you know , the more expensive one.

But , like , people do , they do forget all of the costs associated , like you said , with , like , the LLM costs and things like that.

Claudia Lewis

33:45 - 33:48

Everything that we've got everything we've talked about , agents , state's foundations ,.

Kelly Luney

33:48 - 33:49

Okay.

Claudia Lewis

33:49 - 34:32

use cases , none of that delivers if our people aren't on board or don't feel confident working alongside. And that's been a theme , I think , across both of your answers answers as we we have this conversation.

And I think that's where a lot of the important conversations are happening in in all industries when it comes to , AI. People are gonna feel uneasy about it.

You know , there's a lot of change happening , and when there's change , there's there's fear. They they come hand in hand , unfortunately.

Kelly , how do you think about the relationship between your people and AI and your business ? There's anxiety , and , you know , what does it mean for roles and ways of working ? So really keen to get your perspective on that side of it.

Kelly Luney

34:32 - 37:28

Yeah. I mean , I I think first , we do have to to acknowledge that there is a genuine anxiety around , you know , around AI.

You know ? And we shouldn't dismiss that. You know ? Whenever there's any major technology shift that happens , people always naturally you know , we all do it.

We all think , first of all , what does this mean for me ? What does this mean for my future ? And the skills that we'll need. And I think , you know , as as leaders , it's really important that we're really honest and open about that.

You know , have open conversations rather than pretending those concerns don't exist. You know ? My view is that AI should alleviate , alleviate , you know , stress of the day , not with for people , not actually just replace the person.

I think the opportunities we're exploring here are focused on those low level , repetitive tasks that consume a lot of your you know , a lot of the time in the day that don't necessarily create a lot of value. You know , and if we can automate some of those activities , our consultants , our account managers , you know , our finance , our operational teams , they can actually spend a lot more time making , like , bigger differences , solving complex problems , building relationships , you know , advising customers , talking to them , you know , and actually handling conversations or situations that require , you know , empathy and judgment.

With all that being said , though , you know , I do think roles will evolve. You know ? People will spend their time differently and develop new skills.

You know ? But that , you know , if I look at the last thirty years , that's constantly happened , you know , throughout those last thirty years. You know ? Technology has continually changed the way we work.

And I think the most successful people are the ones that who have adapted and embraced those new opportunities. I do think one area we need to think really carefully about as an industry is about how AI will impact those entry roles , you know , those entry level roles.

You know , many of us started out doing tasks that help us learn the fundamentals of , you know , fundamentals of a business. You know , if we're talking AI and taking away , you know , some of those low level activities , we really need to deliberate about how do we continue to develop , you know , the the future , you know , the future leaders of the industry.

You know , how we're gonna develop the apprentices or the graduates or those , you know , those junior talents. Where are they gonna come from ? So , again , it's all about communications.

You know ? We've been very clear with our teams here. This isn't about replacing people.

It's about helping them to be more effective , giving them tools , allowing them to focus on that. You know ? Using their skills , using their expertise and their empathy.

You know ? AI can support that experience , but it can't replace that human connection. Now that sits the heart of , you know , great customer service.

So for me , you know , that's why I'm seeing AI as a partner to our people and not a replacement for them.

Claudia Lewis

37:28 - 37:53

Yeah. I think that's a a really , really important message there.

And a big thing that , yeah , we do need to talk about more is what happens to those those entry level roles. Helen , from from your side as well , you know , what's changed in terms of how people spend or , you know sorry.

What's changed in terms of , you know , how people spend their day to day , using data and AI , and , you know , what's that change for people in your business ?

Helen Mairs

37:53 - 39:17

I think the the big thing for us , certainly from a team from our team perspective , like I said before , we it helps us to work on more of the interesting stuff. And the account managers and salespeople are having those different conversations with customers in terms of what can we do for you in terms of data.

What are the big questions that you have that we can now potentially respond to ? So when we think about TMCs , we generally think about travelers and bookers and travel managers , etcetera. But we work with finance teams , and so can we help with reconciliation ? Can we help with automation around that ? What are our processes around reconciliation and BSP , etcetera , that we can potentially automate ? So there's a lot of things that we can look at from an AI perspective where we are not doing those boring sort of automated tasks , that we are improving the way that we do those those big reconciliation jobs for argument's sake and then just looking at the the exceptions.

The big thing I think for me is AI is now everywhere. It's ubiquitous.

So my phone updated at the weekend , and I've now got a lot more AI AI functionality on it. So we have to move with with how the sort of wider industry and and the general day to day is is moving as well.

So people are asking more of AI in the day to day lives , and that's certainly feeding into what we're being asked as a business as well.

Claudia Lewis

39:17 - 39:47

We're nearly at time for for the discussion , guys , so I'm gonna rush to our last question. But I just want , you know , one last question for both of you.

If you could give , you know , one really important advice piece of advice to a TMC who wants to start getting AI into production in the next twelve months , you know , one key sentence , what is that ? Kelly , you first.

Kelly Luney

39:47 - 40:52

Couldn't find the mute button. Yeah.

So so , I would say , you know , start with a business problem , not AI. There's loads of excitement around AI right now , but before invest in any solution , you know , be absolutely clear what you're trying to achieve , what is the problem that you're , you know , trying to , what the problem you're trying to achieve.

Again , I'd also encourage organizations to get the foundation right. AI is only gonna be as good as it , you know , as a clean data.

Don't try and do everything at once. You know , just pick one or two high value use cases where you can demonstrate success really quickly , bring those people on that journey , and build the momentum from there.

And , again , I think for me , the organizations that are gonna succeed with AI over the next twelve months won't necessarily be ones adapting , but it's about adapting the most technology. It'll be the ones that are applying it in the most purposeful way.

Claudia Lewis

40:52 - 40:55

Love that. And and , Helen ?

Helen Mairs

40:55 - 41:07

Yeah. I I think.

mentioned before , be clear on the use case. If you have your strategy in place , you'd be clear on the use case.

Talk to your end users about what they're looking at , you know , what they want and what they're look looking towards delivering.

Kelly Luney

41:07 - 41:07

Alright.

Helen Mairs

41:07 - 41:41

Build small use cases as well. So start off with those small use cases , and then you can learn from those in terms of delivering those projects and then build on that moving forward.

I think also be clear on ongoing costs because everyone thinks about the initial development cost. But , actually , what are the ongoing costs in terms of , you know , building , maintaining this ? If you've worked with a third party , do the costs sit with them ? Do they sit internally , etcetera ? So it's all very well building things , but , actually , the ongoing maintenance and also , you know , sunsetting any projects as well , I think you've got to be clear on that end to end journey.

Claudia Lewis

41:41 - 42:30

Guys , thank you both so much for for joining us today. I think we're really honored , transparent around what it takes , what the journey actually looks like , and and the considerations you have to bear in mind.

So , hopefully , the the audience is leaving with a lot to to think about. Few things that that stood out are data foundations , the human side of.

the work , and , of course , proof is in the pudding. Show people , you know , what you're doing and why it's important , and let them see the results of themselves.

So I wanna jump into a couple of questions from from the chat if there are any in there. Yeah.

So any advice on how to successfully make a case to senior leadership for investment on data and AI ? What works for you ? We've been talking about the cost a lot here , so I think that's a a poignant one.

Helen Mairs

42:30 - 42:58

If , if one thing senior leadership like , it's return on investment. So I think be clear on the return on investment.

So can I sell this ? How much how many hours is it gonna save ? What are the benefits of actually implementing , implementing it ? So be clear on the investment , like I said , from the outset as it goes on and what it is to sunset. Be really clear on that.

But , actually , what can we you know , what benefit it it gives to us ? So I think be really clear on that.

Claudia Lewis

42:58 - 43:13

Perfect. And then we've got , with how fast tech and AI is moving , how do you decide what to pay attention to and what to prioritize ?

Kelly Luney

43:13 - 44:08

I'll answer that one. Yeah.

It's incredible how fast it's moving. If you're not careful , you can end up really distracted by the latest headline rather than focusing on what genu is gonna create value.

Again , it always comes back to the business outcome. You know ? Don't start with technology.

Start with a problem that you're trying to solve. Is it improving customer experience ? Is it improving productivity , giving your teams better insights ? If it's a new technology or a k I pay capabilities , you know , does it support one of your business objective ? You know ? If not , it probably isn't a priority right now.

And I would say spend a lot of time listening to your teams , your customers , and your partners , and just listen about what's going on. You know , often your best ideas will come from understanding where the biggest pain points are rather than looking for that latest innovation.

Claudia Lewis

44:08 - 44:50

Brilliant. Love that.

Thank you again , Kelly , Helen , to to have you both with us. And for for anyone in the chat right now or , you know , watching , if if you want to have a further discussion on how to start your AI journey or wanna learn more about where Helen and Kelly are seeing the most value when it comes to to AI , my email has been sent into the chat.

So please don't hesitate to reach out. If you've got questions that you didn't wanna put in the chat , drop us an email as well.

I'm sure Kelly and Helen will be happy to to type out a little answer over email and get those over to you. So thanks again , guys.

Really , really appreciate everyone joining and and taking the time out of their day to to jump on with us.

Helen Mairs

44:50 - 44:52

No. Thanks for having us.

It's , been good.

Claudia Lewis

44:52 - 44:52

guys.

Kelly Luney

44:52 - 44:56

you. Have a nice day.

Thank you.

Claudia Lewis

44:56 - 44:56

Bye bye.

‍