Subscribe
Share
Share
Embed
In this podcast, we talk to some of the greatest founders and leaders about their journey to where they are as well as discuss their companies and many other subjects depending on the guest.
We are aiming to create meaningful content that everyone can get value from. We hope you enjoy 😁
you're listening to the Far Side
with founders and Leaders Podcast
the podcast that gives you a behind the scenes
look of some of the world's most
amazing founders and leaders
looking at their journeys
and how they got to where they are today
hello everyone
and welcome to the latest edition of the fireside
with founders and leaders podcast
today I am joined by the brilliant Chris Storey
who is a tech director
over at a company called Travelopia
so Chris has been there for over 11 years
so he knows the company inside out
and but he's seen some massive changes
especially over the last few years
where we've been through transformation
Travelopia is effectively
is a P E backed and owned organization
and so we talk a bit about sort of what it's like to
to be in that type of environment
but really delve deep
into how they've been transforming the business
utilizing AI to deliver more outputs
working with their software teams
their engineering teams
and empowering the rest of the business
to be able to get the benefits that they see
today through AI
loads of really
interesting insights and nuggets in this one
so I know you're gonna love it
all that's left for me to do is to let you sit back
watch and listen and enjoy
hello Chris welcome to the podcast
good to meet you yeah
good to meet you too a glorious
sunny day in London
in what is currently an air conditioned podcast studio
so a bit of relief from very much the sun
typically we like to complain about it being British
but yeah lovely to have you here today
thank you for making the effort
and the journey to come in
and we've had a conversation off air
talking about some of the things we're gonna discuss
today so
I'm actually
really looking forward to getting into some of these
topics it'll be of no surprise to to most people
it'll involve some AI
along the way
but the things that we've been talking about
your relatively sort of far into your
your exploratory journey um
and again we'll get into some of that later
but that
if you could start by introducing sort of who you are
a bit about Travelopia as well
for those that don't know about the business
and we'll go from there okay
great so my name is Chris Story
I'm a technology director for Travelopia
I've been at Travelopia for 11 years actually yep
as it happens yeah
yeah
in a few different roles and I'm a technology director
so essentially that means I spend
most of my time partnering
with the businesses in our portfolio
making sure I understand
what they're trying to achieve
what their business outcomes
what are their aims
and then helping to translate that into a
a roadmap and a set of activities
for our technology team um
I sit as part of the tech leadership group
and partner with uh
two other technology directors
so yup Shree who's in India
who looks after all of our engineering and product and
and Kaz who looks after our information security
and we all work for our CTO
who's Mike yup
so so yup
so so that's me and
and Travelopia most people haven't heard of actually
travelopia is a portfolio of travel businesses
we were created
I guess off the back of a sale from thwii
so most people have heard of thwii
yup large sort of German
um tour
tour operator uh
it was about eight years ago now I think yeah
we we had that about right yeah
that sale feels like a long time ago yeah
yeah lots changed and yeah
so we created a travel operator at that time so
so we brought together some of the smaller uh
compared to two
e specialist travel brands into a portfolio
and we've got about eight uh
business units yup in our group uh
about 20 or so brands uh
we serve about 230,000 customers a year
yup in them on trips uh
we're about $1 billion of revenue
um so sort of big enough to be interesting and exciting
but too small you know
so it's a nice sort of sweet spot
and yeah so some of the businesses in our portfolio
we have a business called Quark Expeditions
so quark do Arctic and Antarctic a trips on um
on the three vessels that we
that that we have in the fleet
which is amazing a TCS which is a private jet
so we have two uh Airbus A THREE TWENTY ONE jets okay
we will fit it out nice
and you can go on some incredible uh
trips if you can afford it um
but then we also have uh some yacht charter companies
so sunset
and the mornings
are two of the largest yacht charter companies
in the world we have Le Boat
I'm at I'm at the risk of listing every yeah yeah yeah
yeah and then I'll be in trouble for the ones I forget
yeah yeah
Exodus Le Boat
Enchanting Travels Kay Loss Golf yeah
so poor them lots of variety
lots of variety yeah
and generally experiential yeah
experiential sort of trips for people to go on
so do you get to go on any of these these trips as I
yeah I've been on a few yeah
um unfortunately not the TCS trip around OK
around the world uh
but yeah
it's a great uh industry sector to work in
yeah
I'm sure but there's plenty more on the bucket list
maybe at 12 years they they'll give you something
it's a commitment to service
we'll see we'll see
and you mentioned so P backed right as an organization
and we talked about a bit about AI transformation
it's been a big part of the
the work that you've been doing recently over the last
well certainly the last two
three years
I think even even you're further back than that yeah
um like
can you talk to me a bit about the importance of
your transformation in an organisation like yours
especially given that you're
you're pee backed
because it can't just be experimentation
you've got to be commercially focused and minded
yeah so talk to us a bit about you know
the the drive and some of the things you're doing
yeah so it
it it is
and whether we're talking about AI
or just transformation in general
actually that that that's a journey
particularly with AI sure last three
3 years or so
but actually if you look during our sort of ownership
the transformation journey itself has been fantastic
and and we
we often talk about three phases of transformation
that we've been on okay
you know the first phase was about dealing with
just some of the legacy
uh sort of tech debt type type stuff
so you know a
a lot of on premise uh
yeah infrastructure
old applications
not necessarily talking very well to each other
so the first phase of our transformation
was dealing with that mm hmm
second phase
we sort of refer to as the digital transformation bit
yeah so that's where we really focused on
no customer experience websites um
you know making the most of CRM platforms
getting the data plumbed in together
that now it feels like we're in the third phase yeah
which is the the AI phase right
yeah OK
yeah everything that we're doing in and around AI and
and you know
as you'd expect
I think most businesses are on this journey yeah
it's interesting sometimes when you speak outside uh
of your normal bubble right
and you sort of interact with other companies
I think we're I think we're doing well yep
our journey and but
but we're by no means experts yeah OK
I certainly can't sit here today and say I'm an expert
you know we know we have all the answers
we're figuring it out pretty much like most people yeah
I think at the moment yeah
I think that's fair but yeah
but we've got you know
a healthy mix I think
of experiments proof of concepts
so the curiosity mm hmm
emerging across the business
yep and
but then we have um
some things that we've scaled uh
plenty of things that haven't quite worked yep
you know I
I think a good mix but we're actively managing it as a
as a portfolio of change OK
and as a transformation journey in itself
so I think we're sort of positioned well
but there's so much
so much more to learn and to do and to achieve
it's gonna be an interesting journey
there's lots going on at the moment
as you say
I think like when you start actually speaking to people
you realize that no one's got a clue
everyone's still figuring it out
it's the reality of it all yeah
um
there are people that will try and sell you a product
saying that they figured it all out
but I don't think they have at all
they're just trying to sell you their product
yeah which is fine
like that's how that's how it goes in business yeah
um
and it's so
it's interesting to hear that
even though you're quite far on the journey
you're still in that
that sort of zone of like figuring things out
like working out like where does this go
but I suppose what
what's your view on sort of where things are going
over the next uh
I don't know
I don't know where to time it really next week
two or three yeah
the timeline yeah
we can joke you know
the the some it
it felt like so February time January
February this year we
we noticed that there was a
almost like an inflection point
so so something changed yeah
I think it was around February
there were a couple of um
anthropic and um
open AI models launched around that time yeah
particularly in our software engineering
so landscape really became truly transformational
I'm reminded of there's a
there's a great article um
Greg Schumer I think the kind of um
you know and he
and he
and he talked about this is in the middle of February
he talked about things are going to change
yup and he talked about how
you know for software engineers and
and software engineering and product development
we've kind of been on this AI journey now for two
two three years and
and as a piece of transformation
that's affected those teams significantly
right now it's about what
what next yeah
you know it's about finance
it's about legal it's about
you know every part of the organization um
and and so that's
that's where we are where
where are we are at
at the moment actually in Travelopia is that
you know we've been through huge transformation
of our technology function
mm hmm terms of how we go about software
engineering and product um
now it's about actually what else
what next where outside in the organization
and how do we take the learnings
that we've been through as a team
yeah and make sure that we can help to lead the charge
with the with the rest of the business nice
and when we spoke uh
previously you mentioned something to me that
that sort of stuck with me
you said the technology that's the easy part right
so I wanna dig into like what do you mean by that
because I think I know yeah
but it'd be great to try and sort of
understand a bit more about
at your sort of you know
drive again towards towards that sort of stuff
I think if if you look at it in the context of
where we are as a technology team
I think it becomes apparent I
when I look at AI now and the way we use it in our
software and product teams
I am
99.9% confidence
that we can solve any problem that the business has
okay and we can solve it really quite quickly yeah
ha ha um
and and
you know with the right sort of controls
and quality and guardrails etcetera yeah
as a really small example I'm
I'm not a software engineer
yep right
I'm I
I I can't write code
you and me both yeah
despite being in a technology role
and I occupy the space
somewhere in between the business and tech
and I'm quite happy there
but I partnered with um
with my colleague last week
yeah and in the space of about four hours
we wrote an application so I
I spent honestly spent about half an hour
I had an idea in the middle of the night
yeah um
and said you know
it'd be great if we could do this
spent about half an
hour going backwards and forwards with uh
with with a with chat GPT as it happens and
and just sort of built a prompt yeah
OK and then I needed a bit of help with the how do we
how do we make this happen yeah
you know whatever tool set
and in about four hours
I had a sort of message phone call and said
you know jump online and it and it was there and
and the product was built and they
they build it or did you use AI to to deliver the code
all AI yep
code hundred percent
and what did you use for that
was it hundred percent AI cloud code or cloud code yep
um
uh yeah
it was cool code yeah
yeah cool code
but nice I think
we're good at experimenting with lots of different yeah
yeah sure
it's cool code yeah yeah
um but that's the point right it
it is the from an idea to having and
and it's not 100% perfect
it needs some iteration some yeah
refinement but even a year ago
I could never have dreamt of just saying
you know can we build an application
and it's ready a few hours later yeah
yeah it's crazy right when you think about it yeah
that is and and so that's the
if you start to think about the power of that
then me as a non developer
non technologist really can just talk to the AI and
and and refine my idea
and then have it built in a few hours
if I can do it then anyone in our business can do that
yeah right
whether you're in a finance team
a HR team a sales team
you can do that
and that starts to present very different challenges
for us as a technology team at yeah
as an organisation
cause all of a sudden we've gone from
we're 200 people in a group tech team
and our job is to write code and build products
to actually that's not our job anymore yeah
OK job is probably much more about enabling
you know the right people in the
right organization to do that work
but then our role becomes much more about
a little bit of control good governance data
you know thinking about information security
etcetera etcetera yup
it's it's a
it's just such a dramatic shift yeah
yeah in mindset
so to bring it back to your point about the people
and the technology bit yeah
the technology is easy right
I'm you know
from my perspective I look at it
and I think we can solve any problem
that's happened to us the hard bit is figuring out
what does that mean to the organization yeah OK yeah
how does it change people's roles
how are we going to control this
it's not just chaos
how are we going to put the right guardrails
and controls and think about data
and information security
and all the super important stuff that we have to do
yep um
so yeah that
that's when I OK
I'm absolutely certain the technology is the easy bit
yeah makes sense
and look it's always the
one of the things
the challenges around any organisation
but especially sort of enterprise level
is the fact that you've got so many people
and trying to move all those components around
and make sure that
the right people are speaking to the right people
etcetera you talked about
you know maybe finance marketing
whoever it might be being able to pick up uh
you know an AI model and build something as
as an engineer would
but one of the things I see is actually sometimes
a bit of resistance from those people
they don't want to do it cause we get excited
cause we're in the tech industry yeah
um and specifically
you're in the tech industry more than I am
so you get probably more excited by
and people who are writing code
probably get very excited by the fact that
they can build something in
4 hours that would have taken them four weeks before
but someone in finance is not necessarily
they might be
but they're not necessarily as excited by that
so how do you
how do you navigate that sort of challenge
where you've got resistance from adoption
in a large organization it's really hard
yeah technology easy people hard yeah
really hard I
and I don't think we've got the answer figuring it out
but I think there's you
you said software engineers get really excited actually
we when we've been on our journey
lots of software engineers have got really
excited by it
but actually some of the software engineers didn't get
yeah that's fair right yup
and same with QA and same with product owners
now with fortunate we've built
I think a capability of technology
a travolopea that's
that's really strong with fantastic people who
you know
generally we've all leaned very heavily into it
like you say we're all technologists
you know so
so there's a there's a natural interest there
I think it's much harder if you then sit with
you know someone in a finance team
for example and say right
you know we're gonna start using AI because I think
you know in the back of their mind
they're thinking well
what does that actually mean
you know you've got the CEOs of all these big AI
companies coming out openly talking about
you know we're gonna
confidently eliminate 50% of white collar jobs
in four years time or whatever
you know and
and and it
it it
it's difficult right
and
and I don't think we've got an answer for what it means
I think at the moment we're very much in that phase of
we want to drive curiosity
we could talk a lot about curiosity right
we wanna drive curiosity interest
we want people to experiment we want people to you know
we've we've
you know we've chosen to equip people with with tools
mm hmm um so
you know a lot of our organization use copilot or clod
you know whatever the tool is
tomorrow it will be something different yeah
and and and we want them to experiment
but what we say is experiment
and then get to a point
where you recognize that you might need a bit of help
to to take it to the next stage and
and then that's where as a technology team
our role is to go in and partner with them
and figure out what we do
and the other thing we've done is
we've tried to identify in the organization
who are those really key people
who are gonna help to drive change yeah
so we refer to them as business technologists okay
which I think is a gardener term actually yeah
that that we um that we stumbled across
so business technologists are the people who are
you know
really great grounded understanding of the business
mm hmm and what it is you know
whether it's an operations team or finance processes
or whatever it is and
but they also have an interest a curiosity
a willingness to experiment with AI and actually
when you bring those two things together
that's really quite powerful because yeah
then
back to the point about somebody in a business team
building AI themselves yeah
you've
got the context and the knowledge and the interest
then you can do great things
so we've identified across the business
where those business technologists are yeah
and then really partnered them with our tech team again
to make sure that the guardrails and the safety
and the security is all there um
but it's a it's a really hard journey yeah
it's a really really hard journey and I I
I wouldn't believe anybody who
who says they know how to solve this
or even where this is gonna go in the next six months
and I think to your point of um
like some software engineers not being so excited by
I've I've seen and heard from
from various people so questioning like
what does the future look like
there's genuine concern from some people going well
if anyone can just pick up a
an AI tool and build what I can build
like what does that mean for me and my journey
have you seen that within within the business and from
you know wider wider sort of network as well
yes I think so and and and again
you know
software engineering in particular is the industry
the sector that's felt this change the most yeah
that's right so um
but what we've
we've done is we've evolved our operating model
so as a as a technology team
we used to have squads and that supported our business
so you might have a website squad or
a squad that supports applications
for particular business
and those squads would typically have a product owner
have some engineers they have QA
yep have some Devops help
etcetera etcetera
sometimes these teams were eight 10
12 people big right
and we've evolved that now so
so so our squads now are much smaller in nature OK
and much tighter so a product owner still um
but then generally fewer engineers
fewer QA and
but actually we have more of these squads OK
so more squads so not less people
not less people yup yup
more squads so of course
there's been some compression of our team
that has happened but
but we've been able to really focus on
how do we deliver more yup
rather than delivering less for less money yeah
yeah yeah
yeah it's
it's not a cost saving exercise really purely right in
in that play and
and it's
and you mentioned about sort of P E backing as well
I think that's where the you know
really focused on output delivering value yeah
it's important so we
we now say as as businesses
you know we do our strategic planning cycles
and build roadmaps and all that stuff
you know don't be constrained anymore by the technology
hmm dream as big as you possibly want
if you wanted to you know
recreate your business from scratch today
what would it look like
what would the technology look like to enable that
that's pretty much the starting point now yeah
as opposed to in the past where it would be oh
you know we're gonna be a bit constrained by budget
you know I'm not sure
if the tech team's gonna be able to deliver it
it might take six months or
you know those conversations pretty much now are gone
yeah it
it's about
how do you dream as big as you possibly want to yeah
and then
and then we'll create a team setup around it that
that helps to enable that um
by the way this might prove to be wrong yeah
yeah who knows what next week holds yeah
you know this
this is what we feel is the right thing to do
and that's what we're doing at the moment
but it it comes from experimentation right
yeah and we've been through iterations of this
as we will do over the next few weeks
and the next few weeks right and
but I think being willing and
and able and sort of open to experimenting
yeah in all of this has got to be super
super important yeah
and if you've got that kind of closed mindset
you're gonna struggle with with
with with all this change and
and so you seem as we said
like you're not perfect in terms of ways of working yet
and things can change um
but you're
you're quite far down the line in terms of your
your metaphorical journey with uh
with AI and
experimentation and moving beyond experimentation
into actually implementation
yeah of a lot of the things that you're doing
why do you think it is that you've been able to do that
whereas
I see still a lot of companies that are stuck in
experimentation mode um
and I've got my views on it
but I'd love to understand like
why you think that
the companies sometimes get stuck in that
that sort of mindset we
we we talk
it's a really good question
I think we we talk about sort of different phases of AI
and in most instances
it starts with what we call like personal productivity
so it's you as an individual
sat there with copilot or chat
GPT or whatever just using it right yeah
the hard bit is then
how do you scale that to something meaningful
because if you
if you're only focused on the individual productivity
it'll be great
you probably save yourself a few hours a week
because you're you know
so summarizing your emails yeah
maybe your content's a bit better
because you're using clout to build yeah
OK Powerpoints and that kind of stuff
but it's not gonna transform the business
yeah right
the hard bit is then taking it from that to something
which has a meaningful impact across the team
mm hmm right
so um
taking an idea proving it
understanding the value of it
and then partnering
with the technology team to then scale it right
that only
happens if you have the support from the leadership
so we again
one of our kind of principles we talk about a lot is
if this has to be LED from the top okay
so we're really fortunate right
we've we've got a CTO who's
you know really passionate about this my um
tech director colleagues equally passionate about it
so as a technology function
we've really been able to embrace it
because it's been LED from the top yup
and similarly now in our organization
our chief exact talks very openly about you know
we we want to be the best
hmm when it comes to the combination of
technology and humans right
and and really sort of helping to push the boundary
if you speak to the leaders across our businesses
you know equally so they
they believe in it and they create the time
and the space to make this stuff happen
and we talk about
you can't delegate the responsibility for AI yeah
one of the debates we have actually is
I use my example of uh
you know creating an app yeah
somebody who doesn't write code
we've believed for a while
that in order to truly understand
the potential of this technology
you have to experience it yeah OK
so we say to our senior leaders
you have to go and spend some time
really experiencing it
even to the point of come and join us
you know come to visit our team in India
you know come and cocreate something you know
sit there writing some prompts with claw code
you imagine
a managing director on our businesses doing that right
yeah but I think we think it's a
and it's not to say they're gonna do that forever yeah
not to say they have to suddenly
you know become software engineers overnight
they understand it a bit more though
they understand it a bit more yeah
and they understand the power of it
they understand the pace of it yeah
OK and it starts to open up their mind they go oh
wow OK actually this is great yeah
oh what you know
and then they're thinking what can we do with this
exactly yeah nice
I think it just releases some of the constraints yep
shackles now that that in itself is
is a hard conversation to have
it's like I'm a really busy person
what do you mean yeah
go and spend four hours you know
sat in a dark room
but but it but it but we think it's important actually
we've seen that where people have that aha moment
you know when the penny drops it's because they've
they've invested for eight 12
20 hours over a few weeks
to really partnering with our teams
and seeing what can be done yeah
so yeah so I think it's really about leadership
and it's about so trying it and understanding it hmm
and I think if you haven't got those two ingredients as
as a company yep
wanting to go on this journey
I think you're probably gonna struggle yeah
I think that's absolutely spot on
I mean my again
my view and things that I've seen is the
the organizations where the
the top are resistant
and then it doesn't filter down so much
you've got people you know
coming in at at sort of more uh
contributor level
and they're saying they wanna use these tools
they wanna go faster
they wanna do more and they're saying nope
you can't use it we're not getting you any license
it's just a slow burn sometimes
to try and get people to see the benefits of it but
and there's an also it's also an interesting debate
isn't there about to
to what degree you enable an organization to experiment
versus blocking it and locking it down
so yeah
I've heard of companies where it's like
no we're not
you know the tools are blocked
yeah or
or there's one tool you can use and that's it yeah
right
and we've been a little bit more flexible than that and
you know we have
we have preferred tools of course we do
we have quite a strict process in terms of
if you wanna use a new tool
you share it with us we'll have a look at it
we'll let you know what you think
and then we might enable it um
but that's the best way to enable experimentation
we don't know what all the best tools are
that's out there right
there might be something released last week
there's gonna be transformational of the business
that we don't know about yet
but somebody in our business does know about it
we wanna be able to enable that
as long as it ticks the boxes of yeah
safety and security so
you know we have a sort of again
a we have lots of frameworks to manage this stuff
as you can imagine but you know
we we think about
there's some people and tools
that we want to really be completely free and open
to experiment with yeah
right you know
and they're trusted people that we allow to do that
there's a bucket of people who are
you know
just happy to use the approved tools that we've got
right and that's fine uh
and then
there's a bucket of tools that are just blocked right
you know yup
it's just a banned list right
yup and that's fine
so we we try and kind of
then make sure our policies and training
and everything matches that kind of high yup um
but again experimentation that
that hierarchy the latest situation
that hierarchy
was something we discussed at the end of last week so
OK always sort of evolving as the
as the need the the tools change and everything
everything changes as we said
sort of week to week day by day
I'd love to talk a bit about some of the like
experimentations that you've
you've done if we can yeah
um like what's worked what's what's not as well um
so talk us through like say that that journey
some of the things that you guys have been
been working on over the last few years
yeah gosh um
so we've done quite a lot of work with our sales teams
actually which has been great
so if you imagine a a
a sales agent for one of our businesses
pulling together a proposal
you might have contacted us about yup
you know wanting to go on a an amazing trip
yup and that used to be quite a manual job actually
if you know we're gonna pull some
you know pricing information from this system and some
maybe some visuals from the website and
you know you kind of copy paste
you imagine copy pasting yup
stuff into a big word document
and then pdfing it and sending it out right
several hours work actually in some instances and so
so we've had some really interesting
experimentation around
you know using AI to do that right
so so essentially generating quotes using um
a sort of transcripts from calls with customers
where you've you know
cause
if I'm talking to you on the phone for 45 minutes
to build out your dream holiday
that then actually
we pretty much got all the information that you need
you just need to get it
you just plug from there in with some yeah
yeah and then
and then off you go so
so that that's been a real game changer for us um
similarly or in a related field
we've done a a
a really great
so the piece of work that we built ourselves
which is again
taking the transcripts of calls with with customers
but then using it to train a coach sorry
our sales agents OK
so you know
against the framework that we use like all good sales
yup
businesses will use a framework for for for selling and
and but then using the transcripts
and then scoring and helping to coach and guide and
you know which is a transformation again
for our sales managers and leaders in the past
they would have had to have listened in to
recordings of calls you could probably only do a few
yeah right
you know out of hundreds of calls potentially
but with the AI
it'll take all of the data from all of the calls
and then produce these really fantastic sort of outputs
yeah
that's a great example of something that started as one
idea that one person had in one of our businesses yeah
so fleshed it out built it
we actually showcased it
at one of our leadership conferences
you know everybody then wanted it yeah
now everybody has got it yeah
which is the
fantastic bit about being part of a portfolio actually
as you you know you
you can spread this technology out yeah
across businesses so yeah
so lots of stuff in sales
obviously software engineering we've talked about right
yep
transformational in the way that we've used it there
and you know
we're doing some interesting stuff in
in in the finance space as well
you know particularly we have a lot of these big
so generally big complicated finance systems
mm hmm and I
I won't name them but yeah yeah
you know you don't want to
particularly invest in their own technology yeah
their own AI cause it'll
it'll generally be to be a bit expensive
but but how can you get some of the efficiencies
by taking data out right
and yeah and doing stuff ourselves and um
to that and marketing
there's not a single part of the business right
there's not touching on everything yeah
touching on everything I think
those sales experiments are the ones that stand out
hmm
and because we've then been able to scale them across
across yep
some or all of our businesses
I can speak first hand from our experience
we're effectively a sales business as well
in recruitment
and we have the same sort of tooling system
where we use something to yeah
transcribe the the calls that the consultants have
and then look at being able to
to coach based on the transcription
and it's just saving my hours and hours of my life
which is great the other
the other example that springs to mind actually is
and it's kind of
somewhere between the business and tech
is we we have a couple of legacy systems
you know
one of them in particular is like 40+ years old
so good good green screen right
great green screen application
you then it's been challenging
right to
to drive change into an application like that
in the past because everything's just interlinked
you change one thing over here
it breaks something over here
none of the people obviously
who created the application are still in the business
the team's been through several iterations
so it's always just been a little bit too big
and scary to do something about
yep um
now the ability for us to either reverse engineer
some of that using AI
or the pace that we can start to build
functionality outside of the application
and just have it sort of talking
mind blowing again right
so we we had a team last week actually who did a
we call it a dark mode when they just go into
think of it like a hackathon
but OK yeah
two 3 days short
intense focus no distractions just
just focus on outcomes you know
and they're able to create these new
um screens
giving this all this new functionality
something the business had talked about for years
that they wanted
and just with a little bit of focus
yeah the right team
a little bit of focus the right AI in in a few days
they're able to smash out five or six of these features
you know
away from the green screen application and just
just just transformational yeah
really is yeah amazing
and what are the things that
that haven't worked out that you've killed alright
can you give us any examples of that
yeah we did we did some chat bots earlier on
mm hmm um
there was one that springs to mind
and I think
I think it was a so this was probably two and a half
two two and a half years ago now
so it was a while ago and we created a chat bot
whatever the LLM was that
that sort of under underpinned it
mm hmm and and it just it wasn't great
it was OK but it wasn't great
but also
it didn't have the right sort of buying adoption
understanding of why it was there from the
from the business itself yeah
so it
it was almost the case of we've built it because we can
yup not necessarily because the business wants it
or our customers wants it
so we ended up you know
we did good things like we tracked usage
and we could see how many chats were happening a day
and it just kind of
it just kind of trickled down to nothing really yeah
OK turning it off
so on on the face of it
it it just wasn't it wasn't right
or it wasn't easy yeah
something great experiment
cause then it taught us when we do our next chatbot
maybe we need to do something different yeah
definitely not wasted effort yep
but it's an example of something that just didn't
just didn't quite click yeah
it feels um
very much like two years ago
when this was all really starting to kick off
and that's when people were like
they were starting their AI journeys
and putting things in
and building things that people didn't want
need get any value out of
and we talk yeah
we we talk a lot about sometimes the answer is not AI
hmm by the way yeah
yeah sometimes the answer
it's often not actually I just
I just need this problem solving yeah
and we spend a lot of time coaching our
sort of business stakeholders to don't
don't come to us and say
I want this fantastic piece of AI that you've just seen
and then we spend a bit of time reverse engineering
well like
why what yeah
actually trying to solve you know
and you get back to the problem statement
and then build it back up from
once you got the problem statement well
what are we actually gonna do about it
might or might not be AI
might be the SAS software that you found
you know that somebody's tried to sell you
and or it might not be
it might be something that we can build ourselves
yup actually
cause we've got you know
strong capability
we can build it ourselves and that'll be quicker
cheaper in the long run not tied
you know yup
particular vendor etcetera so
um so yeah
thanks and
and you have to ask the right questions though
to get to those
those answers when you know that AI is not the answer
cause I think we get caught up in the hype bubbles
or whatever and it's constantly talking about it
talking about it talking about it
so you're thinking about it yeah
and everyone's trying to solve everything
with AI at the moment but sometimes it's great
as you say take a step back and think about
what's the problem we're really trying to solve here
exactly
let's actually look at what the best solution might be
rather than just jump straight into yeah
it's automatically AI yes
overall yes yes
um and
and have you seen sort of other instances whereby like
AI hasn't been the answer for
for things that you've you've been building
yes there are um
I I just have to have a little think about what
about some examples of that
what I do see a lot of though
is there's quite a lot of fatigue around AI
I experienced it myself I said to
I said to my to Sherry
my colleague last week I'd like
I just I just need to stop
I need I need a weekend
yeah so the Saturday
Sunday
long bank holiday weekend to just decompress a bit
cause I think the intensity
it's intense right
if you're in that AI bubble
particularly when you're leading
teams who are going through it
you you're also trying to transform the business
yeah it
it it's
you know it's
it's there's not a single conversation
that doesn't end up being an AI conversation
and there's a level of intensity to it
I think that that sometimes is quite hard to
to to
to deal with
see I'm not entirely sure that answers your question
yeah that's just something that kind of came to mind
that it's back to the human technology bit yeah
if you assume the tech is solvable on the human side
when there are people involved in this journey
either they're creating the software
or they're on the receiving end or
or their you know
their team processes are changing hmm
it's just a real sort of reminder dying
you know of making sure the people aren't burning out
yep being fatigued by it
but also just some good change management principles
right good
good 1:01 around
you know thinking about your people
the coms the yep awareness
the training the all
all the all the good
so the change management stuff you get taught um
is is
is really important and as a leader
is that how you think you can solve that
that potential issue sort of coming up
by just making sure you're putting structure in place
guard rails etcetera yeah
that's what that's what we're trying to do
so we've got a really good partnership actually
we talk about this is it's people and tech right
so it's as much as
about our tech teams as it is about our people teams
yep so we've got an amazing people team
you know and people HR directors across the group um
who who are part of this journey as well right
so whilst we're thinking about all
this is how we're gonna solve the tech hmm
you know the people teams will be thinking about all
this is what it means
this is how we're gonna solve this
for our organization these are the teams that
whose processes are gonna have to change
or the ways of working are gonna look different yeah
or in some instances
the structures are gonna look different
as we've experienced in the technology team
and so so it has to be both and
and I think they bring a really good grounding in
change management yeah
naturally there are a bit more people focused right
and and and so yeah
that's our that's our sort of formula at the moment is
is the combination of the two and and yeah
not not thinking about one in isolation
cause it's really interesting that balance
as you say cause there's
there's lots of change going on
in businesses at the moment
and I imagine most people teams are probably
quite busy yeah
trying to solve a lot of this
this stuff out and having to work with technology
um even more than they were previously to
to make sure that things are moving
in the right direction looking after the people um
that you worked with
especially when there's any transformation and change
that that will often lead to changing job roles and
you know changing headcount
things like that across the board
so definitely an important one to make sure that people
team are involved on a day to day basis to
to try and balance it yeah
um I've seen that other companies have
perhaps got that a little bit wrong
where they haven't involved those teams
and they're sat there going but
but what are we meant to do
and talking about all these people
creating all this extra work
so you may have reduced some of the
the workload by doing this
but you've increased the workload of the business
overall yeah
because you've created some sort of
it's not technical debt but some sort of debt
workload debt we call it
it becomes a it becomes a technology thing then right
yeah it comes a
the technology team have done this thing to us
yeah yeah
and then you battle against and it's back and forth
and we again
we talk about push and pull a lot
so we really don't believe that pushing technology
onto our businesses is the right thing to do
yup far
rather have a situation where they're pulling from us
and saying oh
we really like this we really need this
this is the problem we need solving
can you come and partner with us and and and do that
hmm and yeah
particularly as a group business
with a portfolio of businesses um
you know each of our businesses is fantastically
unique and different yeah
you know one size doesn't fit all
and from a technology sort of strategy perspective
we have to be OK with that
you're not just gonna be able to say
every single brand
is gonna have this bit of technology
and it's gonna work and that's a bit of flexibility
but the push pull mechanism is yeah
is really important for us
you can't push technology
and without the people team in the heart of it
it becomes a technology push
and and you talked a bit about um
you call it so got dark mode and yeah
the um
where you look at really focused time for outcomes hmm
um and I was wondering if you've seen a bigger drive
for more outcomes now
especially maybe from the top like board investor level
are they saying OK
well you're telling us all about all this great stuff
you're doing with
with AI and technology
we expect more outcomes coming down yeah
I think you know
as a as a business we're we're an ambitious business
yeah I think we grow every year um
and and there's always gonna be a
a pressure to to do more
I think the interesting formula
I guess is is to what degree you
sort of drive more outcomes through greater tech
so greater tech investment
leading to even more great outcomes
mm hmm we can do that
but does that mean we should do it
because actually as a
if you're a particular business
you know
we could churn out millions of lines of code a week
if we wanted to right
doesn't necessarily mean we should yeah
yeah because it
it it's only good if the business is ready to
instead of a
partner with us to make sure it's the right thing
and then b you know
drive the adoption or drive the usage
or drive whatever it is
to make sure that the outcome is
is is delivered yeah
because there's no point
if there's no business outcomes right
we will know that so
I think the pressure for outcomes is probably the same
as it has been talked about
our different phases of transform
the transformation journey right
whether it's transforming legacy debt
whether it's digital transformation
or whether it's AI
there's always a pressure to make sure it leads to some
we shouldn't be doing it right
if it doesn't lead to some kind of tangible outcome
doesn't have to be about increasing revenue
or reducing cost it could be in our business
about improving customer experience
and then there's sort
of secondary and tertiary benefits of that
cause in theory that drives yep
retention or or repeat
you know increase repeat rates
etcetera and so
so that focus on outcomes has always been there
I think the difficult bit now is
is is balancing actually
cause I think we could deliver so much in theory
there could be many many great outcomes for a business
but can we cope with that yep
change is a is a question Mark right OK
to to be decided yeah
I think again again
another one of those things that we'll have to
look at in the future when
when things change again yeah
yeah and
and the fact that you guys are p backed
as an organization there's
there's differences
from the things I see in terms of how you can
implement experiments
my do things comparatively to a definitely a startup
right but even somewhat in that some angel invested or
or VC backed organizations
there are different ways of working
so what are the and again
appreciate you've
you've not necessarily sort of worked in
in the former environment um
so but from your experience
working under a P backed
environment over the last 11 years
like what are the things that you see
that you guys need to do because of
you're in that environment
comparatively to someone else who's not
I think it's a great question
I I think the
there's a higher expectation of outcomes
mm hmm so so I think there's a
there's a higher expectation that the
investment that's made in our technology
actually drives a meaningful business impact
right or
or or outcome
whatever that measure is um
I think equally being part of that sort of environment
where you're working
perhaps with other portfolio businesses in the
in the P um
sort of fund gives you access then to
to insight gives you access to partnerships with
whether it's with you know
the big AI companies or or
or someone else yep
that you
obviously wouldn't get if you were in a sort of smaller
smaller startup environment
and I think it also attracts talent in a way that
that perhaps would be difficult
so I think being part of
it's a self fulfilling prophecy
right if we
we've been on this fantastic tech journey at travel
yeah
we've really transformed everything from our tech stack
and our team and the way we do it
so it's a very uh
attractive place to work I'd say for
for for technologists
um and
and and I think that wouldn't have been possible
if we didn't have the backing yup um
of you know of
of our board and our owners ultimately um
because because they see the value of it yup
and I think that's the the other important bit is
you know whether your
whether it's a startup or
you know um
um private equity or VC or anything else
if your leadership don't believe yeah OK
the power of technology
or more recently in the power of AI then
then yeah
you're not you're not gonna get the investment
you're not gonna be able to
have the freedom to experiment
cause there's gonna be different types of pressures
yeah
I don't know if it's a sort of a good answer really
but I think it's all of those things come together
mm hmm um
to to
that to sort of make it a slightly unique environment
yeah um
but the single thing that stands out is the buying
from the leadership theory yeah
I think without that buying from the leadership
and the understanding of the need to experiment
you wouldn't particularly in an AI world right
you have to be able to experiment yeah
which makes sense
and I suppose if you're working in a smaller startup
you get the buying from the leadership anyway
because they probably own the business
and they're doing it because they're bought into it
whereas
it's different when you've got layers and layers
of people above and boards and things like that so um
yeah I definitely think that's
that's something that's probably a a fair answer yeah
uh given the things that we've seen and
and heard over the last few years
and I suppose in terms of like
where things are going as well
and here's here's the crystal ball moment right
get it out um
but what do you think that you guys
will see over the night next
let's call it six months
because I think looking any further in the future yeah
it's probably unrealistic
and it'll have changed and be outdated very quickly
but certainly
with the things that you can do to get more outcomes
and deliver better outcomes outputs
etcetera with the way technology is going
what are you what are you seeing
what are you hedging your bets on at the moment
there's probably two or three two or three areas um
I think the first is
I talked about individual productivity before
I think for us at least the quality of the models now
and I'm certain the next anthropic or open AI
or Microsoft models will be even greater
the power it gives to individuals actually is
is is quite mind blowing right
so even a few weeks ago right
we were talking everyone was excited about Claude right
cause yes
the the number of show in town and then you know
now you can actually get some of the claw models inside
copilot which yeah
great for us because yeah
we you know we're a Microsoft um business
yup and and we have lots of copilot users
so it gives us all that sort of Protection and safety
um you know
within that Microsoft estate so it
it seeing over the next six months
I think for us it's gonna be key to see how people's
individual productivity really starts to
to ramp up
because they're realizing the power of the tools
and actually I'm not hearing so much
actually a year ago
I had a lot of oh
I used I used AI once and it wasn't very good
or I used this tool or that tool and it wasn't great
and so I've kind of stopped using it
I I don't hear that anymore
people that we've recently added to
you know having co pilot
for example they
they they don't say that right
that because the quality I think of the output and
and just the maturity I guess
of the models yep
means they can
they can get the value straight away and they see it
so I think there's definitely
next six months
is something about individual productivity
and really turning the dial on that yep
I think the second thing is then
from a software engineering
and a product development perspective
how do we get our
sort of operating model working really well
consistently across all of our businesses yeah
obviously not all of our businesses is the same
some are more mature some are less mature
in the way that they sort of work with us and and
and their ambition around AI and but
but getting that operating model to a point where it
it it just works across all of our businesses
yep so we talked about smaller squads
tighter teams you know
that dark mode idea all those things coming together
but really accelerating it so
so challenge for us as a tech team yeah
suddenly everybody wants everything
yeah and so we're trying to figure that out
but but that's
that's the sort of second
third thing I think the final thing is then as a
to the technology leader I think it
it the
the owners is on
you know me
the team to
to kind of be a step ahead of everybody else OK
in our organization
rightly or wrongly
everybody looks at the tech team as well
you guys know AI so what should we be doing next yeah
what's the you know
what
what does this new release from anthropic actually mean
you know so this part of the challenge
I think is is how do we
how do we scale
what we're trying to do with all our businesses
whilst at the same time
just trying to stay a little bit ahead of
the rest of the business yeah
OK understanding
so
how do we invest a little bit of time in understanding
the model that was released yesterday
or the new tool that's on town
in town that's and
and that's hard right
cause that that
cause you're trying to run really fast over here
and you're trying to stay ahead um
so so yeah
but it's just impossible to predict right oh
I think the only thing that is predictable is
you know every three to six months
it just feels like it's getting better
and faster yeah
right
I I can only assume that's gonna carry on going right
so in a world where everything
is getting better and faster
and every model is you know X
y Z what does that mean for us as a business
you know I think we've
we've often thought actually as a technology team
we're really quite efficient now
I don't know how much more efficient we can be
with all these new models
I'm more interested
in what it means for the rest of our business yeah
so how do we really start to use the power of these
models in our finance team
or our legal team
or our HR team or our marketing teams
you know that that that's the real power because we're
we're ahead of the journey yeah
OK as a tech team
it's it's how we scale it more consistently
across the business
where do you think we are generally speaking as a uh
society I guess big question here yeah yeah
uh but just on the maturity levels of
of AI from where we've been
cause you mentioned obviously people not adopting it
cause they go
I use it once it's a bit rubbish to now going actually
it's alright I can yeah
I can use this and get value
so where do you think we are
if we were saying sort of some of
a scale of one to ten
it's um I wouldn't even know how to put a number on it
I I suspect we're like one or 2 yeah OK
really low down I
I would I would think so yeah
when when you just think it's such a relative small
relatively small amount of time right yeah
since we first heard about chat GPT right
which is for most people was probably that yeah
entry point so what was that three years ago yeah
so a bit more yeah
so in the last three years you look at where we've
we've come I
I can only assume therefore in the next three years
six years yeah
you know it's gonna continue um
I think the harder bit for society
is when I sit there and think about
what do I tell my kids yep
right when
when they're growing up so I've got three boys
yep Alfie's 11
Jake's eight and Tom is five okay
you know when
when I look at them and think
you know gosh
in 10 15 years when you're
you know you want to go to university
or you want to do your first job what
what what's that gonna look like yeah
because I I
you know
I don't think I'd be saying be a software engineer no
I'm not sure
I'd say you can probably list off all the jobs right
I I think that's the hard bit
I was chatting to our
we're having some work done at home at the moment yep
an extension built so I was chatting to our builder um
about this topic actually about AI and yeah um
what you do and and
and and it's
you know do you
do you advise your children to go into something
that's physical labour because actually
we'll always need builders or plumbers or electricians
and going down that kind of apprenticeship route is
is is
is a stronger safer career and
or is it is it actually overhyped
yeah and actually
of course we'll still need software engineers
because the world just looks slightly different
for a software engineer
cause you're partnering with throw all changes yeah
yeah you're partnering with a number of agents and
and you you're delivering more output
but ultimately
you're still gonna be the human in the loop who's yep
orchestrating all this stuff
and so I think and
and it's it's
if you listen there's just so much material out there
right yeah
when it's talking about AI like we are now
right that you read
that's either
a real doomsday scenario about what it means for
human existence or it's
you know other end of the spectrum
it's a load of hype
and it's all gonna go away in a few years time right
yup it's clearly not either of those
it's probably something in the middle hopefully
yup something in the middle right
um but it still means quite profound changes
I think hmm
as a society yup um
and yeah I
I ground myself the sort of checkpoints along the way
and grounding is what am I telling my children yeah
OK when they're saying
I want to be a you know
an a a a
a a scientist when I'm older
I wanna be a maths teacher when I'm older
I wanna be and it's always a well
you know can I
how do I foresee that might look different yeah
in you know 10
15 years time what do you think we'll
we'll look back on in maybe like five years time ago
we misunderstood that about AI
I think there is a misconception that AI
can do everything without a human
mm hmm I
I don't think we've quite as a society or even in a
in a business sense actually
um or even at travelopia quite
quite grasped yet the the
the power of
the combination of having a human in the loop
but then the sort of you know
partnering with an AI a set of agents
yep whatever
that can deliver vast amounts of
of work and output but still with the human in the loop
and I think there is a misconception that
all of that can happen
without having a human in the loop and
and I yeah
I think it works in some instances
it doesn't work in other areas
and some people understand it
some people don't
but I worry that there's a perception that the AI
will just do everything without ever needing a human
so I don't fundamentally believe that's true
the AI doesn't always get it right right
no of course
see yourself I'm sure right
and and it needs that back and forward some fine tuning
yeah to
to get to a point it's like ah
now now it's saying what I wanted it to say yeah
and so I think the power of the human in the loop
I'm not sure has is
is really very well understood by everybody yet yeah
I think that's it's a key point
and I think people need to
need to remember that
and have it in the forefront of their minds
that we're we're very good at making change
and we create problems by solving problems
etcetera as
as humans but we always have gone through cycles
where the internet was one thing
where it came and it was gonna change everything
there'll be no more jobs
there'll be no more shops on the high street
with e commerce it's still there
exactly right
we're not gonna destroy everything
no uh
and I don't think the governments would let us no
as well no
so no rightly so do do their jobs yeah
um so yeah
I think we're I think we're into
a position whereby we can get the benefit from it
and use it to help us as humans ultimately yeah
well Chris
thank you very much for for coming on the show today
it's been great having you
I think we've
discussed pretty much everything we can have
yeah so solved it
move on what's the next thing
but thank you again for coming in
and really enjoyed talking to you
likewise thanks very much
you've been listening to the Far Side
with founders and leaders podcast
we really hope you've enjoyed this episode
if you have
make sure that you hit the subscribe button now
so that you can get alerted
every time that we launch a new episode
meaning you'll never miss your favourite session
we really hope you can join us next time
and thanks again for listening to the fireside
with founders and leaders podcast