August 20, 2025 • 27 mins
In this special bonus episode, Georgie Holt (co-founder of Flight Story alongside Stephen Bartlett) takes us behind-the-scenes to look at how her team is redefining media and podcasting using AI.
Georgie shares the story of an internal AI competition that led to the creation of 48 new tools, saving over 62,000 hours of work and $1 million in efficiencies. She takes us inside the tools powering Diary of a CEO, including “Unicorn Hunter” and “Prewatch,” and explains how AI helps the team identify cultural moments, book high-impact guests, and tailor content to maximise engagement.
You’ll learn:
- How Diary of a CEO finds the right guest at the right moment using AI
- The strategy and psychology behind a company-wide AI challenge
- How Diary of a CEO uses data to refine content before it even goes live
- The importance of community and loyalty in building media that lasts
- Georgie’s candid feedback for How I Work
This is Part 2 of my chat with Georgie Holt. Be sure to listen to Part 1 if you haven’t already.
Follow Georgie on LinkedIn: https://www.linkedin.com/in/georgie-holt/
Listen/watch Diary of a CEO: https://www.youtube.com/@TheDiaryOfACEO
My new book The Health Habit is out now. You can order a copy here:
https://www.amantha.com/the-health-habit/
Connect with me on the socials:
LinkedIn
Instagram
Want more tips to improve the way you work and live?
Subscribe to my weekly newsletter for practical, science-backed strategies:
https://amantha-imber.ck.page/subscribe
For full show notes from all episodes, visit:
https://www.amantha.com/podcast
Get in touch: amantha@inventium.com.au
Credits
Host: Amantha Imber
Episode Producer: Sam Blacker, The Podcast Butler
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:01):
Imagine if you could predict with data which podcast guest
would blow up your audience before you even invited them.
Georgie Holt's team at Flight Story built that tool, and
it's changing the game for Diary of a CEO. In
this second part of my conversation with Georgie, she takes
(00:22):
us inside the twenty thousand dollars AI Agent challenge that
transformed how her company works, creating forty eight tools, saving
sixty two thousand hours and sparking a culture of cross
team breakthroughs. She reveals the Unicorn Hunter, the AI driven
guest radar that tracks cultural trends in real time to
(00:45):
identify the voice as everyone will be talking about next,
and she shares the behind the scenes system they use
to test sometimes six hour long interviews with thousands of
people before they're released on Diary of a CEO, which
kills the guest work and make sure that every episode lands.
Speaker 2 (01:06):
And if you missed.
Speaker 1 (01:07):
Part one of this interview, I'd recommend going back to
that first. It'll give this conversation even more depth and context.
Welcome to How I Work, a show about habits, rituals,
(01:27):
and strategies.
Speaker 2 (01:28):
For optimizing your date. I'm your host, doctor Amantha Imber.
I want to talk.
Speaker 1 (01:37):
About the people now and the challenge that you're co founder.
One of your co founders, Steven Bartlett set and I
think this happened maybe two or three months ago now,
which was an AI agent competition.
Speaker 2 (01:49):
Across the entire organization.
Speaker 1 (01:52):
I heard there was a twenty thousand dollars prize for
the winning team, and you basically asked every team and
I've got the words.
Speaker 2 (02:00):
Can you tell me this is correct?
Speaker 1 (02:01):
What could you automate redesign or reimagine if AI was
your co partner not your competitor. Tell me how that
competition played out? Like what, like what happened?
Speaker 3 (02:14):
Okay, So Stephen is a genius. He had this idea.
I think it was classic US. It was sort of
a two am idea. And you know his sort of
first principle is when you express cognitive dissonance around anything,
when you sort of feel attention mentally around particularly a
new technology tool future state, it's a signal to not
lean out, is it's to lean in. And we've discussed
(02:38):
at length one of the biggest mistakes that are board
or a leadership team will be making is to not
talk about AI just sort of to think of it
as something that couldn't possibly compete, It couldn't possibly be
something that could be better than them for us, and
not to be talking about that as a company and
not to be looking at its potential and the opportunity
it creates us would be a huge mistake. And we
(02:59):
all completely agreed immediately, and he said, I want to
set a challenge for the company, and I want to
run a competition. It's a great way to bring teams together,
and our teams love to compete. We're extremely competitive people
in the best possible way. When we socialize and spend
time together, it's usually in some type of competitive environment.
And I think we went through a range of emotions
at the beginning. I think there was excitement. I think
(03:20):
that there was nervousness, a little bit of fear. I
think that's absolutely fair to say, because you might be
building something that quite frankly, could replace you. But we
all came to very quickly to the understanding is that
the best skill set in the world is actually someone
who knows how to direct the AI tools or the
tool itself. And you had to know what great looks
(03:40):
like in order to be able to qualify with whether
you're building something that is good. So I think we
went through a gamut of emotions and we got into
our teams, and then it got competitive, and then we
would also set hang on well the research, you know,
see the research team at nine pm at night working
on them at all, and then the social team will
(04:01):
be like, or we'll be in at seven am tomorrow
working on ours. And it was extremely energizing, a great
culture experience as well, because it was time bound. It
was clear we all knew within a few days that
we were building things that could radically change and improve
how we worked. And I think people could immediately see
(04:22):
the real world impact. And it didn't matter whether you'd
been working for twenty five years or two years. Everyone's
contributions had impact and meaning then everybody was able to
share and build collectively. And we're sort of working cross
functioning as well, which is really interesting, working with people
we hadn't worked with before. And people started to build
(04:44):
and they started to make breakthroughs. So the experience of
breaking through something I kind of took away. I was
like that it's actually experienced that we don't sort of
put in teams enough because it's rewarding and you actually
feel like you're making progress and sort of the reward
and progress making are so important in human psychology to
kind of feel that you're working with purpose and you're
(05:06):
pursuing something. You're pursuing something important and it suddenly became
a pursuit. And we hosted an amazing day on the
sort of the end of the competition, and we got
every team to present their ideas. We had independent judges
come in, We have the Chief Technology Officer WPP come
in and judge the work. We had our director of
(05:27):
Innovation and our lead data scientist and Stephen reviewing all
of the entries and they were judged on multiple scores,
not just obviously efficiency impact, but social impact, human impact.
It's a usability. Is this something that we would be
proud to say that we use that We wouldn't sort
of be hiding it that we use an efficiency tool,
(05:50):
but we're actually saying it's the human beings working on
something that we can see it would make an impact
on our purpose and long term goal. It was such
a powerful reminder that this industry of media still attracts
the most creative, the most curious, the most driven, and
the most passionate people in the world. Is just the
game has slightly changed and we need to play it
(06:12):
and play it better than anybody else. And it was
a wonderful experience, probably one of the greatest I've had
in a very long time, watching everyone present the tools.
And I think we made forty eight tools across all
the different departments. We say sixty two thousand hours, which
is that seven years of efficiencies found at about a
million dollars of cost impact which we can now reinvest
(06:33):
in other things.
Speaker 1 (06:33):
I love what you said about just that experience of
teams making a breakthrough and making progress. It reminds me
of research from Professor Theresa and Marbala that shows that
the thing that leads to the biggest impact on engagement
and motivation at work is having that sense of progress.
And I love how you described you know, that process
(06:56):
that teams went through.
Speaker 2 (06:58):
I have to ask what was the winning idea or was.
Speaker 3 (07:00):
Research and guestbooking, So they had four ideas in total.
We have evidence to show that the impact of a
great guest can be transformational on so many different critical
factors in the company, and it can be as high
impact as the audience. Having a transformational experience or making
(07:22):
a huge life choice or changing the way they eat, live, think,
have opinions or ideas and particular topics. Start a business,
so we can have a profound impact on an audience,
but it can have a profound impact on the metrics
of what we measure at flight story in terms of
how it is growing. So that can be show growth,
it can be revenue growth, it can be subscriber growth,
(07:44):
it can be views ABD so we can see that
the guest is one of the greatest transformational decisions that
we can have, but it creates extremely high human effort,
so it is a lot of researching, sourcing, emailing, scheduling,
(08:04):
and then of course once that's put through researching, that
has to go in. So it's one of our most
high human effort but high payoff areas of the business.
So they've got four different tools that could improve the
efficiency of the process. So they've built a tool called
Unicorn Hunter, which is going into a tool that we've built,
which is guest Radar. So we've built we'd already previously
(08:25):
built a tool called guest Radar, which essentially now can
identify a moment at which a channel whether it's YouTube
or on social media, has somebody appear on that channel
and it overperforms versus it's sort of i'd say current state,
and it will point towards who that person is and
what they were talking about. So that tool has existed
(08:47):
and it is extremely effective and I think completely unique
and proprietary in the industry, and we can now look
at a guest and say, do they overindex on performance
when they appear on a show where they kind of
standard We had a great example Withdr Eric Weinstein who
went on The diver CEO, who had eleven x the
performance on any channel that he went on, So he
(09:07):
was performing eleven times and the standard performance of a channel.
So we booked him on the show and again had
exactly the same impact on the divers O. So we
kind of had this phrase of killing the guestwork. What
was happening is sort of finding those unicorns still required
human effort because you're kind of looking for cultural trends,
high impact moments in news and culture and society. So
(09:29):
they built a layer onto that tour that was now
monitoring culture and social trends to understand what topics were trending,
what was not just uniquely interesting to the world in
that moment in time, but who was uniquely interesting to
the world at that moment in time, and then being
able to action that decision immediately. So Steven's testing out
new debate formats on the back of this idea and
utilizing this idea to find voices you can immediately speak
(09:52):
on complex topics that human beings are curious about and
trying to solve and have opinions on. So we've had
debates world War three on feminism, on lost boys, you
know how young men are being completely a finding sort
of societal struggle incredibly difficult, and the debate format needs
(10:13):
people who can speak with immediacy and expertise on a topic.
And now that tool, the Unicorn Hunter, is built into
guest Radar to help them find it by monitoring culture
and social trends on the Internet and voices who are
speaking on that topic. So it allads an immediacy to
the guestbooking team that they can now sort of say, hey,
we've seen that you are extremely well first on this topic.
We can see you're talking about it right now. Can
(10:35):
you be in LA in two days time to talk
about it with Stephen? So the speed is extremely fast.
Now they built out research tools and briefs that can
deal with extremely complex research topics, which then have a
human lens going over the top. They were the ones
that really stuck out to us with the research and
the Unicorn Hunter.
Speaker 1 (10:55):
So far, you've heard how flight Story turned an AI
challenge into million in impact. In the second half, Georgie
reveals the tool that lets them test a four to
six hour interview with thousands of people before it's released
and how that data changes everything. Plus she gives me
(11:16):
feedback live on air about what I should be doing
with this very podcast. If you're looking for more tips
to improve the way you work can live. I write
a short weekly newsletter that contains tactics I've discovered that
have helped me personally. You can sign up for that
at Amantha dot com. That's Amantha dot Com. I have
(11:44):
to ask, because I know that before you co founded
flight Story with Stephen, I think.
Speaker 2 (11:50):
You and your business partner Christiana.
Speaker 1 (11:52):
Yeah, Christiani listened to every episode of Diary of a CEO.
Speaker 2 (11:56):
Am I right in saying that.
Speaker 3 (11:58):
I will be completely honest, Christiana did. I would find
very high interest in the business and the entrepreneurial side
of his guest, and of course I could see an
episode was sort of important in the culture zeitgeist. I
would listen, but I would say I was listening at
least one episode a week. I would definitely be that
Christiana was listening to everyone.
Speaker 1 (12:19):
I've got to ask, then, like, what would be your
top three if you had to.
Speaker 3 (12:22):
Choose top three episodes? Dr rris wat I think she
is a generational voice, and I think she radically changed
how people thought about stress and the impact of stress
on their life and the things that they could do
to move them out of stressed states and into a
more balanced way of living. I absolutely loved Dr Christen
(12:44):
Holmes's episode on circadian rhythm and her focus and relentless
commitment to human beings being able to find the democratic
and easily accessible ways to improve their health and well being,
and her ability to try late again complex science and
understand deeply and intuitively how to translate that to audiences
(13:07):
to help them understand how to better live, to just,
for example, the moment you wake up, we should get
natural light into your eyes, and how important it is
to take and make those decisions every single day, and
the compounding impact that can have over time. We absolutely
loved her episode. I loved Divina mccaul's episode as well,
which was again her story through addiction. I don't think
(13:30):
we'd ever heard her talk like that before. It had
a profound impact on me. I was lucky enough to
be in the room with Missus Michelle Obama's podcast episode
as well, just to be in her presence and to
experience I think her greatness honestly, and to watch a
woman who has had the eyes of the world on
her twenty four to seven ever since Barak decided to
(13:50):
run for office, to still come back to humanity and
love and childhood experience and family, to kind of hold
those principle so close to her despite everything was one
of the greatest experiences of my life to kind of
watch that happen. I think a lot of people talk
about no Gowdat's episodes the AI Emergency, which was you know,
(14:11):
Stephen was one of the first people to sort of
bring that into the public consciousness. This is something we
should be paying attention to. And if we don't pay
attention to it, if we don't become good stuarts and
parents to AI, what will happen how will it now
move forward? And what that was well over two to
three years ago. I think so he was bringing it
into the attention of the world. I think sort of
(14:31):
Moo's first episode on Happiness. I think people mentioned a
lot because it again his personal story and how we
shared it with Stephen, and how that marriage knitted up
to his purpose. Again, I think had a profound impact
on people. And I think my dream guest, if he
ever had it on all the way, I don't know
whether he would ever happen. I would be Dolly Potton, I.
Speaker 1 (14:52):
Know that about it, But I think we'd have to
find his interest point in that.
Speaker 3 (15:01):
When we look at Yes to Burkett, it comes back
to it a very simple framework, which is, is the
audience interested in this topic or person? Can we prove
that through evidence and data? Is this person the best
person in the world to talk about their topic or
their life experience? And do we have evidence and data
to prove that? And as honestly, the host, are you,
(15:23):
Stephen or any of our new show hosts interested enough
in both that topic and that person to sit down
with them for six hours maybe even longer. He dedicates
you know, the highest portion of his time to recording
these episodes. It's not a thank you for coming in
for ninety minutes and we'll shake hands and appreciate the thoughts.
It's sometimes sort of six, seven, eight hours of conversation.
(15:45):
I do know one that went on for eight hours
in total, and then it's edited down into two to
two and a half hours of conversation. So there has
to be extreme interest from both those people in that seat.
I go back to the Mister Beast episode of being
One of the greatest examples of that is that I
think the story has been told already, but he canceled
the morning of the record because his flights have been canceled,
and he'd been a guest that the team has been
(16:07):
working on for years. You know, when these people arrive
in the seats of these shows, it's not one email exchange.
I think there's sometimes two to three years of work
that have gone in. I think with a doctor Andrew Huberman,
I think it was two years before we eventually were
able to schedule it and move towards having him in
the seat. So it's not sort of oh high, where
(16:27):
the divers, Yo, do you want to come on the show.
This team has compounded the impact over years. You know,
Stephen is a success he is now, and that show
is a success now because of the compounding dedication that
team has had, and many years ago they weren't the
biggest show in the world. They were on their way
to be it, but it wasn't that. It wasn't at
that point. So these episodes take years to come to life.
(16:48):
And mister BEA's canceled in the morning is playing a
big canceled and he was recording the finale of Beast
Games that afternoon. So for all intents and purposes, this
was an impossible task and the team just didn't believe it.
They just we don't believe that this is impossible, and
I believe that we can make this happen. And they
made the impossible possible, and they ended up recording with
mister Beast. I think at two am in the morning,
(17:08):
after you've recorded the finale to Beast Games, and I
think they went on till like five in the morning,
recorded all night. Jimmy is one of the most extraordinary humans.
I think he then took Stephen to seven eleven to
show him his chocolate because he was again as he's
obsessed with the one percents and that's the level of
commitment it takes. So I think those guests kind of
stand out for all different unique reasons.
Speaker 2 (17:30):
Wow, that is quite the story.
Speaker 1 (17:33):
I've also heard that there's a process called I think
it's called the pre watch where yes, you take that
six to eight hours of sound and you get that
into two to two and a half hours. Can you
tell me what that involves exactly?
Speaker 3 (17:46):
Yes, It again goes back to one of our first principles,
which has killed the guest work. You don't need to
guess anymore, and we have so many tools and opportunities
to not have to do that. And one of the
even though YouTube and YouTube studio is one of the
greatest data sets you can get as a creator, and
it's incredibly important in terms of understanding audience behaviors and
(18:09):
understanding when people are watching, how long they're watching for,
when they stop watching, when they start watching again, when
they share things, when they comment, and buying things interesting.
It was still after you had done all the work,
So you'd have done all the work on the episodes
you've recorded with mister Beast at two in the morning,
you'd have moved maybe impossible, possible You've just spent two
(18:31):
years getting a guest into the diver Ceo chair or
any of our new show chairs, and you still don't know,
and you're still you've done everything, and then you publish
on YouTube and you hope that everything you've done is right.
Stephen was very frustrated by that, and also he obviously
presents on TV shows and there's no methodology and science
and TV there's you know, quite often you record months
(18:53):
and months previously, they then do all the editing, and
then six months later it goes onto the TV and
it knows them. It's like we did the ratings, and
how can that be? That is a six month gap
between what we're recording and what happened now. So he said, well,
what if we could build something that could give us
data and insight about what people enjoyed, what they found interesting,
Because wouldn't it be a shame if we put our
(19:16):
own human bias on what was interesting and what was
not interesting because of our own lived experience, Because you,
as a forty five year old woman and me, as
a thirty one year old man, had uniquely different interests
in that episode, and I edited it through my lens
and I completely ignored an entire section of that content
that would have been extremely interesting to an audience that
(19:37):
maybe I didn't understand or I didn't have the lens
of experience or empathy around, and we can't edit that way.
So he built a talk called pre Watch, which means
that we have anywhere between sort of five to seven
thousand people watching episodes of the Diarver CEO, particularly where
across a cohort of people, we maybe don't have a
(19:57):
unique experience of that person's expertise or area or field
of work or human interest story. And it's eye tracking software,
so it can look and understand. We can see when
people have looked away, so it can see where attention
may be dropping. And then people can interact with the
tool as well. They can comment, they can say that
(20:18):
something was interesting. You get this incredible interest cup, and
you'll see that there's high interest at the beginning, which
you would expect to see on YouTube do anyway. You know,
there is natural drop off over time, but you might
see something that happens in the middle of the episode
and at the end of the episode that has the
highest interest of all and you're like, well, we definitely
want to make sure that we talk about that in
(20:39):
the trailer, we definitely want to make sure that's in
the packaging of the episode. Perhaps a section that we
thought was very interesting hasn't had the payoff of the
audience that we wanted it to have, maybe in a
particular demographic, because we can understand the demographics of the
audiences that are watching. We thought it was going to
be very much good towards this audience, but it looks
like this audience over here is much more interested in
that section. So it helps us go into the editing
(21:02):
process with unbiased thinking. And it's not that we are
a biased team. Human beings have biased it's just your
own lived experience often comes to create your own unique
interest point in the world, and I don't think that
there has been or is yet a tool that does
it in real time in the same way that we
do it now. And it also means that we're playing
that tool into other areas of our business, like it's
(21:23):
used by the Brand Plantnerships team to a B test creative,
so we can now use it for Brand plantners who
are doing creative executions with Stephen and other hosts that
they can test which creative is going to be the
most effective, so it actually gets confidence in ROI and decisioning,
and if the client and the brand plancial have slightly
different opinions on what they think will be effective, this
tool is a great way of sort of being a
(21:44):
mediator between the two and demonstrating the value that it
would have. So it's been an extraordinary change in our company.
Speaker 1 (21:51):
The first time I heard about that tool, my mind
was absolutely blowing. Now I know that we are nearly
at time and I need to let you get on
with your David Georgie, I have to ask, like us,
someone that is so passionate about podcasts, that is just
so clear from all the research that I did before this,
I would love to know what feedback do you have
for me now that we are at the end of
(22:12):
the interview.
Speaker 3 (22:14):
Gosh, I would say, how much attention are you paying
to your data? It's a really important component. Of course,
as host, you have made a warm, welcoming environment for me.
I can tell your extremely well research which makes me
feel seen and understood, and that's an amazing experience for
only human being to think that someone would take the
(22:36):
time to understand them and their motivation. So I want
to say thank you for doing the work and doing
the research pre this. This is extremely evident. You ask
naturally curious questions, you let me answer them in good time.
You ask me to qualify some things and go a
little bit deeper. So again, great hosting. I think that's
absolutely brilliant. The thing I will ask you is how
when do you look at your data and how often?
Speaker 2 (23:00):
That's a great question.
Speaker 1 (23:01):
I look at my data generally about once a week,
and then I look at it in depth about once
a month with Ihat or Aaron that look after the podcast.
But I could look at it more. I could absolutely
look at it more. I'm so inspired by hearing what
goes on behind the scenes of Diary of a CEO,
and obviously you know Flight Story is more than just
(23:22):
that show.
Speaker 2 (23:23):
And I am a data.
Speaker 1 (23:25):
Nerd at heart as well, so I love getting stuck
into the data.
Speaker 2 (23:28):
So I think that's great feedback.
Speaker 1 (23:31):
Anything else I can do, Georgie from your one hour
experience with.
Speaker 3 (23:35):
Me, lean into community. It's one of for me and
for the team. Community is what created podcast media for me.
With one of the profound and impactful reasons why I'd
love to pocast so much, the fact it gave hours
of nuanced long form storytelling in a world that was
(23:55):
trying to constrain and reduce the time, an effort that journalists, creators,
and storytellers were being told at that time that they
should be putting into their content like no one had
any attention anymore, that you should be in one hundred
and forty characters or less. I don't believe it, and
I still don't believe it that people's attention doesn't exist.
(24:15):
I think it's rubbish. I think that it does, and
I think podcasts proved it time and time again. What
you must do, more than anything is value that attention.
But also if someone is willing to give you attention,
that starts to tell me that you have a community
and a community that needs to be nurtured, a community
that needs to be talked to, a community that maybe
(24:37):
needs to talk to each other. And how are you
nurturing that community outside of the walls of this show.
How are you showing up on social media? How are
you interacting with them? When are you talking to them?
What are you saying? What are they asking you for?
That's an amazing advice from John Hugh, who is the
founder and CEO of a company called Stand that Stephen
is the co owner of, and that really is around
(24:58):
democratizing creator entrepreneurial space where creators can create tools, products,
subscriptions for their community. And he said to me, what
are the three questions you get asked the most as
a creator? What is your community asking you? What are
they asking you to answer for them? And what they
have questions about? And within that is your three product ideas,
(25:20):
your three things that you should be building for your community.
So if you're asking them what do you need? What
do you want more of? And you're getting the same
repeated questions back, it's evidence that either you need to
put more into the IP and the show itself, or
you start to well build in your social media in
real life, how do you bring these people together? And
(25:42):
for me, attention is community. If audience is the mote,
we talk about that all time, audience being the mote
attention and therefore community and probably the sort of the
twin flame of that is loyalty is the castle and
the walls, and you need to take care of those
as much as you just take care of the audience
in the reach. So I would say the community is
(26:02):
the thing that you need to be spending the most
time and how you build for them amazing.
Speaker 1 (26:07):
Georgie, thank you for spending your time with me. It's
just been an absolute pleasure.
Speaker 3 (26:12):
Thank you, thank you.
Speaker 1 (26:15):
I love that Georgie reminded us that attention is so
much more than numbers. It's community, and community, when it's nurtured,
becomes your strongest asset.
Speaker 2 (26:26):
And if there's one thing to.
Speaker 1 (26:27):
Take away from today, it's this, stop guessing, whether it's guests,
content or strategy.
Speaker 2 (26:34):
Use the tools and the data to make confident, high
impact decisions.
Speaker 1 (26:39):
Now, if you've enjoyed this episode and you want to
upscale yourself even more in AI, you will probably like
some of the episodes that I've released on how to
turbocharge your AI skills. A great place to start is
the conversation I had with Neo Applin on how to
create an AI briefing document.
Speaker 2 (26:57):
And if you have no.
Speaker 1 (26:58):
Idea what I'm talking about, you should definitely listen to
that episode because it will completely change how you interact
with AI.
Speaker 2 (27:06):
And if you know someone worried about AI or.
Speaker 1 (27:09):
Struggling to grow their audience, share this episode with them.
It might just change their perspective and make sure you're
following How I Work so that you never miss an
insight that could change how you work.
Speaker 2 (27:21):
And how you lead.
Speaker 1 (27:24):
If you like today's show, make sure you hit follow
on your podcast app to be alerted when new episodes drop.
Speaker 2 (27:31):
How I Work was recorded
Speaker 1 (27:32):
On the traditional land of the Warrengery people, part of
the Coulan nation.
Imagine if you could predict with data which podcast guest
would blow up your audience before you even invited them.
Georgie Holt's team at Flight Story built that tool, and
it's changing the game for Diary of a CEO. In
this second part of my conversation with Georgie, she takes
(00:22):
us inside the twenty thousand dollars AI Agent challenge that
transformed how her company works, creating forty eight tools, saving
sixty two thousand hours and sparking a culture of cross
team breakthroughs. She reveals the Unicorn Hunter, the AI driven
guest radar that tracks cultural trends in real time to
(00:45):
identify the voice as everyone will be talking about next,
and she shares the behind the scenes system they use
to test sometimes six hour long interviews with thousands of
people before they're released on Diary of a CEO, which
kills the guest work and make sure that every episode lands.
Speaker 2 (01:06):
And if you missed.
Speaker 1 (01:07):
Part one of this interview, I'd recommend going back to
that first. It'll give this conversation even more depth and context.
Welcome to How I Work, a show about habits, rituals,
(01:27):
and strategies.
Speaker 2 (01:28):
For optimizing your date. I'm your host, doctor Amantha Imber.
I want to talk.
Speaker 1 (01:37):
About the people now and the challenge that you're co founder.
One of your co founders, Steven Bartlett set and I
think this happened maybe two or three months ago now,
which was an AI agent competition.
Speaker 2 (01:49):
Across the entire organization.
Speaker 1 (01:52):
I heard there was a twenty thousand dollars prize for
the winning team, and you basically asked every team and
I've got the words.
Speaker 2 (02:00):
Can you tell me this is correct?
Speaker 1 (02:01):
What could you automate redesign or reimagine if AI was
your co partner not your competitor. Tell me how that
competition played out? Like what, like what happened?
Speaker 3 (02:14):
Okay, So Stephen is a genius. He had this idea.
I think it was classic US. It was sort of
a two am idea. And you know his sort of
first principle is when you express cognitive dissonance around anything,
when you sort of feel attention mentally around particularly a
new technology tool future state, it's a signal to not
lean out, is it's to lean in. And we've discussed
(02:38):
at length one of the biggest mistakes that are board
or a leadership team will be making is to not
talk about AI just sort of to think of it
as something that couldn't possibly compete, It couldn't possibly be
something that could be better than them for us, and
not to be talking about that as a company and
not to be looking at its potential and the opportunity
it creates us would be a huge mistake. And we
(02:59):
all completely agreed immediately, and he said, I want to
set a challenge for the company, and I want to
run a competition. It's a great way to bring teams together,
and our teams love to compete. We're extremely competitive people
in the best possible way. When we socialize and spend
time together, it's usually in some type of competitive environment.
And I think we went through a range of emotions
at the beginning. I think there was excitement. I think
(03:20):
that there was nervousness, a little bit of fear. I
think that's absolutely fair to say, because you might be
building something that quite frankly, could replace you. But we
all came to very quickly to the understanding is that
the best skill set in the world is actually someone
who knows how to direct the AI tools or the
tool itself. And you had to know what great looks
(03:40):
like in order to be able to qualify with whether
you're building something that is good. So I think we
went through a gamut of emotions and we got into
our teams, and then it got competitive, and then we
would also set hang on well the research, you know,
see the research team at nine pm at night working
on them at all, and then the social team will
(04:01):
be like, or we'll be in at seven am tomorrow
working on ours. And it was extremely energizing, a great
culture experience as well, because it was time bound. It
was clear we all knew within a few days that
we were building things that could radically change and improve
how we worked. And I think people could immediately see
(04:22):
the real world impact. And it didn't matter whether you'd
been working for twenty five years or two years. Everyone's
contributions had impact and meaning then everybody was able to
share and build collectively. And we're sort of working cross
functioning as well, which is really interesting, working with people
we hadn't worked with before. And people started to build
(04:44):
and they started to make breakthroughs. So the experience of
breaking through something I kind of took away. I was
like that it's actually experienced that we don't sort of
put in teams enough because it's rewarding and you actually
feel like you're making progress and sort of the reward
and progress making are so important in human psychology to
kind of feel that you're working with purpose and you're
(05:06):
pursuing something. You're pursuing something important and it suddenly became
a pursuit. And we hosted an amazing day on the
sort of the end of the competition, and we got
every team to present their ideas. We had independent judges
come in, We have the Chief Technology Officer WPP come
in and judge the work. We had our director of
(05:27):
Innovation and our lead data scientist and Stephen reviewing all
of the entries and they were judged on multiple scores,
not just obviously efficiency impact, but social impact, human impact.
It's a usability. Is this something that we would be
proud to say that we use that We wouldn't sort
of be hiding it that we use an efficiency tool,
(05:50):
but we're actually saying it's the human beings working on
something that we can see it would make an impact
on our purpose and long term goal. It was such
a powerful reminder that this industry of media still attracts
the most creative, the most curious, the most driven, and
the most passionate people in the world. Is just the
game has slightly changed and we need to play it
(06:12):
and play it better than anybody else. And it was
a wonderful experience, probably one of the greatest I've had
in a very long time, watching everyone present the tools.
And I think we made forty eight tools across all
the different departments. We say sixty two thousand hours, which
is that seven years of efficiencies found at about a
million dollars of cost impact which we can now reinvest
(06:33):
in other things.
Speaker 1 (06:33):
I love what you said about just that experience of
teams making a breakthrough and making progress. It reminds me
of research from Professor Theresa and Marbala that shows that
the thing that leads to the biggest impact on engagement
and motivation at work is having that sense of progress.
And I love how you described you know, that process
(06:56):
that teams went through.
Speaker 2 (06:58):
I have to ask what was the winning idea or was.
Speaker 3 (07:00):
Research and guestbooking, So they had four ideas in total.
We have evidence to show that the impact of a
great guest can be transformational on so many different critical
factors in the company, and it can be as high
impact as the audience. Having a transformational experience or making
(07:22):
a huge life choice or changing the way they eat, live, think,
have opinions or ideas and particular topics. Start a business,
so we can have a profound impact on an audience,
but it can have a profound impact on the metrics
of what we measure at flight story in terms of
how it is growing. So that can be show growth,
it can be revenue growth, it can be subscriber growth,
(07:44):
it can be views ABD so we can see that
the guest is one of the greatest transformational decisions that
we can have, but it creates extremely high human effort,
so it is a lot of researching, sourcing, emailing, scheduling,
(08:04):
and then of course once that's put through researching, that
has to go in. So it's one of our most
high human effort but high payoff areas of the business.
So they've got four different tools that could improve the
efficiency of the process. So they've built a tool called
Unicorn Hunter, which is going into a tool that we've built,
which is guest Radar. So we've built we'd already previously
(08:25):
built a tool called guest Radar, which essentially now can
identify a moment at which a channel whether it's YouTube
or on social media, has somebody appear on that channel
and it overperforms versus it's sort of i'd say current state,
and it will point towards who that person is and
what they were talking about. So that tool has existed
(08:47):
and it is extremely effective and I think completely unique
and proprietary in the industry, and we can now look
at a guest and say, do they overindex on performance
when they appear on a show where they kind of
standard We had a great example Withdr Eric Weinstein who
went on The diver CEO, who had eleven x the
performance on any channel that he went on, So he
(09:07):
was performing eleven times and the standard performance of a channel.
So we booked him on the show and again had
exactly the same impact on the divers O. So we
kind of had this phrase of killing the guestwork. What
was happening is sort of finding those unicorns still required
human effort because you're kind of looking for cultural trends,
high impact moments in news and culture and society. So
(09:29):
they built a layer onto that tour that was now
monitoring culture and social trends to understand what topics were trending,
what was not just uniquely interesting to the world in
that moment in time, but who was uniquely interesting to
the world at that moment in time, and then being
able to action that decision immediately. So Steven's testing out
new debate formats on the back of this idea and
utilizing this idea to find voices you can immediately speak
(09:52):
on complex topics that human beings are curious about and
trying to solve and have opinions on. So we've had
debates world War three on feminism, on lost boys, you
know how young men are being completely a finding sort
of societal struggle incredibly difficult, and the debate format needs
(10:13):
people who can speak with immediacy and expertise on a topic.
And now that tool, the Unicorn Hunter, is built into
guest Radar to help them find it by monitoring culture
and social trends on the Internet and voices who are
speaking on that topic. So it allads an immediacy to
the guestbooking team that they can now sort of say, hey,
we've seen that you are extremely well first on this topic.
We can see you're talking about it right now. Can
(10:35):
you be in LA in two days time to talk
about it with Stephen? So the speed is extremely fast.
Now they built out research tools and briefs that can
deal with extremely complex research topics, which then have a
human lens going over the top. They were the ones
that really stuck out to us with the research and
the Unicorn Hunter.
Speaker 1 (10:55):
So far, you've heard how flight Story turned an AI
challenge into million in impact. In the second half, Georgie
reveals the tool that lets them test a four to
six hour interview with thousands of people before it's released
and how that data changes everything. Plus she gives me
(11:16):
feedback live on air about what I should be doing
with this very podcast. If you're looking for more tips
to improve the way you work can live. I write
a short weekly newsletter that contains tactics I've discovered that
have helped me personally. You can sign up for that
at Amantha dot com. That's Amantha dot Com. I have
(11:44):
to ask, because I know that before you co founded
flight Story with Stephen, I think.
Speaker 2 (11:50):
You and your business partner Christiana.
Speaker 1 (11:52):
Yeah, Christiani listened to every episode of Diary of a CEO.
Speaker 2 (11:56):
Am I right in saying that.
Speaker 3 (11:58):
I will be completely honest, Christiana did. I would find
very high interest in the business and the entrepreneurial side
of his guest, and of course I could see an
episode was sort of important in the culture zeitgeist. I
would listen, but I would say I was listening at
least one episode a week. I would definitely be that
Christiana was listening to everyone.
Speaker 1 (12:19):
I've got to ask, then, like, what would be your
top three if you had to.
Speaker 3 (12:22):
Choose top three episodes? Dr rris wat I think she
is a generational voice, and I think she radically changed
how people thought about stress and the impact of stress
on their life and the things that they could do
to move them out of stressed states and into a
more balanced way of living. I absolutely loved Dr Christen
(12:44):
Holmes's episode on circadian rhythm and her focus and relentless
commitment to human beings being able to find the democratic
and easily accessible ways to improve their health and well being,
and her ability to try late again complex science and
understand deeply and intuitively how to translate that to audiences
(13:07):
to help them understand how to better live, to just,
for example, the moment you wake up, we should get
natural light into your eyes, and how important it is
to take and make those decisions every single day, and
the compounding impact that can have over time. We absolutely
loved her episode. I loved Divina mccaul's episode as well,
which was again her story through addiction. I don't think
(13:30):
we'd ever heard her talk like that before. It had
a profound impact on me. I was lucky enough to
be in the room with Missus Michelle Obama's podcast episode
as well, just to be in her presence and to
experience I think her greatness honestly, and to watch a
woman who has had the eyes of the world on
her twenty four to seven ever since Barak decided to
(13:50):
run for office, to still come back to humanity and
love and childhood experience and family, to kind of hold
those principle so close to her despite everything was one
of the greatest experiences of my life to kind of
watch that happen. I think a lot of people talk
about no Gowdat's episodes the AI Emergency, which was you know,
(14:11):
Stephen was one of the first people to sort of
bring that into the public consciousness. This is something we
should be paying attention to. And if we don't pay
attention to it, if we don't become good stuarts and
parents to AI, what will happen how will it now
move forward? And what that was well over two to
three years ago. I think so he was bringing it
into the attention of the world. I think sort of
(14:31):
Moo's first episode on Happiness. I think people mentioned a
lot because it again his personal story and how we
shared it with Stephen, and how that marriage knitted up
to his purpose. Again, I think had a profound impact
on people. And I think my dream guest, if he
ever had it on all the way, I don't know
whether he would ever happen. I would be Dolly Potton, I.
Speaker 1 (14:52):
Know that about it, But I think we'd have to
find his interest point in that.
Speaker 3 (15:01):
When we look at Yes to Burkett, it comes back
to it a very simple framework, which is, is the
audience interested in this topic or person? Can we prove
that through evidence and data? Is this person the best
person in the world to talk about their topic or
their life experience? And do we have evidence and data
to prove that? And as honestly, the host, are you,
(15:23):
Stephen or any of our new show hosts interested enough
in both that topic and that person to sit down
with them for six hours maybe even longer. He dedicates
you know, the highest portion of his time to recording
these episodes. It's not a thank you for coming in
for ninety minutes and we'll shake hands and appreciate the thoughts.
It's sometimes sort of six, seven, eight hours of conversation.
(15:45):
I do know one that went on for eight hours
in total, and then it's edited down into two to
two and a half hours of conversation. So there has
to be extreme interest from both those people in that seat.
I go back to the Mister Beast episode of being
One of the greatest examples of that is that I
think the story has been told already, but he canceled
the morning of the record because his flights have been canceled,
and he'd been a guest that the team has been
(16:07):
working on for years. You know, when these people arrive
in the seats of these shows, it's not one email exchange.
I think there's sometimes two to three years of work
that have gone in. I think with a doctor Andrew Huberman,
I think it was two years before we eventually were
able to schedule it and move towards having him in
the seat. So it's not sort of oh high, where
(16:27):
the divers, Yo, do you want to come on the show.
This team has compounded the impact over years. You know,
Stephen is a success he is now, and that show
is a success now because of the compounding dedication that
team has had, and many years ago they weren't the
biggest show in the world. They were on their way
to be it, but it wasn't that. It wasn't at
that point. So these episodes take years to come to life.
(16:48):
And mister BEA's canceled in the morning is playing a
big canceled and he was recording the finale of Beast
Games that afternoon. So for all intents and purposes, this
was an impossible task and the team just didn't believe it.
They just we don't believe that this is impossible, and
I believe that we can make this happen. And they
made the impossible possible, and they ended up recording with
mister Beast. I think at two am in the morning,
(17:08):
after you've recorded the finale to Beast Games, and I
think they went on till like five in the morning,
recorded all night. Jimmy is one of the most extraordinary humans.
I think he then took Stephen to seven eleven to
show him his chocolate because he was again as he's
obsessed with the one percents and that's the level of
commitment it takes. So I think those guests kind of
stand out for all different unique reasons.
Speaker 2 (17:30):
Wow, that is quite the story.
Speaker 1 (17:33):
I've also heard that there's a process called I think
it's called the pre watch where yes, you take that
six to eight hours of sound and you get that
into two to two and a half hours. Can you
tell me what that involves exactly?
Speaker 3 (17:46):
Yes, It again goes back to one of our first principles,
which has killed the guest work. You don't need to
guess anymore, and we have so many tools and opportunities
to not have to do that. And one of the
even though YouTube and YouTube studio is one of the
greatest data sets you can get as a creator, and
it's incredibly important in terms of understanding audience behaviors and
(18:09):
understanding when people are watching, how long they're watching for,
when they stop watching, when they start watching again, when
they share things, when they comment, and buying things interesting.
It was still after you had done all the work,
So you'd have done all the work on the episodes
you've recorded with mister Beast at two in the morning,
you'd have moved maybe impossible, possible You've just spent two
(18:31):
years getting a guest into the diver Ceo chair or
any of our new show chairs, and you still don't know,
and you're still you've done everything, and then you publish
on YouTube and you hope that everything you've done is right.
Stephen was very frustrated by that, and also he obviously
presents on TV shows and there's no methodology and science
and TV there's you know, quite often you record months
(18:53):
and months previously, they then do all the editing, and
then six months later it goes onto the TV and
it knows them. It's like we did the ratings, and
how can that be? That is a six month gap
between what we're recording and what happened now. So he said, well,
what if we could build something that could give us
data and insight about what people enjoyed, what they found interesting,
Because wouldn't it be a shame if we put our
(19:16):
own human bias on what was interesting and what was
not interesting because of our own lived experience, Because you,
as a forty five year old woman and me, as
a thirty one year old man, had uniquely different interests
in that episode, and I edited it through my lens
and I completely ignored an entire section of that content
that would have been extremely interesting to an audience that
(19:37):
maybe I didn't understand or I didn't have the lens
of experience or empathy around, and we can't edit that way.
So he built a talk called pre Watch, which means
that we have anywhere between sort of five to seven
thousand people watching episodes of the Diarver CEO, particularly where
across a cohort of people, we maybe don't have a
(19:57):
unique experience of that person's expertise or area or field
of work or human interest story. And it's eye tracking software,
so it can look and understand. We can see when
people have looked away, so it can see where attention
may be dropping. And then people can interact with the
tool as well. They can comment, they can say that
(20:18):
something was interesting. You get this incredible interest cup, and
you'll see that there's high interest at the beginning, which
you would expect to see on YouTube do anyway. You know,
there is natural drop off over time, but you might
see something that happens in the middle of the episode
and at the end of the episode that has the
highest interest of all and you're like, well, we definitely
want to make sure that we talk about that in
(20:39):
the trailer, we definitely want to make sure that's in
the packaging of the episode. Perhaps a section that we
thought was very interesting hasn't had the payoff of the
audience that we wanted it to have, maybe in a
particular demographic, because we can understand the demographics of the
audiences that are watching. We thought it was going to
be very much good towards this audience, but it looks
like this audience over here is much more interested in
that section. So it helps us go into the editing
(21:02):
process with unbiased thinking. And it's not that we are
a biased team. Human beings have biased it's just your
own lived experience often comes to create your own unique
interest point in the world, and I don't think that
there has been or is yet a tool that does
it in real time in the same way that we
do it now. And it also means that we're playing
that tool into other areas of our business, like it's
(21:23):
used by the Brand Plantnerships team to a B test creative,
so we can now use it for Brand plantners who
are doing creative executions with Stephen and other hosts that
they can test which creative is going to be the
most effective, so it actually gets confidence in ROI and decisioning,
and if the client and the brand plancial have slightly
different opinions on what they think will be effective, this
tool is a great way of sort of being a
(21:44):
mediator between the two and demonstrating the value that it
would have. So it's been an extraordinary change in our company.
Speaker 1 (21:51):
The first time I heard about that tool, my mind
was absolutely blowing. Now I know that we are nearly
at time and I need to let you get on
with your David Georgie, I have to ask, like us,
someone that is so passionate about podcasts, that is just
so clear from all the research that I did before this,
I would love to know what feedback do you have
for me now that we are at the end of
(22:12):
the interview.
Speaker 3 (22:14):
Gosh, I would say, how much attention are you paying
to your data? It's a really important component. Of course,
as host, you have made a warm, welcoming environment for me.
I can tell your extremely well research which makes me
feel seen and understood, and that's an amazing experience for
only human being to think that someone would take the
(22:36):
time to understand them and their motivation. So I want
to say thank you for doing the work and doing
the research pre this. This is extremely evident. You ask
naturally curious questions, you let me answer them in good time.
You ask me to qualify some things and go a
little bit deeper. So again, great hosting. I think that's
absolutely brilliant. The thing I will ask you is how
when do you look at your data and how often?
Speaker 2 (23:00):
That's a great question.
Speaker 1 (23:01):
I look at my data generally about once a week,
and then I look at it in depth about once
a month with Ihat or Aaron that look after the podcast.
But I could look at it more. I could absolutely
look at it more. I'm so inspired by hearing what
goes on behind the scenes of Diary of a CEO,
and obviously you know Flight Story is more than just
(23:22):
that show.
Speaker 2 (23:23):
And I am a data.
Speaker 1 (23:25):
Nerd at heart as well, so I love getting stuck
into the data.
Speaker 2 (23:28):
So I think that's great feedback.
Speaker 1 (23:31):
Anything else I can do, Georgie from your one hour
experience with.
Speaker 3 (23:35):
Me, lean into community. It's one of for me and
for the team. Community is what created podcast media for me.
With one of the profound and impactful reasons why I'd
love to pocast so much, the fact it gave hours
of nuanced long form storytelling in a world that was
(23:55):
trying to constrain and reduce the time, an effort that journalists, creators,
and storytellers were being told at that time that they
should be putting into their content like no one had
any attention anymore, that you should be in one hundred
and forty characters or less. I don't believe it, and
I still don't believe it that people's attention doesn't exist.
(24:15):
I think it's rubbish. I think that it does, and
I think podcasts proved it time and time again. What
you must do, more than anything is value that attention.
But also if someone is willing to give you attention,
that starts to tell me that you have a community
and a community that needs to be nurtured, a community
that needs to be talked to, a community that maybe
(24:37):
needs to talk to each other. And how are you
nurturing that community outside of the walls of this show.
How are you showing up on social media? How are
you interacting with them? When are you talking to them?
What are you saying? What are they asking you for?
That's an amazing advice from John Hugh, who is the
founder and CEO of a company called Stand that Stephen
is the co owner of, and that really is around
(24:58):
democratizing creator entrepreneurial space where creators can create tools, products,
subscriptions for their community. And he said to me, what
are the three questions you get asked the most as
a creator? What is your community asking you? What are
they asking you to answer for them? And what they
have questions about? And within that is your three product ideas,
(25:20):
your three things that you should be building for your community.
So if you're asking them what do you need? What
do you want more of? And you're getting the same
repeated questions back, it's evidence that either you need to
put more into the IP and the show itself, or
you start to well build in your social media in
real life, how do you bring these people together? And
(25:42):
for me, attention is community. If audience is the mote,
we talk about that all time, audience being the mote
attention and therefore community and probably the sort of the
twin flame of that is loyalty is the castle and
the walls, and you need to take care of those
as much as you just take care of the audience
in the reach. So I would say the community is
(26:02):
the thing that you need to be spending the most
time and how you build for them amazing.
Speaker 1 (26:07):
Georgie, thank you for spending your time with me. It's
just been an absolute pleasure.
Speaker 3 (26:12):
Thank you, thank you.
Speaker 1 (26:15):
I love that Georgie reminded us that attention is so
much more than numbers. It's community, and community, when it's nurtured,
becomes your strongest asset.
Speaker 2 (26:26):
And if there's one thing to.
Speaker 1 (26:27):
Take away from today, it's this, stop guessing, whether it's guests,
content or strategy.
Speaker 2 (26:34):
Use the tools and the data to make confident, high
impact decisions.
Speaker 1 (26:39):
Now, if you've enjoyed this episode and you want to
upscale yourself even more in AI, you will probably like
some of the episodes that I've released on how to
turbocharge your AI skills. A great place to start is
the conversation I had with Neo Applin on how to
create an AI briefing document.
Speaker 2 (26:57):
And if you have no.
Speaker 1 (26:58):
Idea what I'm talking about, you should definitely listen to
that episode because it will completely change how you interact
with AI.
Speaker 2 (27:06):
And if you know someone worried about AI or.
Speaker 1 (27:09):
Struggling to grow their audience, share this episode with them.
It might just change their perspective and make sure you're
following How I Work so that you never miss an
insight that could change how you work.
Speaker 2 (27:21):
And how you lead.
Speaker 1 (27:24):
If you like today's show, make sure you hit follow
on your podcast app to be alerted when new episodes drop.
Speaker 2 (27:31):
How I Work was recorded
Speaker 1 (27:32):
On the traditional land of the Warrengery people, part of
the Coulan nation.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
New Heights with Jason & Travis Kelce
Football’s funniest family duo — Jason Kelce of the Philadelphia Eagles and Travis Kelce of the Kansas City Chiefs — team up to provide next-level access to life in the league as it unfolds. The two brothers and Super Bowl champions drop weekly insights about the weekly slate of games and share their INSIDE perspectives on trending NFL news and sports headlines. They also endlessly rag on each other as brothers do, chat the latest in pop culture and welcome some very popular and well-known friends to chat with them. Check out new episodes every Wednesday. Follow New Heights on the Wondery App, YouTube or wherever you get your podcasts. You can listen to new episodes early and ad-free, and get exclusive content on Wondery+. Join Wondery+ in the Wondery App, Apple Podcasts or Spotify. And join our new membership for a unique fan experience by going to the New Heights YouTube channel now!