Welcome to Inside Marketing with MarketSurge — your front-row seat to the boldest business insights, marketing breakthroughs, and entrepreneurial real talk.
Hosted by Reed Hansen, Chief Growth Officer at MarketSurge and a digital marketing veteran who's helped scale everything from scrappy startups to Fortune 500 giants, this podcast dives deep into what’s really moving the needle in today’s marketing world. Find us at Marketsurge.io
Each week, we’ll break down the latest marketing and business news (minus the fluff), explore tech trends you actually need to know, and feature unfiltered conversations with the most interesting minds in entrepreneurship and marketing.
Whether you're a founder, a marketer, or just a curious hustler looking to level up, this is where growth happens—loudly, smartly, and with just the right amount of sass.
Subscribe, tune in, and let’s scale something legendary. 🚀
Speaker: Welcome to Inside
Marketing With Market Surge.
Your front row seat to the
boldest ideas and smartest
strategies in the marketing game.
Your host is Reed Hansen, chief
Growth Officer at Market Surge.
Reed Hansen, MarketSurge: Hello
and welcome back to Inside
Marketing with Market Surge.
Today's guest is Ari Sanju,
founder and CEO of Omni Key, an
AI driven creative platform that
helps brands generate, analyze,
and optimize ad creative at scale.
trained computer scientist
turned entrepreneur.
mission is to bridge the gap between
generative AI and performance
marketing, turning machine creativity
into measurable business results.
Under his leadership, Omni Key has
helped brands use AI to produce
high converting ad variants.
Discover insights from data that
humans would miss and scale creative
faster, all while staying on Brandand.
And on message.
Today we'll dive into how teams
can build AI powered creative
pipelines that actually perform.
Welcome to the podcast, Ari.
Hikari Senju: Thank you Reed.
Really excited to be here
Reed Hansen, MarketSurge: Thank
Hikari Senju: and thanks
for the amazing intro.
Reed Hansen, MarketSurge: Absolutely.
You have a very impressive
resume and a great
Hikari Senju: Thank
Reed Hansen, MarketSurge:
I'm excited to dive
Hikari Senju: you.
Reed Hansen, MarketSurge: So let's start
from the top, and I'd love to talk about
the platform, Omni Key for instance.
So what inspired you to start
this and what was the key problem
you were trying to solve with?
Hikari Senju: Sure.
So I was a performance marketer
like many of your listeners.
Right before starting on Mickey,
I was running ads, I was running
meta ads, I was running Google ads.
I was working with a
designer to create content.
And really felt that, you
know, this was in 20 16, 20 17.
I really felt like it was broken.
The system was broken.
There's that, a famous Watermaker
quote, which is, you know, I
know my marketing is working.
I just dunno which half of
my marketing is working.
And so it felt true today more than
ever, or back then more than ever.
Uh, and I think probably true today
as well for many, for many marketers.
And so, at the same time.
I studied, you know, as you noted,
computer science, the focus of machine
learning at Harvard and had this
deep passion, uh, for generative
AI because my dad's a painter.
And so the idea of an AI being artistic,
being creative is, was something
I got really hooked on in college.
And This potential technology.
And so, I saw this opportunity to use this
new technology, which was generative ai,
to solve this marketing problem, which
is driving more efficiencies in your ads
budgets, by being more effective at ab
testing, scaling AB tests, and, uh, and
having more of a, you know, we describe
ourselves as a creative operating system.
A systematic approach for optimizing
creative versus a historic approach,
which is more kind of, kinda shot
in the dark, I would say almost
like a scattershot approach.
But really more of a systematic
approach to, almost like brute forcing
performance from these ad channels.
And so that was kind of the idea.
and then we've, were the first
mover in this space and been
growing and scaling ever since.
and particularly now it's a
very exciting time for us.
we're growing faster than
we've ever, done before.
Reed Hansen, MarketSurge: I think
generative AI can be a little
controversial and I think your
unique background, a son of a
painter, I think you're probably
more aware of this than most.
there's a lot of issues
with, generative ai.
You know, there's the, creatives
that, naturally bristle against it.
They, and then there's, are
they going to be displaced?
Or but then they also push back
and say many cases, the generative
AI just isn't a very high quality.
What are some of the issues
that you push up against, what
are some of the most common
Hikari Senju: Yeah.
Reed Hansen, MarketSurge: get?
Hikari Senju: Sure, for sure.
So I've heard every single, we've
been operating this company for
many years, and so I've definitely
heard every single feedback possible
regarding generative creative I
think it's really changed this year.
2025 was the year where
I would say AI became.
Maybe as good as the
average designer, maybe.
the quality of the ads that these
models are generating, the quality
of the designs, the graphic designs
that these models are generating is
almost indecipherable from a human
creating this content themselves.
And I think, you know, particularly
if you're a small business owner and
you maybe can't hire a designer or
you can't hire many designers or even
if you're a performance marketer and
you need to create that volume of
creative needed nowadays for driving
performance in your ad campaigns.
You know, meta recently updated
their algorithm to and drama data,
and now it's all about creative
testing and creative journeys.
Most marketing teams just aren't staffed
sufficiently to produce that volume
and diversity of creative content.
AI really is a lifesaver for them
because how else are you gonna produce
the volume of creative you need to
deliver performance in your marketing?
And so I think it really does
solve a really critical problem for
marketers, in terms of replacing,
amazing creative designers.
I don't think, creative designers
are gonna go away in a broader sense.
I think there's always gonna be need
for good ideas and big ideas and,
and I think the best stories and the
best creatives are gonna continue
to, uh, you know, do amazing job,
and continue to, to do very well.
But yeah, I mean, I think the average,
design that like somebody who isn't
even a designer is using to create
a underperforming creative like that
can and should be automated by an ai.
And, and furthermore, then you need to
turn one creative into dozens of different
variations for different dimensions and,
you know, horizontal, vertical, et cetera.
Image video, you need to format
it for Reddit, you need to format
it for TikTok, you need to format
it for Instagram and YouTube.
I mean all this kind of grunt work.
That probably isn't the best use of
a human creator anyway, and a human
creative person, should be automated.
And so I think it's gonna let people
focus on the areas that they're
strong in, which is inspiration.
And, the grunt work is gonna
be automated more and more.
Reed Hansen, MarketSurge: That,
that's a great explanation now and
I know that you're not necessarily
trying to compete against the great.
Master artists in the world, of course.
And, and so there's no threat of that.
But and we get down to, the
kinda the utilitarian focus
of, generating ad assets.
Now I'm sure, so you've been a
performance marketer in the past.
You so you have a lot of expertise.
Does your platform.
Incorporate best practices from the
universe of ads marketing and how,
so where do you draw on for the
quote unquote creative that converts?
Or is it a little bit more
comparative against past, like
within an ad account, past assets.
Hikari Senju: Yeah, so it's both.
we have different levels of rules.
So there's the rules relating to what's
good creative design, and we have
thousands of, reference ads that we,
recommend to the user and to the ai
depending on the industry, depending
who their target customer is, depending
on which country this brand is in, and
the target audience of their buyers.
And so we have a whole
recommendation system in terms of.
You know, when you're prompting
an AI to generate creative, what
is the best example that's most
likely to drive conversions?
And so.
We have data regarding, historic
data regarding for every
industry and for every country.
These are the types of
ads, uh, that convert.
Separately there is the customer
connects their own data.
They connect their meta, their
Google, their Reddit, their
TikTok, their LinkedIn, et cetera.
Uh, we tag all those creatives.
We build, predictive models based on
the data that we're collecting from
that customer's ad account plus, The
tags to say, okay, well what is it?
That's, what kinda designs
are driving more conversions?
What types of designs are
getting more impressions as well?
Uh, and then generating new
creative based on those suggestions.
And so it's both.
And then also third, I would
actually, there's also like industry
related regulations and roles.
Uh, if you're in healthcare or if you're
in FinTech you know, and, and, and, um.
I think brand safety and brand compliance
is another big key focus of ours.
We have, we announced something called
the brand LLM many years ago, and it's
all about maintaining brand safety,
uh, and that particularly regarding
different regulated industries or,
you can and can't say certain things.
And so all that is incorporated
in terms of our generative model.
Reed Hansen, MarketSurge:
That, that's great.
And I think that's really important.
you know, I think you've written
about an AI creative loop, and I think
that's, illustrative to that, where you.
You.
And maybe you could talk a little bit
about like, how you incorporate that in
Hikari Senju: Yeah sure.
Reed Hansen, MarketSurge:
how does that work?
I think that's a fascinating concept.
Hikari Senju: Yeah, absolutely.
The old era of creative was, you know, you
have an app and you design an app and it's
a kind of in this vacuum of information,
you know, it's just an inspiration.
You're creating an image, you're
creating a video, and you launch it.
But I think, uh, what AI and gene
AI enables is actually more of a
systematic approach to creating content.
So it's no longer about just, you're in a,
you know, vacuum chain, but, but actually
constant feedback from data regarding
these are the designs that are resonating.
And then using that data in real time
to generate new creative, um, and have
this kinda end-to-end workflow from
brand to insights to then generation
and launching and deploying those
creative, and kind of this kind of.
Connected line.
Whereas historically it
wasn't very connected at all.
These are almost operating
silos really when it comes to
marketing, analytics and creative.
And so by.
Closing the loop between the data and
the creative process, you can drive
more effective ad campaigns, uh,
that drive higher return on ad spend.
So how that works within our platform
is customer signs up, they connect
ad accounts, they immediately
get insights about the types of
creative that's driving results.
They generate new creative on our platform
based on those insights, and then they
can launch those creative in our platform
Integrations with Meta, Google,
LinkedIn, TikTok, Reddit, et cetera.
And so, and then they can
see those insights again.
And so it's this closed feedback loop of
insights, generations, launching brand
information then, and then and it's all
anchored based on the customer's brand.
Reed Hansen, MarketSurge: Okay.
Now you work with so much,
you've ingested a lot of data.
You have a lot of clients been
doing this for many years.
And we talk about the data aspect of this.
What are some of the more
surprising insights you've found?
As you've, had your engine analyzed,
all these creative assets have anything
that sticks out to you even, recently
or putting you on the spot, but,
Hikari Senju: josh, I mean it's,
through the years there's been,
numerous types of these insights.
I think it is so contextual to
the brand and the audience and the
country that you're operating in.
So for example you know, we have a
large customer base in the United
States and it's English, but we have
a large customer base also in Japan
and in Asia and completely different,
creative styles, text density,
and we do those very well as well.
and it's based again, on just letting
the AI automate based on the data
and suggesting its own reference ads.
Yeah, I mean, it is really more of it's
so contextual depending on the customer.
That I'm not sure if there's a specific
takeaway, but many years ago, for
example, it was like, the learn more
CTA was one of the best performing
CTAs out of all any of the CTAs or
but yeah, also these algorithms are
constantly changing as well yeah.
Reed Hansen, MarketSurge: Do.
So let's maybe zoom out a little bit.
I'd love to, considering your position
in the marketplace AI is just like
such a compelling topic overall,
but generative AI is a nice little.
Subsection of it.
What I guess my experience with generative
ai, it started with Jasper, if you, I
don't know if you are familiar with that
Hikari Senju: Yeah.
Reed Hansen, MarketSurge: Generating
blogs three, four years ago
just was, barely passable at,
or barely legible at the first.
And has greatly improved of course.
And now there are so many other
people, not people systems and
platforms that are doing that.
I have less experience in
creating visual assets.
But tell us a little bit about
what are some of the interesting
generative AI trends that marketers
or business owners should be?
Looking at, paying attention
to that are going to be more
important coming in the future.
Hikari Senju: I really, I
remember Jasper very well.
we had also a copy generation tool
powered by GPT-3 and even before
GPT-3, we were also an open AI beta
partner for GPT-3 before it launched.
definitely.
It was very impressive growth at the time.
I think at some point Jasper was
one of open AI's largest customers.
But since then, things have changed.
Prior to Jasper we were fine tuning
models, and taking those, weights
and beddings from hugging face even a
precursor we're using customer's brand
and information, fine tuning and writing
prompts based on each brand to generate
on-brand copy, back in GPD two, prior
to uh, GP I mean that was definitely,
an interesting inflection point, right?
when it comes to trends we're seeing
today, suggestion to marketers
first and foremost is just create
and test more variety of content.
our customers see on average,
significantly higher return ad spend
using our platform, based on the data.
But just testing more creative also
will drive higher, performance as well.
the new update in particular, all
about the sequence of creative and
lifecycle of that user journey.
And it's about showcasing a
different, you know, in many ways
the algorithm is ab testing different
sequences of creative to show a
potential buyer to have them convert.
Um, and so.
One is you know, there's
a couple of metrics here.
One is you know, the average conversion
takes about seven touchpoint.
So you really want each one of
those customer touchpoints to be
unique and kind of move the user
forward in that buyer journey.
Two, the metric is, uh, if a
user sees the same creative more
than four times conversion rates
didn't decrease by about 50%.
And so, in those seven impressions, you
know that you're giving, uh, that user if
you're showing the same ad more than four.
times you're lowering
your return on ad spend.
Um, and so just having, I dunno, at
least a dozen ads in your ad set, uh,
to start with so that a customer can
have each, you know, unique touchpoint
for every single one of those seven
impressions, uh, is gonna dramatically
improve your return on ad spend.
But then, yeah, after that then
it's about, well, how do I create
new creative test and how do
I have a systematic approach?
I'm not just throwing
random stuff up there, but.
You know, incrementally
improving the creative content.
And that's really where the data loop
comes in, as well as transferring insights
from other ad accounts because I think,
and you know, I run our performance
marketing today at Omni Key and the
performance of, you know, meta Google,
Reddit, TikTok Link, et cetera, changes.
So much week by week, month by month.
And so I'm constantly shifting
budgets across platforms as well,
depending on where we're seeing
performance and, and so being able
to transfer those insights across
platforms is also so important as well.
Therefore that omnichannel approach for
marketers I think is the other key thing.
So one is volume of creative, and
then two is variety and diversity
of channels you're advertising
on is more important than ever.
As as there's more options than ever
for advertisers to run media on, but
also as the advertising ecosystem
gets more volatile and competitive.
Reed Hansen, MarketSurge: So got a thought
in my head, but I'm trying to think of
a good way to ask it because one of the
issues with ai generative content is.
Sometimes people can tell that
it's AI generated and that's one
factor, can they tell or not?
But then the other factor
is do they care or not?
And you find that those two factors,
like a cultural tolerance for.
Either being able to spot AI
generated content or, being
concerned with it or embracing it.
Do tell us a little bit about
how those factors affect,
how you generate ad creative.
Hikari Senju: First is I think,
you can go to our homepage.
You can see the ads are being
generated by our customers.
You can sign up and start Generat ads.
The ads that we're generating you,
it's very hard to differentiate
that it was AI generated.
They look like it was a real photo shoot.
A real a real.
Not AI generated.
so I think that line is
almost indistinguishable now
between AI and human created.
Like when it comes to graphic design.
It's almost indistinguishable not a
hundred percent of the time, but if you
create enough, definitely substantial
percentage of the ads that you're
generating on our tool is, in industry
control from human created content.
So that's that.
And then the second thing is, I think
related to, you know, advertising a lot
of it is about trust and authenticity.
that's what your brand building is.
Our brand is essentially
a symbol of trust,
If your brand is about innovation
and ai, then that could work.
If you're trying to, make your brand
seem cool and edgy even if it's obviously
AI generated, that still adds extra
kudos to your brand because it seems
like you're trying new things and I
think consumers appreciate when they.
You know, when their brands innovate,
if that's kind of, the domain.
So like if Coca-Cola is testing out
new AI things or if then, the NB
is, I think that's cool, but if it
might, you know, my, um, I dunno,
the company that does like my surgery
is doing AI generated content.
Maybe, you know, a little bit more,
it depends right on the context.
And so, one thing we're seeing
also in this is again, because.
to Japan is, uh, this, which
is, you know, again, somewhat
of a smaller, uh, percentage of
our revenue comes from there.
But, you know, they're not
even generating, like it's all
cartoon and animated stuff.
So in that realm it is indistinguishable.
I mean, 'cause these are all cartoon
characters that are being generated
and they're running those ads.
And when it comes to that kind of stuff,
it is a hundred percent distinguishable.
And so, yeah.
I think it also depends on like the
cultural context and rules and regulations
of that country where we're seeing
maybe a little bit more of an innovative
approach to using AI generated ads by
large enterprises in the wild and maybe
more of a, depending on the industry, a
bit more of a conservative approach when
it comes to embracing this technology.
Reed Hansen, MarketSurge: Interesting.
I love talking to people that work in
this space, like you and I think this has
been a really interesting conversation.
I would love to point people to Omni Key.
Where can people find it?
where are the best places to follow you
for thought leadership and, what where
are the best platforms to, to find you?
Hikari Senju: Yeah you can search
for us on Google, O-M-N-E-K-Y
Omni Key, or omni key.com.
You can also probably just spell Omni Key.
It'll probably, send
you to the right place.
You can follow us on LinkedIn.
You can follow me on LinkedIn where I
post, some thought leadership content.
Our Instagram, our Facebook, our
Twitter X you can actually join
our, Reddit sub channel as well.
you can follow us on YouTube and then you
can sign up for free trial on our account.
So oniki.com.
Seven day free trial, you can
see the tool for yourself.
I think that our product is a
lot better than our competitors.
I think we generate higher
quality ads than our competitors.
I think we provide better insights
than any of our competitors.
I think we have a better workflow product
in the ease of use of generating, editing
and launching better than our competitors.
I would just try software free for
seven days and, make the call yourself.
Reed Hansen, MarketSurge: That,
it looks like a great tool.
I've signed up for the trial.
I'm very excited about it.
And please reach out.
This was so great to
have you on the podcast.
Thanks for joining Ari.
Speaker 2: Want to stay ahead of what's
actually working in marketing right now.
Head over to Market surge.io
and see how we're helping businesses
grow smarter, faster, and louder.
That's market surge.io
because your next breakthrough
shouldn't be a guess.