Latest / The Prompt / Ep. 586 | Stop bolting on AI: Rebuild your go-to-market from the foundation up

Transcript

Ep. 586 | Stop bolting on AI: Rebuild your go-to-market from the foundation up · The Prompt

0:01
This is on base your one stop shop for solving your most
0:04
pressing B2B go to market challenges.
0:07
Each week, host Chris Moody and Paul Gibson talk to sales and
0:09
marketing leaders to get you in depth insights and creative
0:12
strategies to set your revenue teams up for success.
0:18
Hey everybody, welcome back to the own Base podcast.
0:20
I'm your host, Chris Moody and today we're diving into a topic
0:23
that's top of mind for every single B to B leader AI and your
0:28
go to market strategy. But more specifically, why
0:31
simply adding AI to existing go to market or account based
0:35
marketing workflows rarely changes outcomes.
0:38
And we're going to talk about what it actually takes to make
0:41
AI drive real performance. And I just realized that we had
0:45
diving into a topic Yas, as they start to introduce you.
0:49
So that will be relevant. Our guest today brings a
0:52
perspective that you don't hear often.
0:54
Yaz Dalkovich is the head of AI at Rab to B, where she's leading
1:00
the redesign of how a modern B to B agency operates in an AI
1:03
first world. Instead of layering AI onto
1:07
legacy processes, which is not what a either of us are going to
1:10
recommend today, Yaz has rebuilt the foundation itself,
1:14
rethinking the roles, work flows, collaboration models, and
1:17
even how data flows between strategy, ABM, media, and sales.
1:22
Here's the diving part here. She's also a former professional
1:26
free diver who set 8 world records, including dives to 400
1:31
feet, which is a background that is shaped how she approaches
1:34
complex problems under extreme pressure.
1:37
That's literal and figurative. But Yaz, welcome to the show.
1:41
Hello. Hello.
1:42
Very nice to join you, Chris. I'm excited to have you here in
1:46
the 8 World Records, and we have to start with that.
1:49
That kind of competitive mindset, doing things that are
1:53
challenging, taking risk, pushing yourself as far as you
1:56
can. What are some of the connections
1:57
between the free diving and what you do professionally?
2:02
Yeah, yeah, you got that very right.
2:03
So I've been trying to figure out this answer.
2:06
I mean, how is it that I am taking just as much joy and
2:09
drive so much passion out of AI, you know, leveraging AI for
2:13
business transformation as I did for free diving when back, back
2:16
when I was a child, I could not imagine a world like outside of
2:20
that. I was so passionate, so driven.
2:22
And the commonality is the, the complex challenge.
2:25
I just have the passion to solve very tough puzzles.
2:28
And if the solution is not clear, it just drives me crazy
2:32
until I figure out what is it that the pieces I have to put
2:35
together to overcome that. So as a child, I would be
2:39
fascinated by the ocean on the water documentaries.
2:42
And I will say, Oh my God, like this is like a place, like the
2:45
the space, like an alien environment that very few people
2:49
are Privy to see. And it's fascinating.
2:51
And it demands us giving up our most important survival instinct
2:56
of breathing to be able to get to see it.
2:59
I'm like, what can I do to figure out how I can spend more
3:01
time on the water, how I can go deeper?
3:03
And that just drove me like, I wanted to find out how deep I
3:07
could dive and how, how I could improve myself and how much time
3:10
I could spend. So I will spend, you know,
3:12
summer after summer, you know, practicing, trying to get
3:15
better. And then when it was time to
3:17
pick a career, the university, it was very interesting.
3:20
I had a second passion that was math actually.
3:24
I was thus just this nerd that through high school, I would
3:27
just borrow like calculus books from the the university and
3:30
solve it on the weekend for fun. It just was, you know, between
3:33
all my swimming trainings, that's what I would do.
3:36
And I said there was one school that was the hardest to get to.
3:40
It was like the I grew up in Turkey and it was like the
3:43
Harvard of the Middle East region that had a free dining
3:48
club. And I said like, I have to join
3:49
these people. This is my way to figure out how
3:51
I can dive deeper. So I put everything on pause and
3:55
spend like a year and a half studying like crazy.
3:57
And and I had to rank among like a few 100 people, you know, like
4:01
a two and a half million student exam.
4:04
And I got a rare like perfect math score.
4:06
And I got into the math department.
4:08
I picked math because I couldn't pick a career yet.
4:10
I just wanted to free dive. And I said, like, this is like a
4:13
foundational career. I think later on when I decide
4:16
what I'm going to do, I can pivot this to do something
4:19
useful, you know, like foundational understanding.
4:22
I believed in that and my plans worked exactly how I imagined.
4:26
So I got into that club and we got more and more into it on
4:30
until I set my first world record.
4:32
And the interesting that happened with that is free
4:34
diving is not an organized sport.
4:36
It's not a sport where there are competitions.
4:38
You don't have anything organizing anything for you.
4:42
So if you want to do a record dive, you have to put a team
4:46
together that will do the safety for you.
4:48
Scuba divers that will be stationed at different depths,
4:51
you have to hire them for a month so you can train with them
4:54
because it's a very risky sport. If I need help underwater, I
4:57
need these people. So you have to rent a boat, you
5:00
have to put them in a hotel, pay salaries, and I had to finance
5:03
this. To finance this, I had to find a
5:05
sponsor and to convince a sponsor that this is worth
5:09
financing, I had to create like a media image.
5:13
I had to record images underwater because nobody's
5:15
going to get to see it. Everything's happening in a
5:17
place that nobody get to see. So I had to put a camera crew
5:21
together. And again, this is like around
5:23
year 2000. It was not, there were no GoPro
5:26
cameras or anything like that yet.
5:28
So our team became like a 30 person team of like cameras and
5:32
like a live transmission and and I had to go like company to
5:36
company with a file under my arm, like selling myself,
5:38
convincing them that, you know, this was worth financing and
5:42
then organizing those things like impressive, impactful, do
5:46
all of my interviews afterwards. You know, I made appearances in
5:50
a very strategic way so that it became more appealing to the
5:54
sponsors and it did work out. It did make a quite a bit of it
5:59
generated a lot of interest in Turkey and Europe in general.
6:03
And I ended up doing 8 riggers. But all of a sudden I found
6:05
myself like 40% training and maybe doing running like 60% of
6:10
my time running a a marketing business almost.
6:13
I was a salesperson, a project manager, you know, reviewing my
6:17
legal contracts because there was no like sports management
6:20
for this unknown sport. You know, nobody knew about it.
6:23
I was APR person. It was interesting.
6:26
And then one day came that I said, I think I'm satisfied.
6:29
I found out how deep I can dive. You know, I'm a little bit bored
6:32
with all this chaos. I want to open a new chapter.
6:35
And I said, so let's go back to my second passion, math.
6:39
And I said, what is similar to math that I can enjoy that
6:41
cannot put into practice. And I decided that was
6:44
programming. So I taught myself, found myself
6:47
a programming job. I moved to the US and soon after
6:51
I said programming became a little bit easy.
6:54
So let me take it to the next level.
6:56
I said, I started getting on client meetings and saying, do
6:59
you have a tool you need, a challenge that you need to solve
7:02
that doesn't have like an out-of-the-box solution.
7:05
So I could invent the solution, put this unusual set of tools
7:08
together, solve problems and I started getting so much
7:11
satisfaction and joy out of that.
7:12
So I realized this kind of uncomfortable place, this
7:16
discomfort my was always my signal that I searched for soon
7:21
after I built this innovation team, like minded individuals
7:24
that I picked out of our development team that were like
7:27
curious, that was brave to face challenges and they took over.
7:32
And I said, how can I take this in the next level?
7:35
So I start tackling business problems to solve.
7:38
And when AI came out, it was like the business problem from
7:43
heaven, you know, like I have to figure this out.
7:46
I have to take 100% of my time. How do you leverage AI to solve,
7:50
you know, to, to transform a business in the best possible
7:53
way? So I've been focused on that
7:55
100% now and having just as much as fun as free diving.
7:59
Wow, that's an incredible background.
8:01
And I feel like any, anything we talk about the sense of
8:05
perspective going from true life or death, planning to cover
8:11
yourself, going 400 feet into the water versus sales and
8:16
marketing and generating revenue and pipeline.
8:18
It's a bit humbling, but we'll, we'll try to pull us through.
8:21
I mean, one, one thing that I'm really curious about that
8:24
founder story you were talking about with building the team and
8:27
getting the right people and planning the strategy and making
8:30
sure that you cover every single detail to literally keep you
8:34
alive and to train for a month and everything there.
8:37
As we've embraced AII think a lot of people have just bolted
8:42
it on to what they're doing. So it's, oh, I need to use some
8:45
some of the tools to help me a little bit with this versus
8:48
thinking about how that changes the entire foundation of their
8:52
go to market strategy. Why does adding AI often fail to
8:57
change the outcomes for B to B teams?
8:59
So adding AI to two components, what we do is definitely
9:03
helpful, obviously. And if you're not doing this as
9:06
where you got to start. And yes, a copywriter can
9:09
efficiently create 50 emails at once.
9:11
A strategist can research like hundreds documents at once,
9:15
while at the time they could do 10.
9:17
And if you're not doing this, you're missing out on
9:19
efficiency. But I think you're missing on
9:22
the biggest picture when you approach it this way.
9:24
And I think there's a huge opportunity to, to connect the
9:28
streams together. What we try to do is we use AI
9:32
to start with like a common brain on every project.
9:35
And, and this brain starts with wherever you're coming from.
9:38
What is the story of our client and what is the business
9:40
problem? And now all the strategies
9:43
copyright there's, you know, the media campaign sales, they all
9:47
drive from this common goal and they are all focused on the
9:50
problem statement. And instead of putting AI onto
9:54
the hands, we are putting it onto the handoffs in a way.
9:58
So we keep collecting all, all of our knowledge, our meeting
10:01
knows our findings into this common brain.
10:04
And this is how we drive the true efficiency.
10:07
And this is how we made the the foundational model change that
10:11
change things a lot for us. Yeah, I, I love that.
10:14
I'm curious too, when we use AI to solve some of the legacy
10:19
things that we do, some of those processes, I mean, we certainly
10:22
can see the efficiency gains. And you know, I'm wondering why
10:27
does that not typically move the business forward in a dramatic
10:32
way, right. I, I think from a lot of these
10:34
episodes, people are tackling AI to solve things that consume a
10:39
lot of time and it's not always attached to what drives the
10:43
business forward or what's the big strategy improvement or what
10:46
gets us into the next go to market.
10:48
I'm just curious how you see the juxtaposition between the
10:52
efficiency gains of putting AI into what you're doing today
10:56
versus planning to move the business forward with AI?
10:59
One thing I want to say is so you, you find an AI tool
11:03
integrated into the tool that you're using and it's very
11:06
impressive. Like you're, you're opening a
11:08
little chat on the side and you generate some insights out of
11:11
it. And it gives you something that
11:13
you may not have seen right away, but this gets just locked
11:18
in that area like you cannot, you still go back to saying what
11:22
you know. I drive more inside out of it
11:24
myself. I've been doing this for a long
11:25
time. And I don't think this this
11:27
little thing is helpful enough to change how I'm doing it.
11:32
And it kind of stays on as a sighting.
11:35
And everybody is very busy and they don't want to invest more
11:39
time into what I can do. And, and these little tools and
11:42
this little additions are not game changing enough to to drive
11:46
its change and impose upon you. And also people are a little bit
11:50
overwhelmed with the puzzle of how I can truly use this.
11:54
Now, something that happened that I think was game changing
11:58
is when after ChatGPT came on and everybody started believing
12:02
that this is this is our business is changing forever.
12:04
Many thought that and, and this was affected by the news.
12:08
Obviously, many thought this is going to be the way that all
12:11
that giant data set that we have all of our documents, species,
12:15
infographics, we can throw AI at it and make sense out of it.
12:19
And the news made it look like this was possible and this is
12:22
still not truly possible. Right now.
12:25
You need to put like an agentic AI structure layer after layer
12:29
to clean up the data, extract A logical sense out of it, build a
12:33
rag out of it, build agents that check that.
12:35
So now it's starting to become possible due to processes, but
12:39
AI is not capable of just putting that together.
12:41
A lot of companies that have the power to do so, you know,
12:45
deployed an LLM and through all the documents and it's like, why
12:49
is it hallucinating? Why am I not getting my strategy
12:51
or why can't I pull this documents out of here just like
12:55
that? So that was a really, really
12:57
tough moment, especially with big companies that have like
13:00
legacy workflows, Vegas processes.
13:03
Something that disheartens me is something that happened with
13:07
Microsoft, for example, with Copilot, like this was a really
13:11
sound business choice to go with Microsoft Copilot.
13:14
I mean, you're most likely in that Microsoft 365 workplace.
13:19
You have all your files there and it's built on ChatGPT.
13:22
You see the tool working really good.
13:24
It's utilizing the Open AI API model.
13:27
So you thought very logically, my documents are safe and this
13:31
is going to extract all the information I need.
13:33
I can search through my files. But even that couldn't make
13:36
sense properly and it didn't work well.
13:38
And, and we almost deployed this to the agency.
13:41
But I approach everything with my mathematical background that
13:44
I approach everything with the hypothesis.
13:47
Psych is, is this truly better? And I always run like a side by
13:51
side test. I would test it myself.
13:53
I would expand the test to like 10 different people from
13:56
different roles. Give them like a method about
13:58
how they're going to document how they're going to test.
14:00
Give them like a period of time and, and we report to each other
14:04
on what the outcome is. And we saw that this, this LLM
14:08
didn't work, this copilot didn't quite work.
14:10
Hopefully it's getting there now, but we had to do the hard
14:14
work instead of that. We had to focus on processes, on
14:17
education, on the slower but the harder version of doing this,
14:21
and not expect this magic to happen.
14:23
Yeah, I relate to that. I was trying to build out a
14:26
personal workflow and I'd gone through some of the GEM training
14:30
using GEM and I and chained a bunch of different work flows
14:34
together. So sequentially everything made
14:37
sense. It was doing the research and
14:38
then spending it out. And then I got to the video
14:41
creation and it started with a product and everything was
14:44
great. And then it was a cardboard
14:47
knife. I was just testing it with cool
14:49
little things that I have around the house.
14:50
So, you know, $7.00 product on Amazon and it started, they're
14:54
holding up the product and doing the video and then it goes to
14:56
the box and it switched to a giant machete.
15:00
And it's something that's simple, could throw off the
15:03
entire workflow if you're not checking it and building it from
15:07
the very beginning and looking at every single detail and every
15:10
single step. I was just taking the easy
15:11
approach, like here's what I'm trying to solve, build out the
15:14
workflow. I'm curious for a complicated
15:17
business and the agency you redesigned your entire operating
15:21
system, where did you start? You know, how did you realize
15:24
the system needed to change and not just using AI tools?
15:28
So I realized that there was a huge efficiency in using getting
15:33
good at using an LLM. Like I didn't want to fall for
15:36
this game of this like thousands of products pushing upon us with
15:40
amazing AI capabilities. I told we were not going to get
15:44
out of this puzzle. But when you truly understand
15:47
the value AI and instructing an LLM and and getting some some
15:52
solutions out of it, now you're getting in a foundational
15:55
understanding of how you can utilize this.
15:57
So I said to our CEO, like we need to deploy like an LM
16:01
licenses to everybody. And I ran my test.
16:04
I decided, you know, this is the one that I think we should pick,
16:07
but that's feature proof is smart.
16:09
And you know, until we get all those licenses, I think, you
16:12
know, we should be on hold because this is going to be our
16:14
75% of our efficiency will come from that.
16:18
And he moved with every one of these items like right away.
16:21
And that's why one of my workplace, no red tape logic
16:24
rules. And then I started a training
16:27
program. So I recorded a one hour
16:30
training that covers all the foundational parts of how to use
16:33
the LLM, all the features so that nobody is missing any
16:37
important parts of the puzzle. And I had designed an assessment
16:41
for everybody to take. So that was a requirement to get
16:43
your license. And after that we launched this
16:46
like 15 minute sessions weekly, but where I will cover a
16:49
different topic. Little by little, other people
16:52
started covering these topics and the power of that was
16:55
incredible. I mean, it was just 15 minutes a
16:58
week and we always fit this, this cool utility in that 15
17:02
minutes. And I, I didn't see it coming.
17:05
I didn't think that that was going to be one of the biggest
17:08
parts of our wins. I had this timeline planned
17:11
where after a certain while we were going to start developing
17:14
our commercial package, our our go to market solutions, what
17:18
they're going to be, how they're going to be.
17:19
But this training got us so far that the account managers all of
17:24
a sudden started like extracting this out of me and it came
17:28
before I was even ready that, you know, they want this and
17:31
they want this. We're already proposing this.
17:32
I know this is possible. How do we solve this?
17:34
So we put our expanded our team really fast to meet the demand,
17:39
but that simple act of for good training and this training was
17:43
not run like a normal education plan.
17:45
I thought it was like a promotional campaign.
17:48
Basically it is about look at the cool things I can do with AI
17:52
and you can do them too. Isn't that amazing?
17:54
So that's how we approached it and that's how we spread and
17:58
became like a workplace culture. I love the workplace culture.
18:02
I also love the consistency, the 15 minutes a week and making
18:05
sure that people are continuing to educate themselves and do it.
18:09
I'm curious about the pivot from day-to-day.
18:12
So what they did every single day, how you did the pivot from
18:17
just normal work that every single person was doing into
18:21
higher judgement, higher impact work.
18:23
So how did AI help that transition?
18:26
We were organically responding to the changes that was
18:29
happening faster than I thought through this training plan and
18:32
and everybody utilizing their licenses.
18:36
And when I think back at it, I wanted to put one of my blog
18:39
posts together. I realized that the recipe came
18:42
down to this 4P methods, which was like people, process,
18:46
platform and proof where we started with the people, where
18:50
we were, you know, doing this training, we were supporting
18:54
their journey. I also implemented a step that
18:58
keeps track of everything that contributes to our final AI
19:02
goals. All the elements that you can
19:04
do, whether it's a certification you take, a tool that you
19:07
learned about using AI in a job or following some certain leaf
19:12
setters and developing custom GPTS.
19:14
All of those were elements. And I gave them certain weights.
19:17
And with that, I could keep track of like who's doing what
19:20
and find out who are our champions, who are the ones that
19:23
need a little bit of help so that I could trigger a method
19:27
to, to bring them up to date to, to run the promotional campaign
19:30
a little stronger to them. And that was the pupil step.
19:33
And then the process was. So from now on, this is the
19:37
folder that every project that starts, we're going to log all
19:40
the transcripts. You know, we're going to record
19:42
the meetings, we're going to log the transcripts and we're going
19:45
to put these transcripts in a, you know, notebook LM, you know,
19:48
which stays really true to the to the source, right?
19:51
This is the amazing tool to consume massive information with
19:55
as little hallucination as we can get.
19:57
Then we are going to start briefs with.
20:00
Not, not the way we used to write them, but about through
20:03
briefs that that AI can consume and everybody can can merge on
20:07
this true idea. And then are we would wrap up
20:11
our projects with like an AI summary of what happened, what
20:14
didn't work and Add all that collective information.
20:16
So that was a process step and then the platform was like we
20:20
kept testing platforms. I mean we did land on an LLM
20:23
that you know, we facilitate for everybody.
20:25
But then there are a lot of tools that we have to follow,
20:28
but we approach every tool with hypothesis, you know, instead of
20:32
the hype and this tool can do this and let's go try.
20:36
We would come to it with a bit of a test like this is what we
20:39
should be able to do it. This is our problem statement.
20:42
Does this tool really fit this problem statement or it's just a
20:46
little, you know, cute thing to do something.
20:48
And then we would decide whether there's value to using that or
20:51
not and document and make it part of our process.
20:54
And the last one was the proof where every initiative we did, I
20:59
would attach AKPI to it. And, and sometimes it's very
21:02
hard, but AI is helpful with that too.
21:04
I would say like this is what we're going to test, you know,
21:07
what is a helpful KPI for me half the time AI gaming, really
21:11
good ideas. And we would track sometimes
21:13
it's very simple that you check how much time you used to spend
21:17
on it and now how much time you you spend on it.
21:19
And we ran comparisons on that. But sometimes it's how much much
21:24
leads you generate, how much, you know, demand that this gets.
21:28
How are the media campaigns running of the attribution of
21:31
the different channels? How how are they coming
21:33
together? So by attaching a proof to
21:36
everything that we ran, we could see what worked, we could see
21:39
what didn't. And we've been doing nothing but
21:41
moving forward thanks to this little elements, very simple
21:44
elements, a little bit cumbersome compared to the nice
21:47
easy like push the button and make it work way, but not so
21:52
much either. I mean, it becomes like a
21:53
collective passion. I love that.
21:55
And you've talked previously about building an AI fluent
22:00
organization, not just an AI curious organization.
22:04
What's the difference and how does that show up and how teams
22:07
actually operate? So yeah, I mean we started AI
22:10
curious, you know, spent a year everybody testing tools and
22:14
saying, oh, you know, big journey is amazing.
22:16
Look at what I did and and somebody else coming like I did
22:19
this with the guitar copilot tool.
22:21
And and so we got over our curiosity face and mummies start
22:26
to develop this foundational approach on this four piece.
22:31
Then that's that became a fluency thing.
22:34
And that happened thanks to our little 15 minute sessions of
22:38
teaching each other, learning from each other, just sharing
22:41
the good things, bad things we find.
22:43
But this becomes a shared language.
22:45
Right now everybody knows that we can utilize AI to bounce
22:49
ideas, to brainstorm to, you know, The funny thing is like
22:53
you cannot sometimes create an idea, but people are amazing at
22:56
criticizing existing ideas. So there you go.
22:59
Is your, your body to criticize all you want.
23:02
You help AI to create something and you improve it by
23:05
criticizing and making it better.
23:07
So this is the shared language that I'm talking about now from
23:10
all the way from account managers to our coders or
23:14
strategists, we know about this, this process that we're always
23:18
going to contribute to the common brain.
23:20
We're always going to find this resource that we can quickly
23:24
pull utilizing AI that will help our processes.
23:28
And everybody's looking at opportunities to share and lift
23:32
each other up. And the other thing that we are
23:34
doing is with the shared language is every department
23:38
that have a certain expertise, we usually turn them into like
23:42
custom GPTS. For example, our research team
23:45
is, is amazingly good at competitive research and they
23:48
have a methodology about it. None of us know it, but I like
23:52
that me as the AI department or somebody in the media
23:55
department, they, they have to run competitive analysis too.
23:58
So now we get them to develop this custom GPT that maybe it's
24:02
not client grade, project grade like strategy teams, but we can
24:06
run the analysis that can yield the results much better than
24:10
they used to be. So now we are sharing each
24:13
other's expertise, copywriters share tools with the other teams
24:16
to write better and so on. So that is the shared language
24:19
and the AI fluency that's I think we got to, we got to shoot
24:23
for. I love that.
24:24
And yeah, as you mentioned in previous conversations that one
24:27
of the outcomes that's really important is a more accurate
24:30
understanding of buying groups, which we're huge fans of here at
24:34
Demand Base. I'm curious how rebuilding your
24:37
go to market system did that change and if so how the way you
24:42
see and engage buying groups? So with ABM, like your target
24:47
list was more of a static way of going at things.
24:50
You would identify the people, you would do research about it
24:54
and design your ABM at it. But what AI brings, and even
24:59
before AI became a thing, it was like machine learning systems
25:03
that would organically pull clusters out of data, the
25:07
organic real ways of how people are behaving.
25:10
So now with AI, we can go through your entire targets
25:15
company's behavior and see what the story is because you know,
25:19
you're not Privy to the story of who's actually influencing the
25:23
CEOCFO to buy. And with the behaviors now we
25:28
can very quickly and very efficiently target like a
25:31
technical deep dive to person A and while we target the ROI to
25:36
the person C or we can make a business case to person B.
25:39
And these are not necessarily the people that you thought were
25:42
the case. But we are not forcing a message
25:45
to a list, but we are fitting our answers to the conversation
25:49
that AI is revealing for us. So that's how this transforms
25:53
for us that change the game a little bit.
25:56
Well, that and yes, I want to tap into the mathematics
25:58
background here. You've talked about KPIs.
26:01
I'm curious, how did you measure the success of the new AI
26:07
operating system and to see what's working, what signals
26:10
mattered the most and to make sure it's not just a different
26:14
AI exercise? For every different AI infusion
26:18
we did, you know, we always started with the KP is we would
26:21
determine what is it going to be at the start and we will track
26:24
against this. But what is very interesting is
26:27
this one KPI where I developed this system of identifying all
26:33
the things that people do that contribute to their AI fluency,
26:37
that participation to the program that surfaces how well
26:40
they're using chat GPD or you know, how well they're using,
26:44
utilizing the AI tools there. Or you could generate insights
26:47
or or following the news, doing certifications or following our
26:51
trainings, all of those things. Actually, I mean that that is
26:55
the KPI that drives the solution are followed like a score, every
26:59
score for the company, for the departments, for people, and
27:03
somehow everything is completely parallel to that.
27:07
So I think if you just focus, one person forgets about the
27:12
other stories and dedicate a department or person dedicate an
27:16
initiative to educate everybody and support everything they want
27:20
to do with that common goal. I think the rest just follows
27:23
because AI is amazing, impressive, just you got to give
27:27
the people the opportunity to unearth that.
27:30
But now like I, I keep getting back an immense amount of like
27:34
look at the insights I started writing things to AI.
27:37
And the other one is look at the strategical, you know,
27:39
presentation I put together that I couldn't do before.
27:42
Like this is just happening organically based on the
27:44
original initiative. So I think just focusing on that
27:48
KPI alone does a lot of the work.
27:51
Outstanding. And I love what you're hitting
27:53
on there to kind of the benefits of us as humans using AI and
27:58
what that can help us achieve. I'm curious to as AI takes more
28:03
of the manual load and a lot of the time consuming things away
28:06
from us, what becomes more important for us as people and
28:10
go to market teams? And where do you see leaders
28:13
under estimating the value of human judgement?
28:15
The biggest things that we focus as to go to markets with is one
28:20
of them is Geo on your digital appearance.
28:24
And this story is, is funny because technology keeps pushing
28:27
us to do things right. We started with like injecting
28:30
like keywords and the metadata or page and and that used to
28:35
work for a little bit. And then, you know, Google said
28:37
like it's not going to work right, Proper content.
28:39
Let me find those keywords there.
28:41
And then it became like a website.
28:43
Why? It's general understanding that
28:45
serves the purpose and what people are searching.
28:47
And now AI is giving the answers and forcing the business to put
28:52
their whole story together. I mean, how you're cited, how
28:56
people discuss you. So you you have to truly
28:59
transform your business from inside out.
29:02
You need to fix those stories as they appear out there and fixing
29:06
them is there's no hack around this.
29:08
You know, you don't have to do things right if you're not
29:11
driving through gain efficiency. You have to change how you do it
29:15
so that your story is authentic and it's full of substance.
29:19
So I think that AI needs to pull everything boring that people
29:24
do, that everything repetitive and freedom to bring and what
29:28
they do best. You know, the human judgement,
29:30
human uniqueness. One thing I like to talk about
29:34
is how AI, like machine learning in general, is the art of
29:38
averaging. So that is the system is based
29:41
on getting a massive amount of data and cut out the outliers,
29:45
the, the ones on the top, the ones at the bottom and find the
29:48
pattern in the middle and squash it.
29:50
So obviously this is a way to heighten your your skills that
29:54
didn't exist before to an average level.
29:56
But then this is going to kill the creativity and uniqueness
29:59
you bring to the picture. So we always focus on people
30:02
just getting the boring stuff out of their life and do nothing
30:06
repetitive. And now like they have the time
30:09
and the the drive to focus on what is it that they bring to
30:13
the table? What is it that makes them human
30:15
and truly transform your business so it serves the
30:18
clients in every way. And this is how you fix your You
30:22
know Geo and your friends will be represented, right?
30:26
Awesome. Yes, we have a couple quick
30:28
questions. We ask the same few questions at
30:30
the end of every single episode, and it can be as long or short
30:34
as you'd like, so these could be quick hitters or you can expand
30:38
on anything you want. The first What are your favorite
30:40
tech recommendations for everyone listening?
30:43
So I'm not going to bring any any interesting tool I didn't
30:47
know about. I believe that you can drive so
30:50
much efficiency with the good use of an LLM.
30:53
And my choice is either ChatGPT, Gemini or Claude.
30:57
I'm, I'm fascinated with Gemini lately as Gemini 3 came out and
31:01
I feel like that we are now deploying a lot of agentic AI
31:04
structures and a lot of workflows.
31:06
But it always starts with a good use of LLM, like we write the
31:10
recipe and like custom GPTS and instructions and processes and
31:13
steps. And then we turn those steps
31:16
into something more static that that the agentic structure with
31:19
the rack. So that's my belief there with
31:22
the tools. Awesome.
31:23
What about good reads? Now we're very generous with
31:26
reads. It could be book, blog,
31:27
newsletter, podcast, video, any type of format or content.
31:32
But what are some of your standouts?
31:34
So one thing that I will always make sure I listen to every week
31:38
and I recommend everybody is the artificial intelligence show.
31:41
This is our podcast by smartericks and I consume a lot
31:45
of, you know, AI information non-stop podcast.
31:48
But but if you don't follow all of it's like an AI person does
31:52
you. This is the funnel.
31:54
You know, they will filter out everything necessary and bring
31:57
to you what matters most and you will not fall behind with any
32:01
news that matter to your business.
32:02
I like to follow a Gartner Tink cast podcast also.
32:05
I find it's very nice. It's a lot of interesting ideas,
32:09
business ideas. And one thing a book I like a
32:12
lot is Mogul. That's the, what was the name,
32:15
the a little voice in your head. So he has a way of approaching
32:21
business problems like maths, which is something I need a lot
32:25
and I like a lot. And he has this methods in his
32:28
book. It doesn't apply to direct it to
32:30
your business, but it gives you the recipe on how to really
32:35
truly free your mind to think, to put your best up front, which
32:39
as AI helps us with so many things.
32:42
That's going to be the most valuable thing we humans bring
32:45
to the table. Awesome.
32:47
What about good people? Anyone that you would recommend
32:49
we bring on the show or that everyone should follow?
32:52
So again, I come back to Paul Rotzer, the the CEO of Smarter
32:58
X, that's again, I love his approach to AI.
33:01
It's just there's no hype, you know, it's it's amazingly
33:04
helpful methods for companies. And then we have a Co worker of
33:09
mine, my boss, Alexander Rue, she's the SVP of product
33:13
delivery in our team. But what is amazing is, is as I
33:17
put this AI strategy together, I mean, we needed like an
33:21
operational leader that put this interaction that her methods
33:24
were were key to transform this into like how we really truly to
33:30
make it happen with the teams, how we turn them into products,
33:32
how we story tell so that these products are so vague, they're
33:36
not truly invented yet to make sense.
33:39
So that will be very interesting approach that from a business
33:43
perspective, go to market perspective.
33:45
And the last one is Mogul, that guy.
33:47
I love listening to his books and podcasts and his
33:51
presentations. Again, it puts my mind at ease
33:55
and it helps me find what I need to really think to to get the
33:59
best out of myself. Great suggestions, we will
34:02
definitely follow those. And the most important question
34:05
of all, this is the very last one.
34:07
How can everyone get in touch with you after this episode?
34:11
I think LinkedIn will be a really good way.
34:14
I put, you know, a lot of posts there and I'll respond to any
34:17
kind of notices. Yeah, I also have a blog into
34:20
the Deep, ai.com that I share all my thoughts that in my
34:24
inspirations on various topics, but I'll share them on LinkedIn
34:28
too. So that's the best way.
34:30
Awesome. Well, thank you so much for your
34:32
time today. It's been a pleasure.
34:33
Yes, for me too. Thank you so much.
34:36
Thanks for joining us for On Base.
34:38
If you enjoyed this episode, please be sure to share, rate,
34:41
review and subscribe on Spotify, Apple, YouTube, or wherever you
34:45
get your podcasts. This show is brought to you by
34:48
Demand Base, the only account based experience platform built
34:51
for sales and marketing alignment.