January 24, 2023 • 48 mins
Artificial Intelligence (AI) tools lead the forefront of current marketing conversations in 2023. Joey and Nels are diving into how they can be used by industrial marketers and what you need to know to be prepared for the coming revolution.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Joey Strawn (00:00):
Welcome back, everybody to another episode and
another season of the industrialMarketer Podcast your place for
the tips, tech trends andtactics for industrials who care
about driving leads and revenueto their businesses. We are
back, it is another season andanother year, I am one of your
hosts, Jo E. AI, and I am herewith some artificial
(00:24):
intelligence. And we are goingto talk about some fun
technology. Oh Nels. I'm soexcited to be back. And we
decided this quarter, we'regoing to talk about tools in the
tech of marketing. And whatbetter place to start than the
buzziest buzz topic that ishappening right now, which is AI
(00:45):
Artificial Intelligence and thetools that surround it. I mean,
are you excited to dive intothis topic? Are you excited to
be back one for season threeNels?
Nels Jensen (00:55):
Oh, yes. And just to let you know that the title
for this episode should be aImarketing strategies for
manufacturers. Perfect. That'swhat that's what AI tells us.
Joey Strawn (01:07):
We will listen, we are now answering to not the man
but the machine. And so this isgonna be great. Now
unfortunately, for all youlisteners, Nelson, I are still
humans. This is still a humanled podcast, we haven't given it
all over to AI just yet. But weare going to dive in. I mean,
again, this is a topic thateveryone from in the trenches,
(01:30):
to hopefully CEOs and leadersand drivers of their businesses
are thinking about these tools,not so much, you know, the
Minority Report or iRobot of itall, that's not really the the
artificial intelligence thatwe're going to be diving into.
But you know, what tools are outthere that are usable? What
(01:51):
investments should be thoughtof, or structural
infrastructure, ideas andconcepts should be considered
for your organization? Those arethe types of things that we
really want to dive into on?
Where do these fit in? What dothey realistically look like?
And what does it mean for peoplein our industries, which is more
like the industrialmanufacturers of the world? The
supply chains of the world? TheE commerce businesses of the
(02:14):
world? Like what, what is AIgoing to do for us?
Nels Jensen (02:19):
Yeah, and I think, you know, really, this falls
into a larger b2b marketingcategory, like a lot of ours do.
It's not, you know, hugely, it'snot hugely different that we'll
talk about nuances andsubtleties. But let's also be
transparent here, Joey, we arenot experts on the AI
applications here. This is verymuch new to us, as well as new
(02:39):
to, to the people out there whoare listening. We have been
playing around a lot with theseexperimenting, we've used some
of these tools. But we'relearning it too. And we're
learning the power and the theguardrails and the you know,
limitations and all that. Sowe're right, jumping in there
with everybody else. And I wouldsay we are far from the
(03:02):
authoritative sources on whatexactly you should use. So I
think, maybe around one cornerahead of a lot of us, right.
Joey Strawn (03:10):
I think that's a very important point, Nelson, I
do want to back that up and say,Yeah, we're not experts, and
we're not authorized resellersor re dealers have any of these
tools, either anything wementioned in this episode are
things that we've either playedwith, read about, or have heard
about, or have been told toexplore. So these are tools that
we know are in this universe,these are things that we know,
(03:32):
are going to pop up inconversation this year, whether
it is you know, at a meetup, orat an industrial group or
online, or maybe your boss comesto you and says, Hey, what about
this chat GPT thing? You shouldhave some sort of answers and
know what's on the horizon. Andthat's really what we want to
focus on today. I mean, knowswe're not even, we're gonna
mention a lot of tools, I thinkthroughout the conversation, but
(03:54):
really, the heart of theconversation is like, what does
it mean, for us, you know,manufacturers and marketers and
industrials. You know, how doesit realistically going to be
implemented? I mean, like, let'sbe clear. We're not saying that
by the end of this year,everyone's gonna be wearing
virtual reality headsets, and,you know, the haptic hands and,
you know, controlling things inan artificial realm. That's what
(04:17):
realistically, can we thinkabout this year, and then we
just want to go through a lot ofthe types, there's a lot of pros
and cons, and a lot of ways thistechnology is being used. So we
just want to focus on some ofthat. I mean, I'm getting I'm
excited. This is something thatyou and I've been reading a lot
about Nels. So I know you, youand I have each played with
different tools and systems. Andso I'm excited to see what
(04:39):
you've learned and figured outand I'm excited to bring to the
table what I've got, I mean, butjust kind of to start when you
think of artificial intelligenceand these AI tools and this big
trend that's been happening,it's being talked about right
now. What do you feel itactually means for for
industrial marketing, and peopledoing what we do?
Nels Jensen (05:01):
Well, I guess I have a hard time with the big,
grand umbrella of it all. Ithink it's if you if you are in
the industrial sector, youunderstand predictive
maintenance, right? Youunderstand how equipment has
patterns and telltale signs. Andthere are definitely things that
(05:26):
it tells you are going tohappen. And the key thing there
is closer to a timeline for whenthey might happen. And I think
artificial intelligence islargely prescriptive in that,
yes, it's, it's telling you it'sroadmap, it's telling you where
to go. There's also you get intosome of the different
(05:49):
applications about actuallydoing the work for you. But to
me, that's the more cautiousarea to go to me, I view it as
more powerful for a roadmap andguideposts than I do for the
actual execution.
Joey Strawn (06:07):
Yeah, and I think that's an interesting point to
make, too. And, you know, thisis a fancy term, and a lot of
people are talking about thisnow. But even before we started
recording now, so you and I weretalking, saying, well, these are
stuff that we've been using fora while, like artificial
intelligence is the way thatwe're sort of putting it all
under the same umbrella. Butmachine learning has been a
(06:28):
topic that we've been talkingabout for a while it's in a lot
of the tools that we use, likethe SEM rushes and the male
chimps of the world, you know,that's looking for machine
learning. automations, you wetalked about big data, you
remember, like a handful ofyears ago, when big data was the
buzz buzzword? Well, the naturalnext step in the gathering of
(06:48):
big data is that, well, we haveto have computers to compute the
massive big data, and then tellus what that means. And that's
what intelligence is, that'sartificial, right.
Nels Jensen (06:58):
And that's what history of that industry. 4.0 is
basically putting data to workfor insights. And, you know, and
that, and I think the reason whythis is such a big topic right
now is just we've reached atipping point where it's yes,
it's not just for the data nerdsout there, we've now reached the
(07:19):
point where the usability issuch that almost anybody can do
it. And, you know, we've come tothe point where we have to
figure out more about trainingand more about, you know,
applications. And we'll talkwe'll talk more about that, you
know, coming up, but, but wehave definitely reached a
tipping point, in terms of AIwhere it's, it's just not, if
(07:44):
you're just thinking, well,that's for the data scientists
and the, you know, analyticspeople, you know, no, it's we've
reached that point where it'sgoing to be for all of us. And,
you know, if you're, if you'vebeen lagging, and you know, yes,
now, it'd be a good time to getup to speed and begin to stick
(08:06):
your toe in the water and figurethis out.
Joey Strawn (08:08):
Well, and, and that's a great segue to because
one of the things that, youknow, we want to definitely talk
about today is the ways that wecan realistically see this
happening within the nearfuture. So I mean, again, you
don't have to dive in and becomethe biggest data scientist in
the world. But now it's to yourpoint, we've reached the tipping
point where it's going to becomenot only commonplace, but sort
(08:31):
of expected, especially with theproliferation of, you know,
tools, like I mentioned, jetchat GPT earlier. You know, last
year, there was a big socialmedia meme explosion around AI
image generators, like midjourney, and Dali, and some of
those, and, you know, it was anentertainment and meme driver a
(08:54):
lot of last year, but to yourpoint, we're getting in now
where it's going to be a lotmore economical, it's going to
be a lot more systematic and alot more scalable. So a lot of
people will have access to thetools. So sure, sure, chat bots
are a normal thing that I see. Iwould bet by the end of this
year, a lot of chat bots are runby some AI mechanism.
Nels Jensen (09:15):
Oh, and we've seen that we've all seen that on on
websites, too, as the, you know,the volume and importance on
online research takes hold andin the industrial sector
chatbots are I now see them moreoften than I don't exactly that
people people might not think ofthat as artificial intelligence.
(09:38):
But it is a very simple even ifit's just sorting you into one
of four or five differentcategories for you know, some
type of connection somewherethat is still, you know,
artificial intelligence at usright now,
Joey Strawn (09:53):
And especially with tools, you know, we're getting
better about having access tosystems you know, Bambara is a
Big tool out there, but intentdata, you know, the more access
to intent data and historicalrelevance of activities, the
more what do we mentionedearlier, those big data patterns
can be recognized by machinesthat are trained to do it. So
(10:15):
all of these pieces have been inplay for a handful of years. And
I mean, in real in an honestyand in all real NIS advertising
has been doing this for years.
Programmatic advertising isbased around the art, like
artificial intelligence andmachine learning of large
amounts of advertising and bigdata. You know, this is
something that has been in usefor a lot of marketers for a
(10:37):
while whether or not we'veactually thought of it as
artificial intelligence.
Nels Jensen (10:45):
Yes, and, you know, if you are putting your even
lead scoring into a CRM, it isfeeding into an engine that, you
know, HubSpot AI, or you know,some of these other ones as
well. And I think the one theone thing that to me, we're kind
of jumping all over the place,but that's okay, the analytics
(11:07):
side of it. So GA four, we didan episode recently on GA four.
And that's one of the coolpromises of the latest platform
for Google is that they'rebasically going to help you
determine what campaigns have ahigher level of success
predicting audiences, basicallyhelping you, you know, just give
(11:30):
you a much clearer course ofaction to take, you know, for a
campaign, you know.
Joey Strawn (11:37):
That's one of the major tenants of GA for is one
of the reasons the platform isevolving, the way that it is, is
stepping away from individualproperties and managing those
but looking at the entire matrixof everything going on, and then
get gathering learnings fromthose activities in those trends
across the spectrum. And somachine learning has been
(11:58):
something that they're veryheavily pushing, and I'm happy
that you mentioned analytics,because nails this is really,
really where the rubber meetsthe road. And I think where we
should spend kind of up a goodmeat of our conversation today
is what what is this going tolook like in 2023? Like, there
are a lot of different aspectsof our jobs that we you
(12:19):
mentioned analytics. I'vementioned content creation, to
some degree with chat GPT, orchat bots and things like that.
But there are tons of uses forthese, like, you know, just to
go through some of the typesthat you and I mentioned, when
we were planning this is contentgenerators, image generators,
advertising, analytics, CRM, andautomations, devel development
(12:44):
and code generation, a lot ofthese things already exist in
already are out there. I mean,just we can start at the top on
this one, like contentgeneration, what do you Nelson
imagine the future looking likewith these AI tools? And content
generation? Or like, are yougoing to lose your job? Are we
going to have to like, replaceyou with a robot like a Nelson
bot point?
Nels Jensen (13:05):
Oh, yeah. The not yet. And I think there's a
general, there's, there's thebasic, mostly right. View of
content generating AI? And, youknow, it depends, it depends,
obviously, what is the pool ofinformation that this AI is
(13:27):
working from? So you have thethe new, the crazy tipping point
tool is the chat GPT that is...
Joey Strawn (13:35):
Schools are banning, and then everyone's
getting up, and I was like, Oh,it's right, in the presidential
speeches or whatever.
Nels Jensen (13:41):
Right and it's, uh, but it is, if you've never
played with it, do so Irealized, if you jump on it a
lot of times right now and youdon't already have an account,
you're not going to get oneright away. It's really cool.
It's, but it's scraping the webto write these things for you.
You give it a prompt, and we cantalk a little bit more about,
(14:01):
you know, the importance andunderstanding what exactly
you're asking it to do. Becausethat becomes, you know, if
nothing else, I'm not going tolose my job because, you know,
how do you interpret exactlywhat you want? Exactly, putting
that into an ask? So but that'sdifferent than like, Jasper,
which is actually made by thesame people open AI, the same
(14:23):
company. You know, we're, we'reJasper will take for instance, I
can provide, here are threedocuments that I want you to use
to draft and then you ask forwhatever 1000 word blog post,
and you give it obviouslyinstructions, but Jasper is
basically allowing you toprovide the inputs, you know,
(14:45):
chat GPT is not it's basicallyand I'm sure at some point it
will. I mean, this is the peoplerunning this open AI the
company, you know, are verysmart. They basically put chat
GP He out there. And they'veprobably advanced their
development work because ofeverybody playing with it by
(15:07):
yours in a matter of two weeks.
Oh, yeah. Or I don't know howlong it's been out there. But
it's. But yes. So there'sthere's fundamental differences
in terms of what are what arethese tools using to generate
content? And that's superimportant to understand. And
then, yeah, and then
Joey Strawn (15:25):
I think when we compare pros and cons later,
that'll be Yeah, that'll besomething to dive into,
especially with the imagerystuff, because that's got to
come from somewhere. But like,even things like content bot,
you know, I think is a fun one.
One of the things that you and Ihave talked about I, this is the
joke that I like to say rightnow. But I really view
especially the content AIgenerators as like a really
(15:46):
great virtual assistant. Youknow, it's like, give them a
prompt, and they're smart, andthey can go anywhere on the
internet and get everything andput you together something.
Yeah, but it's going to have tobe edited. It's going to have to
be reviewed for accuracy andmake sure that the sources are
verified. There's work to bedone. It's not a final product,
but it's a great assistant.
Nels Jensen (16:08):
A shout out to our peer Virginia Roberson who she
calls it her, you know, her topnotch research assistant. Yes,
exactly. And we'll talk a littlebit more to as well research is
certainly a huge use for it.
There are others as well. Butyes, she's, you're both correct,
right there, virtual assistantresearch assistant. You know, it
(16:29):
is very, these are very, verypowerful tools. And I'm, I've
gone from the oh, well, we'lljust wait and see how this plays
out, too. All right, it's playedout enough now that I need to
get a better handle on it.
Joey Strawn (16:45):
One of the first things that I remember seeing in
the market, or at leastconversations about this, where
the image stuff like mid journeywas one of the first ones I
remember our developer, we'vehad him on the show before Brian
shared a bunch of Dali stuffearly on, like last year. And I
was like, Oh, wow, this ispretty crazy. And then you dive
into like, where's this? Like?
How is this getting? You know,photo, I think is another one.
(17:08):
So there's a lot of the imagestuff that I don't really
understand. But that's where Ikind of first saw it. And that's
gonna be really interesting inmy mind to watch on, like, what
can those images be used inAdWords? What's the copyright on
these like, because images aredifferent than the semantics of
words and how words are puttogether? Because those images
have to be sourced, the pixelshave to be sourced. So it's,
(17:31):
that's an interesting, what'skind of one of the first ones I
ever saw. But getting deeperinto it, like you mentioned,
analytics. And I've mentionedadvertising, like those are two
very in the weeds element ofsomething that we do as a
marketing team, that now AI iskind of dipping their toe into,
I mean, outside of you know,like the ad roles and the
(17:53):
programmatic advertising. Youhave sem rushes and GA fours and
machine learning analyticsplatforms that are giving pretty
good advice on how to updatepages, how to look at
optimizations, and where totarget advertising. So it's
Nels Jensen (18:12):
Yeah, one of the beauties, one of the beauties is
how they link together too. Soprogrammatic tells us, here's,
for instance, you know, avenuesto track users, and where, show
them this message where they go.
Right, right, and then you know,you're using some of these
different channels to do that.
(18:33):
And what you might might findout is that you see analytics
that oh, yeah, there's a hugenumber of people who are seeing
this. And there's very littlestickiness, right, so you're
learning two things that yes,this is it does deliver
audience, it does not delivervalue, you know, I mean, it's
(18:54):
the the, that's the power of theanalytics is isn't just telling
you, Hey, do this and saying,Hey, do this. And oh, by the
way, that wasn't the greatestthing. You know, it'll
ultimately help us, as I like tosay, play offense and defense at
the same time, right, doubledown on what works and cut back
on what doesn't, right.
Joey Strawn (19:12):
It is, it gives us an additional reach as marketers
to get our hands around the bigdata. You know, it's the data
source that we talked aboutearlier. And that's really what
a lot of this comes back to ishaving the history of metrics,
having the access to InternetTrends and culture and the ways
that we use words and images ispulling all of that together,
(19:34):
where we wouldn't have the timeand energy to do so.
Nels Jensen (19:37):
Right. So. So with automations, right. So, you
know, when I first heard aboutthis, you know, I'm much newer
to marketing than you are, butwhen it was like sort of this
personalization is going to beautomated. I'm just like, Well,
that'll be interesting. Andit's, you know, and many in many
(19:57):
cases, it wasn't perinternalized it was targeted.
Right? But it has been it hasevolved to be personal where you
actually can have theattribution sources so that they
can, you know, yeah, deliversomething 100% unique to me
versus you even though we havemany of the same behaviors,
(20:18):
right?
Joey Strawn (20:18):
Well, and you mentioned within automations,
I'll expand that to include likeCRM platforms and marketing
automation tools, too, becausethat's one area that we know
have been playing around withartificial intelligence for a
while. I mean, HubSpot ingeneral has been talking about
their machine learning, theyhave a lot of big architecture
that they're working on, aroundmachine learning to provide
(20:40):
insights on everything that'shappening within their tools.
But the big one right now, Ithink that a lot of people hear
about and hear talked about isSalesforce, Einstein, Salesforce
launched their Einstein platforma little while ago, but any, a
lot of people probably heard itand know what the name is, they
may even use it. But it's basedon the machine learning, it's
(21:00):
based on machine learning andinsights and trying to help make
a lot of those things that usedto be very manual within
automation tools, a lot moremachine focused, or a machine
driven. So whereas, you know, ahandful of years ago, we as a
marketer, you would have to gointo HubSpot, or go into
Salesforce and make filters andrule for every situation that
(21:22):
someone could fall into. Well,now we can use tools like
Einstein or the machine learningfrom HubSpot to identify intent
factors from their users historythat we wouldn't have access to
mixed with location data mixedwith previous activity within
the CRM to provide them a veryuniquely designed experience on
(21:44):
a page. And all of that can bedone with human and machine
hands, you know, collaboratingalong the way. You know, that's
one of the things that it's an,that's an interesting area for
me, as I see it coming togetheralmost in full fruition in some
of those technology platforms.
Because eventually, you know,HubSpot has a chatbot feature,
(22:05):
so does Salesforce andSalesforce marketing hub. So the
chats will be in there, theemail marketing will be in
there, you can connectadvertising platforms and
programs to those types of hubs.
So I see those as being acentral crux of how in the
future we're gonna use this,this universe of AI, if you
(22:27):
will.
Nels Jensen (22:29):
So let's, let's talk about the pros and the
cons.
Joey Strawn (22:33):
Yes, this is, there's a lot on both sides.
This is one of those where Ifeel like if I had been in a
debate team in high school, andthis was the conversation, you
would have to be able to argueboth sides, because there's
validity, I think everywhere.
Nels Jensen (22:50):
Yeah, so to me the most obvious from the benefits,
and I'm looking at it much morenarrowly than you are right, my
role is content, long formcontent. But it's definitely the
the research, and definitely,you know, time and speed and
efficiency. So research can bewhat I've found is the research
(23:18):
to me, tends to be the moremicro or macro you go, the
better, you know, it's not it'snot and that's not meant to be
general, it's meant to be getaway from as vague. At all
costs. Yeah. And then, but thetime and speed is, is it really
gets you started. And in somecases, it delivers great
(23:41):
efficiency, there are somethings, you know, if I'm like,
you know, what's the valueproposition for having XYZ for a
small manufacturer? And it'slike, okay, I know, I could sit
there, oh, yeah, you know,whatever. And it might take me
three minutes to write the twosentences. But this, you know,
might deliver something that'slike, Okay, I could use that.
(24:03):
And, but then that gets to thecons that we'll get to in a
minute. And about, should I usethat exactly there?
Joey Strawn (24:09):
Well, and that that, to me, is the immediate
benefit of all of this as thespeed and the amount that can be
analyzed at once. I mean, it'sjust there is no comparison to
AI and artificial intelligencecan do the work faster than we
could ever do it. So that to meis I think, is to us, the big
pro, and even in my world, likethe the social strategies or the
(24:33):
development code that can bewritten from Ai saves a lot of
time, you know, on an agencyside or you know, a time spent
on a salaried worker, the lesstime you spend doing doing
things you shouldn't have to themore time you can spend doing
things that are reallyprofitable to the company. If
you're, you know, an owner of acompany and you're looking at
your workforce. There are a lotof ways that are probably
(24:56):
appealing to say, well, I couldreally have this core team focus
on these profitable areas andhave ai do some of these menial
tasks and keep the gearsspinning, and I can use my human
capital, if you will, a lot moreeffectively, you know, I think
that is a huge Pro. But on theflip side of that is, to your
point nails on the con is? Well,should you use that, you know,
(25:19):
the is? Is it trustworthy? Wherewas it generated from the, you
know, images, output images, andit's a little bucket because the
that, like, there are artiststhat make art, and they put them
on the internet, and then itpulled together in a new thing.
And it's like, Well, where didthe all that come from? So
that's its own little world, butlike, with the information and
(25:41):
the accuracy, is it trustworthy?
Now, it's like, if you ask it aprompt, can you be sure that if
then you put your name on it andpost it, what it puts out there
isn't going to be ripped apartby the online? You know, truth
patrol, you know?
Nels Jensen (25:57):
Sure, sure,
Joey Strawn (25:58):
How much time should you have to spend to
validate what's in it, asopposed to the speed in which
the prompt was given? Yeah, youknow, that's a big question mark
of I don't know that answer. Butthat is kind of one of the cons
of, it's almost like whenWikipedia was the thing, you
know, was started to be a thing.
Colleges like, wow, that thingis super wrong. And Wikipedia is
way incorrect. And it's like,they're actually it's like, way
(26:22):
more correct. And so I don'tknow, it's out there.
Nels Jensen (26:26):
But there will be there will be a time when a
major brand sends out anautomated email that basically
misinterprets a term. Yep. Thatthat is vague. And there will be
the, you know, mini crisis PR,and all that it will happen. So
um, you know, I can't tell youwhen or where, but it will, so I
(26:48):
wouldn't, I would,
Joey Strawn (26:49):
I'm going to put my money I'm going to do like a
Babe Ruth point now is by thistime next year, you and I will
have a very soft, solid casestudy of a company who did that
within 2020. Yes. Or they sentout an email and mislike
misconstrued a social term thathas a slim meaning.
Nels Jensen (27:07):
Oh, 100 100%. So the trustworthiness to so you
when you talked about the theimage generators? So obviously,
you know, when we when theinternet really took off and
reached its tipping point, youknow, a lot of people talked
about the, oh, there's no morebarriers to publishing isn't
(27:28):
this great? And it's like, assomebody who basically grew up
in the editor, gatekeeper world,it's like, yeah, be careful what
you wish for. And maybe, maybeagree, sure enough, we've seen
lots of abuses of whether it'ssocial media, blogs, news, you
know, whole news channels, youknow, ever so I'm, I'm really
(27:50):
concerned about deep fakes, I'veseen a video that basically, you
know, is a deep fake of apolitician delivering a message
that they never delivered onthings that they would never
have said. And so it's pristine,it looks perfect. And you know,
(28:11):
so there's a whole, and that's,you know, that's less of a
concern in manufacturing andindustrial marketing, that it is
societal. But this at some pointdoes become related. If, if you
can't trust sources ofinformation, or platforms, then
(28:32):
that detracts from the effortsthat we're doing, you know, to
market for the manufacturers.
Joey Strawn (28:38):
And I'll put it in a really, you know, trust is
something that especially sincethe blooming of the internet,
that's become a very valuablecommodity. If people can trust
you, if they can feel like theycan believe what your company
puts forward, that gains thatgets you a lot of capital. So
(28:58):
with this kind of blow ofartificial intelligence, if
there is a layer of untrust, andjust the media in general, I'll
put it in a very real example.
Now, some of our, some of ourclients or some of the people
that are industries, work withaerospace, or work with defense
or work with airplanes, and haveto have very specific, very
pinpoint measurements andspecifications and realities to
(29:22):
their work. And if it's an, youknow, an automatically generated
or AI generated message that'snot true and tells an a layer of
falsity then those could havereal effects and real life
effects for companies that don'tget bids companies whose jobs
get refused on delivery of QA.
(29:46):
There's a lot that can go intobacking up the claims that the
speed in which you know the AIgenerated it sure correct was
given to you. Yeah, I you know,we've already talked about them
not being fine already. It'sVirtual Assistant world like
that, to me is a con is I don'tever want it to become a crutch
where people are just like threewords in a prompt, copy paste
(30:09):
pasted on a website. Like it'snot, it's never fine already.
He's got to have eyes on it.
Nels Jensen (30:14):
That's why my job is not in jeopardy in the near
Exactly. You know, I couldcertainly see where a small
content team might not havequite as many editors
gatekeepers AI strategists, andis relying on AI, especially as
it gets better, and it's gettingget a lot better faster, too.
That's the that's the otherthing. But yes, it's, um, you
(30:37):
know, so at a certain point, ifeverybody is using largely the
same sources, then isn't allthis going to be a certain
sameness? I mean, you know,that's, that's another downside,
that you, how do youdifferentiate your product or
service or your message? So, youknow, there's, there's still
(30:59):
always going to be value andcreativity, there's still always
going to be value and quality?
So, yes. Is AI going toeventually get better at that,
too? Yes, I mean, you can useeven Chet GPT. Now, write me an
800 word blog post on XYZ, andin two minutes, it's done. And
(31:19):
then you can say, make itfunnier, and it does. So I mean,
it's like, and I'm actuallylaughing. So it's, you know, I
realized this all isn't finalready. Now, it's going to
continue to get better. But, andwe'll talk a little bit more
about this coming on, but it'sreally super important to
(31:40):
understand how to use it. And,and where it needs to go into
that. It's not just, oh, let's,you know, let's get the
interface and go. It's like,what is it you really, really
want to ask it to do? Right?
Joey Strawn (31:58):
Yeah. And I think that's, that's, that's a good
point, too. And I think anotherpro that is good about this is
it's it's scalable, you know, itdoes become a layer of
assistance and a layer of helpto a team that, you know,
nowadays is mobile is spread outcan be small. So, you know, this
(32:20):
could be a good layer and helpsmaller teams become a lot more
scalable, and provide scalablepot, good, positive solutions to
their audiences and to their,their constituents. I mean, this
could be really, really good forsmall manufacturing teams, who
have just been struggling tocompete with the big boys for a
(32:41):
lot of years, because now youcan have some help, you can have
a layer of support andscalability that you didn't get
before. So there are ways forsmall industrials, small supply
chain manufacturers to reallytake advantage of the tools that
are out there. But you know, itreally is focusing, there's a
lot that's going to be poppingup. There's a lot of shiny
buzzword effects, fun thingsthat are going to be out there,
(33:04):
and really kind of focusing isgoing to be that I mean, that's
what we're gonna focus on oursecond section of the episode
Nelson, kind of on the shopfloor is really how we're
prioritizing.
Nels Jensen (33:14):
Yeah, well, I'm eager to hear how you're going
to use this. Yeah, I
Joey Strawn (33:19):
have asked you to put the how you're going to use
it together, which I'm eager tohear as well. The last thing I
want to touch on before we moveon this onto the shop floor is
just if, and I come in, I'm verycurious, on your opinion on
this, given your history injournalism, and as an editor and
having seen some of thesemachinations in the world, is,
(33:40):
what is what does plagiarismlook like in this sort of world?
Like if machines are justpulling from everything, and
it's all the same? Pooleventually, like what you said,
there's sameness that'sinevitable. There's that zoo for
something like this?
Nels Jensen (33:55):
Or maybe it is maybe it isn't? I don't know, I
don't know, even if I were toask, you know, a tool to do the
exact same thing. One day later,will I get the exact same
response? You know, that's aninteresting test showed up,
right? Should I is it or, youknow, yes, you can learn more.
So it should evolve. But yeah,and you know, plagiarism is, you
(34:20):
don't hear as much about it, orI don't anyway, in journalism
circles, because I think to someextent, there's an awful lot of
sort of link attribution anyway,so there's less incentive to
actually plagiarize. But wetalked about this offline in
terms of even academia whereit's like, okay, how do you how
(34:42):
do you make sure your studentactually wrote something as
opposed to just pull it off theinternet to begin with? So that
exists now? Will it be harder?
You know, I think it's there'sjust going to be a whole lot
less of authentic IQ, individualwork. I mean, it's just now it's
like keyword. There's so manytools you can use, what is the
(35:06):
keyword strategy for something,and there are tools already that
you can just go and type acouple of things in. And it
helps you do it. So, you know, Ithink we're as long as we're
viewing this tool sets, I don'tknow that plagiarism will become
more of an issue, maybe, maybeit will. But I just find it hard
to believe that a lot of peopleare going to fully automate
(35:28):
blogs, you know, right, or, Imean, it's, I think social
posts, you maybe will see a lotmore similarity in social posts,
maybe No, I don't know. What doyou think
Joey Strawn (35:41):
It's social as well as social kind of is all the
same already. It sort of soundsa lot the same. So that probably
is already happening. I I'minterested to see to see what
happens. I think I agree withyou. I don't I don't have much
more to add, than then what yousaid, I think it will become
kind of a not so much a stickierissue. But it'll be more on the
(36:03):
duplicate content side, likewhat is what is the search
engines? How did the searchengines differentiate between
content that's, you know,generated by the same types of
tools, you know, to generate thesame types of content? That'll
be that'll be an interesting,
Nels Jensen (36:18):
That's a that's a good point. Because that's,
that's already been in play.
Right? You know, exactly howdifferentiation and creativity,
those are always going to betenants of, of messaging, right?
Yeah. Well, okay. Let's,
Joey Strawn (36:34):
I'm excited to hear what you're prioritizing. We've
talked about a lot, there's toomany tools for one person to do
everything. So let's head ondown to the shop floor and else
and tell our listeners how we'reprioritizing things this year.
And maybe it can help them andget some direction on how they
want to prioritize. So let'shead on down to the shop floor.
Sounds good. All right. I don'twant to repeat a lot of what
(36:59):
we've said, this has been a veryin depth and good conversation,
I think. But you and I were bothlike, okay, there's so much
going on? How are we even goingto focus this. And so we each
came up with three ways thatwe're going to be focusing this
year. So I'll go through mythree, and then you can go
through your three, I'm excitedto hear what you've got. So I
mean, for me, we talked aboutit, my big one is just what can
(37:22):
it do in the areas of research,I am very interested to see how
good and helpful it can be inthe areas of like keyword
research and idea generation,especially for content
calendars, or keyword, you know,ranking opportunities. That is,
that's kind of the first thingthat I really want to dive into.
(37:42):
I'm a big fan of sem rushalready. And you know, there's a
lot of the AI tools and machinelearning that it's incorporated.
So just the research angle to meis super exciting. The next one,
the big area for me this year isgoing to be analytics and
feedback loops. You know, withGA four and us having to
transition and a lot of peopletransitioning to GA for that
(38:05):
system and platform this year,I'm really expected to see a lot
of good optimizations andmachine learning analysis to
come back from that. And that'skind of I'm excited to see what
it can do, whether it's, youknow, a big query integration
with GA four, or whether thereare some different sort of AI
(38:25):
software's that need to beincorporated into it? Or if it's
just a lot of combined apps thatwork together to output you
know, machine analytics. I'mjust curious what that what
those feedback loops could looklike. Because that conversation,
if it if speed is involved, andaccuracy is involved on the
data, that could really makethings interesting from a
(38:47):
marketing perspective. And thenlast for me is customer
interaction like engagementmetrics, like how is AI going to
be used in conversationalmarketing and chatbots? Like,
will all chatbots be mostly AIby the end of the year? Or will
still there be people, you know,typing in chats at a chat
(39:09):
center? You know, thepersonalization? How personal?
Can content on a website or inan email be and how relevant to
an audience and how perfectlytimed? Can it get those kind of
those three areas like how toresearch and generate the
directions, how to actuallyengage the customers and keep
(39:31):
them engaged? And then how toget feedback on all that
engagement like that. That is mykind of world of AI this year,
there's going to be a lot of funthings that happen in code
development and image generationand some others but I'm going to
let better qualified peoplefocus on
Nels Jensen (39:47):
You know, I, I love your list, and I especially love
feedback loops. And let me justgo on a short little tangent
here. We'll get to my list soonbut okay, the world of hiring It
is so broken, right? You hearabout these companies that do 10
rounds of interviews and, youknow, you hear about in
(40:07):
manufacturing, you know, we haverunners, the the usually brand
new people to the workforce whodon't even make it through their
first, you know, full day on thejob, the closing the feedback
loops on the recruiting, hiring,onboarding, retention, you just,
it's like you can, you canimprove your communication so
(40:29):
much, if you can close feedbackloops, you don't need to have
the third round of interviews,if you can adequately capture
what happened in the first tworounds of interviews, you know,
it's so to me, and I'm just sortof riffing off this, I hadn't
really thought about it. But AIshould help the recruitment
world, and the onboarding worldso so much, it's Oh, please,
(40:52):
please,
Joey Strawn (40:53):
I will add a layer to that, I will add a layer to
that, as well, as someone whohas been in the seat of hiring
and looking through resumes, Ihave seen how the quote unquote,
automated systems have worked inthe past. And it's a matter of
like, look for these sixkeywords in a resume. And if
those keywords aren't in thatresume, then put them in the Do
(41:15):
not contact folder and yeah, meas a hot that does those to me
eliminate so many goodcandidates, because it's not
learning. It's not. It's notintelligent. It's all just based
around arbitrary placeholdersand milestones. And that, to me,
is an area of just like yousaid, if there's a machine or if
(41:36):
there's an AI that canlegitimately scour through the
resumes and highlight the mostrelevant and helpful ones that
would one get people the rightjobs and help hiring directors
and decrease the amount ofinterviews people have to do it
with just that's a good,amazingly apt area for 2023
(41:58):
exploration for somebody. Allright, app developers get on
their AI. All right. All right.
Now, I want to know, what areyou focused on this year?
Nels Jensen (42:06):
So I what I will be exploring with AI in 2023. And
we've talked a little bit aboutit. So I my list of three, the
the first one is what I callframing and experimentation.
Right? So I'm basically lookingat it from a content strategy
perspective. So I'll beexperimenting with, you know,
(42:26):
because the better set ofinstructions you give, the
better result you get. So I'llbe trying to learn how to frame
the ask to get a better outcome.
And then the the depth part ofthat will be the applications in
the execution, right? Can youreally ask it for this sight
type of task? Or how good is it,you know, when you, you know,
(42:48):
are doing completely different
Joey Strawn (42:50):
Or, or maybe it's just sort of for someone who
tasks. So to me, that's the,that's sort of the big picture.
We've talked about research tome, I'm much more interested in
the micro research, I, I see alot of the macro, to me is not
nearly as helpful, at least inthe things that I've tried and
the things I do. So again, it'sit's really more in depth
(43:14):
research. But I think there'shuge potential from a content
perspective on that. And thethird one is saving time. And
that, for example, where ithelps at a macro is like writing
a summary of something. So itcould be as something as simple
as, you know, okay, I've come tothe point where I need to, you
(43:35):
know, talk about what are whatare the benefits of, you know,
adding 3d printing or additivemanufacturing, you know, just I
need a quick little summaryhere, whatever, that's been
written a million times, right.
So I could go pull somethingdirectly from the web, but it's
actually easier to put it in thetool and frame it a little bit
(43:57):
and get something that I want.
Similar for a list of benefitsor a list of hurdles or
challenges. It's like, thelittle the little list of goals,
if you will, it can be very goodat generating those. The big
mystery for me is the idea of ofnot using it as a research
(44:18):
assistant, but using it for arough draft. So a first draft,
and that's where I think I'll bespending considerable time
including next week, you know,looking at okay, so how
effective you know, can this beisn't, you know, if you're a
really slow writer, it mightsave more time. People, a lot of
(44:39):
people find it easier to editwhat is in front of them instead
of generate something that isn'tthere. Right.
needs a kickstart maybe givingseven prompts around a single
idea is what they need. It saves30 minutes of staring at a blank
screen. And didn't come up withthose seven prompts on their
(45:02):
own. One of the things that yousaid that I also want to loop
back to because I think it'sgoing to be interesting to see
where it goes in the future isthe framing. That to me is going
to be very interesting. Iremember when colleges started
to teach, like the differentBoolean search codes and the
(45:22):
different ways to search Google,I was like, Well remember, if
you put quotations around it, itwill search the exact phrase.
And if you put site colonquotations around a thing, it
will search just that web,there's all these specific ways
you can search. In Google, or insearch engines. We're gonna have
a whole new lexicon and, anddictionary of ways to frame
(45:46):
questions to AI, that pump outthe best answers, and we're
gonna have to as a society ofhumans, be trained in how to do
that. It's almost gonna be likelearning how to communicate with
the robots, but like, ya know,to your point analysis, like in
academia, I, it would notsurprise me in, you know, three
or four years that a part of amarketing class would be, hey,
(46:08):
how do you frame a search withinan AI platform to get the
correct content generated or toget the correct sources
referenced or whatever it maybe? That to me will be an
interesting new area ofexploration in the future.
Nels Jensen (46:26):
Yep. So yeah, it's it's a, it is a new world, even
though it's not totally newworld. But right. There's always
like one, yep, yep.
Joey Strawn (46:38):
Feels Well, it's a new year, a new you a new
Nell's, a new AI. It's a newuniverse of all of this
artificial intelligence. I mean,like, like you said at the
beginning, and I do want toreiterate, we we're not
developers on any of these,we're not experts on these
tools. A lot of this, we'reexploring at the same rate that
everybody else is. And, youknow, this is something that we
(47:00):
know is going to make adifference in all of our realms.
And so if you have tools thatyou've been using, or if you
have things that we didn't talkabout on this episode that you
think are worth exploring, letus know. You know, email us at
podcast and industrialmarketer.com. Follow us on
social media or comments on theblog posts on our website,
industrial marketer.com. Forthis episode, let us know what
(47:23):
you're exploring how you'refinding useful industrial
marketing uses for artificialintelligence, or even some of
the problems that you've runinto. A lot of this year, we're
going to be diving into thetechnology and the tools that
surround us as marketers and sowe want to know those hurdles
that you're running into. Followthe show, share with your
(47:44):
friends, let's pull together andand ask us those questions.
Because those are the thingsthat we want to be answering and
talking about this season on theshow. So again, if you haven't
subscribed to the podcast, whatare you doing? Come on, you know
you want to and email us yourquestions and your and your
comments at podcasts andindustrial market. or.com
Nelson, are you excited? Are youpumped about what the machines
(48:05):
are going to do for us?
Nels Jensen (48:07):
I am putting the machines to work hopefully I
will have let it outcome.
Joey Strawn (48:12):
Lets put them to work and hope that it's not like
a terminator output. But forright now. We're good. They're
going to be our great virtualassistants. And we're going to
put them to work and let's seewhat great marketing we can make
in 2023
Welcome back, everybody to another episode and
another season of the industrialMarketer Podcast your place for
the tips, tech trends andtactics for industrials who care
about driving leads and revenueto their businesses. We are
back, it is another season andanother year, I am one of your
hosts, Jo E. AI, and I am herewith some artificial
(00:24):
intelligence. And we are goingto talk about some fun
technology. Oh Nels. I'm soexcited to be back. And we
decided this quarter, we'regoing to talk about tools in the
tech of marketing. And whatbetter place to start than the
buzziest buzz topic that ishappening right now, which is AI
(00:45):
Artificial Intelligence and thetools that surround it. I mean,
are you excited to dive intothis topic? Are you excited to
be back one for season threeNels?
Nels Jensen (00:55):
Oh, yes. And just to let you know that the title
for this episode should be aImarketing strategies for
manufacturers. Perfect. That'swhat that's what AI tells us.
Joey Strawn (01:07):
We will listen, we are now answering to not the man
but the machine. And so this isgonna be great. Now
unfortunately, for all youlisteners, Nelson, I are still
humans. This is still a humanled podcast, we haven't given it
all over to AI just yet. But weare going to dive in. I mean,
again, this is a topic thateveryone from in the trenches,
(01:30):
to hopefully CEOs and leadersand drivers of their businesses
are thinking about these tools,not so much, you know, the
Minority Report or iRobot of itall, that's not really the the
artificial intelligence thatwe're going to be diving into.
But you know, what tools are outthere that are usable? What
(01:51):
investments should be thoughtof, or structural
infrastructure, ideas andconcepts should be considered
for your organization? Those arethe types of things that we
really want to dive into on?
Where do these fit in? What dothey realistically look like?
And what does it mean for peoplein our industries, which is more
like the industrialmanufacturers of the world? The
supply chains of the world? TheE commerce businesses of the
(02:14):
world? Like what, what is AIgoing to do for us?
Nels Jensen (02:19):
Yeah, and I think, you know, really, this falls
into a larger b2b marketingcategory, like a lot of ours do.
It's not, you know, hugely, it'snot hugely different that we'll
talk about nuances andsubtleties. But let's also be
transparent here, Joey, we arenot experts on the AI
applications here. This is verymuch new to us, as well as new
(02:39):
to, to the people out there whoare listening. We have been
playing around a lot with theseexperimenting, we've used some
of these tools. But we'relearning it too. And we're
learning the power and the theguardrails and the you know,
limitations and all that. Sowe're right, jumping in there
with everybody else. And I wouldsay we are far from the
(03:02):
authoritative sources on whatexactly you should use. So I
think, maybe around one cornerahead of a lot of us, right.
Joey Strawn (03:10):
I think that's a very important point, Nelson, I
do want to back that up and say,Yeah, we're not experts, and
we're not authorized resellersor re dealers have any of these
tools, either anything wementioned in this episode are
things that we've either playedwith, read about, or have heard
about, or have been told toexplore. So these are tools that
we know are in this universe,these are things that we know,
(03:32):
are going to pop up inconversation this year, whether
it is you know, at a meetup, orat an industrial group or
online, or maybe your boss comesto you and says, Hey, what about
this chat GPT thing? You shouldhave some sort of answers and
know what's on the horizon. Andthat's really what we want to
focus on today. I mean, knowswe're not even, we're gonna
mention a lot of tools, I thinkthroughout the conversation, but
(03:54):
really, the heart of theconversation is like, what does
it mean, for us, you know,manufacturers and marketers and
industrials. You know, how doesit realistically going to be
implemented? I mean, like, let'sbe clear. We're not saying that
by the end of this year,everyone's gonna be wearing
virtual reality headsets, and,you know, the haptic hands and,
you know, controlling things inan artificial realm. That's what
(04:17):
realistically, can we thinkabout this year, and then we
just want to go through a lot ofthe types, there's a lot of pros
and cons, and a lot of ways thistechnology is being used. So we
just want to focus on some ofthat. I mean, I'm getting I'm
excited. This is something thatyou and I've been reading a lot
about Nels. So I know you, youand I have each played with
different tools and systems. Andso I'm excited to see what
(04:39):
you've learned and figured outand I'm excited to bring to the
table what I've got, I mean, butjust kind of to start when you
think of artificial intelligenceand these AI tools and this big
trend that's been happening,it's being talked about right
now. What do you feel itactually means for for
industrial marketing, and peopledoing what we do?
Nels Jensen (05:01):
Well, I guess I have a hard time with the big,
grand umbrella of it all. Ithink it's if you if you are in
the industrial sector, youunderstand predictive
maintenance, right? Youunderstand how equipment has
patterns and telltale signs. Andthere are definitely things that
(05:26):
it tells you are going tohappen. And the key thing there
is closer to a timeline for whenthey might happen. And I think
artificial intelligence islargely prescriptive in that,
yes, it's, it's telling you it'sroadmap, it's telling you where
to go. There's also you get intosome of the different
(05:49):
applications about actuallydoing the work for you. But to
me, that's the more cautiousarea to go to me, I view it as
more powerful for a roadmap andguideposts than I do for the
actual execution.
Joey Strawn (06:07):
Yeah, and I think that's an interesting point to
make, too. And, you know, thisis a fancy term, and a lot of
people are talking about thisnow. But even before we started
recording now, so you and I weretalking, saying, well, these are
stuff that we've been using fora while, like artificial
intelligence is the way thatwe're sort of putting it all
under the same umbrella. Butmachine learning has been a
(06:28):
topic that we've been talkingabout for a while it's in a lot
of the tools that we use, likethe SEM rushes and the male
chimps of the world, you know,that's looking for machine
learning. automations, you wetalked about big data, you
remember, like a handful ofyears ago, when big data was the
buzz buzzword? Well, the naturalnext step in the gathering of
(06:48):
big data is that, well, we haveto have computers to compute the
massive big data, and then tellus what that means. And that's
what intelligence is, that'sartificial, right.
Nels Jensen (06:58):
And that's what history of that industry. 4.0 is
basically putting data to workfor insights. And, you know, and
that, and I think the reason whythis is such a big topic right
now is just we've reached atipping point where it's yes,
it's not just for the data nerdsout there, we've now reached the
(07:19):
point where the usability issuch that almost anybody can do
it. And, you know, we've come tothe point where we have to
figure out more about trainingand more about, you know,
applications. And we'll talkwe'll talk more about that, you
know, coming up, but, but wehave definitely reached a
tipping point, in terms of AIwhere it's, it's just not, if
(07:44):
you're just thinking, well,that's for the data scientists
and the, you know, analyticspeople, you know, no, it's we've
reached that point where it'sgoing to be for all of us. And,
you know, if you're, if you'vebeen lagging, and you know, yes,
now, it'd be a good time to getup to speed and begin to stick
(08:06):
your toe in the water and figurethis out.
Joey Strawn (08:08):
Well, and, and that's a great segue to because
one of the things that, youknow, we want to definitely talk
about today is the ways that wecan realistically see this
happening within the nearfuture. So I mean, again, you
don't have to dive in and becomethe biggest data scientist in
the world. But now it's to yourpoint, we've reached the tipping
point where it's going to becomenot only commonplace, but sort
(08:31):
of expected, especially with theproliferation of, you know,
tools, like I mentioned, jetchat GPT earlier. You know, last
year, there was a big socialmedia meme explosion around AI
image generators, like midjourney, and Dali, and some of
those, and, you know, it was anentertainment and meme driver a
(08:54):
lot of last year, but to yourpoint, we're getting in now
where it's going to be a lotmore economical, it's going to
be a lot more systematic and alot more scalable. So a lot of
people will have access to thetools. So sure, sure, chat bots
are a normal thing that I see. Iwould bet by the end of this
year, a lot of chat bots are runby some AI mechanism.
Nels Jensen (09:15):
Oh, and we've seen that we've all seen that on on
websites, too, as the, you know,the volume and importance on
online research takes hold andin the industrial sector
chatbots are I now see them moreoften than I don't exactly that
people people might not think ofthat as artificial intelligence.
(09:38):
But it is a very simple even ifit's just sorting you into one
of four or five differentcategories for you know, some
type of connection somewherethat is still, you know,
artificial intelligence at usright now,
Joey Strawn (09:53):
And especially with tools, you know, we're getting
better about having access tosystems you know, Bambara is a
Big tool out there, but intentdata, you know, the more access
to intent data and historicalrelevance of activities, the
more what do we mentionedearlier, those big data patterns
can be recognized by machinesthat are trained to do it. So
(10:15):
all of these pieces have been inplay for a handful of years. And
I mean, in real in an honestyand in all real NIS advertising
has been doing this for years.
Programmatic advertising isbased around the art, like
artificial intelligence andmachine learning of large
amounts of advertising and bigdata. You know, this is
something that has been in usefor a lot of marketers for a
(10:37):
while whether or not we'veactually thought of it as
artificial intelligence.
Nels Jensen (10:45):
Yes, and, you know, if you are putting your even
lead scoring into a CRM, it isfeeding into an engine that, you
know, HubSpot AI, or you know,some of these other ones as
well. And I think the one theone thing that to me, we're kind
of jumping all over the place,but that's okay, the analytics
(11:07):
side of it. So GA four, we didan episode recently on GA four.
And that's one of the coolpromises of the latest platform
for Google is that they'rebasically going to help you
determine what campaigns have ahigher level of success
predicting audiences, basicallyhelping you, you know, just give
(11:30):
you a much clearer course ofaction to take, you know, for a
campaign, you know.
Joey Strawn (11:37):
That's one of the major tenants of GA for is one
of the reasons the platform isevolving, the way that it is, is
stepping away from individualproperties and managing those
but looking at the entire matrixof everything going on, and then
get gathering learnings fromthose activities in those trends
across the spectrum. And somachine learning has been
(11:58):
something that they're veryheavily pushing, and I'm happy
that you mentioned analytics,because nails this is really,
really where the rubber meetsthe road. And I think where we
should spend kind of up a goodmeat of our conversation today
is what what is this going tolook like in 2023? Like, there
are a lot of different aspectsof our jobs that we you
(12:19):
mentioned analytics. I'vementioned content creation, to
some degree with chat GPT, orchat bots and things like that.
But there are tons of uses forthese, like, you know, just to
go through some of the typesthat you and I mentioned, when
we were planning this is contentgenerators, image generators,
advertising, analytics, CRM, andautomations, devel development
(12:44):
and code generation, a lot ofthese things already exist in
already are out there. I mean,just we can start at the top on
this one, like contentgeneration, what do you Nelson
imagine the future looking likewith these AI tools? And content
generation? Or like, are yougoing to lose your job? Are we
going to have to like, replaceyou with a robot like a Nelson
bot point?
Nels Jensen (13:05):
Oh, yeah. The not yet. And I think there's a
general, there's, there's thebasic, mostly right. View of
content generating AI? And, youknow, it depends, it depends,
obviously, what is the pool ofinformation that this AI is
(13:27):
working from? So you have thethe new, the crazy tipping point
tool is the chat GPT that is...
Joey Strawn (13:35):
Schools are banning, and then everyone's
getting up, and I was like, Oh,it's right, in the presidential
speeches or whatever.
Nels Jensen (13:41):
Right and it's, uh, but it is, if you've never
played with it, do so Irealized, if you jump on it a
lot of times right now and youdon't already have an account,
you're not going to get oneright away. It's really cool.
It's, but it's scraping the webto write these things for you.
You give it a prompt, and we cantalk a little bit more about,
(14:01):
you know, the importance andunderstanding what exactly
you're asking it to do. Becausethat becomes, you know, if
nothing else, I'm not going tolose my job because, you know,
how do you interpret exactlywhat you want? Exactly, putting
that into an ask? So but that'sdifferent than like, Jasper,
which is actually made by thesame people open AI, the same
(14:23):
company. You know, we're, we'reJasper will take for instance, I
can provide, here are threedocuments that I want you to use
to draft and then you ask forwhatever 1000 word blog post,
and you give it obviouslyinstructions, but Jasper is
basically allowing you toprovide the inputs, you know,
(14:45):
chat GPT is not it's basicallyand I'm sure at some point it
will. I mean, this is the peoplerunning this open AI the
company, you know, are verysmart. They basically put chat
GP He out there. And they'veprobably advanced their
development work because ofeverybody playing with it by
(15:07):
yours in a matter of two weeks.
Oh, yeah. Or I don't know howlong it's been out there. But
it's. But yes. So there'sthere's fundamental differences
in terms of what are what arethese tools using to generate
content? And that's superimportant to understand. And
then, yeah, and then
Joey Strawn (15:25):
I think when we compare pros and cons later,
that'll be Yeah, that'll besomething to dive into,
especially with the imagerystuff, because that's got to
come from somewhere. But like,even things like content bot,
you know, I think is a fun one.
One of the things that you and Ihave talked about I, this is the
joke that I like to say rightnow. But I really view
especially the content AIgenerators as like a really
(15:46):
great virtual assistant. Youknow, it's like, give them a
prompt, and they're smart, andthey can go anywhere on the
internet and get everything andput you together something.
Yeah, but it's going to have tobe edited. It's going to have to
be reviewed for accuracy andmake sure that the sources are
verified. There's work to bedone. It's not a final product,
but it's a great assistant.
Nels Jensen (16:08):
A shout out to our peer Virginia Roberson who she
calls it her, you know, her topnotch research assistant. Yes,
exactly. And we'll talk a littlebit more to as well research is
certainly a huge use for it.
There are others as well. Butyes, she's, you're both correct,
right there, virtual assistantresearch assistant. You know, it
(16:29):
is very, these are very, verypowerful tools. And I'm, I've
gone from the oh, well, we'lljust wait and see how this plays
out, too. All right, it's playedout enough now that I need to
get a better handle on it.
Joey Strawn (16:45):
One of the first things that I remember seeing in
the market, or at leastconversations about this, where
the image stuff like mid journeywas one of the first ones I
remember our developer, we'vehad him on the show before Brian
shared a bunch of Dali stuffearly on, like last year. And I
was like, Oh, wow, this ispretty crazy. And then you dive
into like, where's this? Like?
How is this getting? You know,photo, I think is another one.
(17:08):
So there's a lot of the imagestuff that I don't really
understand. But that's where Ikind of first saw it. And that's
gonna be really interesting inmy mind to watch on, like, what
can those images be used inAdWords? What's the copyright on
these like, because images aredifferent than the semantics of
words and how words are puttogether? Because those images
have to be sourced, the pixelshave to be sourced. So it's,
(17:31):
that's an interesting, what'skind of one of the first ones I
ever saw. But getting deeperinto it, like you mentioned,
analytics. And I've mentionedadvertising, like those are two
very in the weeds element ofsomething that we do as a
marketing team, that now AI iskind of dipping their toe into,
I mean, outside of you know,like the ad roles and the
(17:53):
programmatic advertising. Youhave sem rushes and GA fours and
machine learning analyticsplatforms that are giving pretty
good advice on how to updatepages, how to look at
optimizations, and where totarget advertising. So it's
Nels Jensen (18:12):
Yeah, one of the beauties, one of the beauties is
how they link together too. Soprogrammatic tells us, here's,
for instance, you know, avenuesto track users, and where, show
them this message where they go.
Right, right, and then you know,you're using some of these
different channels to do that.
(18:33):
And what you might might findout is that you see analytics
that oh, yeah, there's a hugenumber of people who are seeing
this. And there's very littlestickiness, right, so you're
learning two things that yes,this is it does deliver
audience, it does not delivervalue, you know, I mean, it's
(18:54):
the the, that's the power of theanalytics is isn't just telling
you, Hey, do this and saying,Hey, do this. And oh, by the
way, that wasn't the greatestthing. You know, it'll
ultimately help us, as I like tosay, play offense and defense at
the same time, right, doubledown on what works and cut back
on what doesn't, right.
Joey Strawn (19:12):
It is, it gives us an additional reach as marketers
to get our hands around the bigdata. You know, it's the data
source that we talked aboutearlier. And that's really what
a lot of this comes back to ishaving the history of metrics,
having the access to InternetTrends and culture and the ways
that we use words and images ispulling all of that together,
(19:34):
where we wouldn't have the timeand energy to do so.
Nels Jensen (19:37):
Right. So. So with automations, right. So, you
know, when I first heard aboutthis, you know, I'm much newer
to marketing than you are, butwhen it was like sort of this
personalization is going to beautomated. I'm just like, Well,
that'll be interesting. Andit's, you know, and many in many
(19:57):
cases, it wasn't perinternalized it was targeted.
Right? But it has been it hasevolved to be personal where you
actually can have theattribution sources so that they
can, you know, yeah, deliversomething 100% unique to me
versus you even though we havemany of the same behaviors,
(20:18):
right?
Joey Strawn (20:18):
Well, and you mentioned within automations,
I'll expand that to include likeCRM platforms and marketing
automation tools, too, becausethat's one area that we know
have been playing around withartificial intelligence for a
while. I mean, HubSpot ingeneral has been talking about
their machine learning, theyhave a lot of big architecture
that they're working on, aroundmachine learning to provide
(20:40):
insights on everything that'shappening within their tools.
But the big one right now, Ithink that a lot of people hear
about and hear talked about isSalesforce, Einstein, Salesforce
launched their Einstein platforma little while ago, but any, a
lot of people probably heard itand know what the name is, they
may even use it. But it's basedon the machine learning, it's
(21:00):
based on machine learning andinsights and trying to help make
a lot of those things that usedto be very manual within
automation tools, a lot moremachine focused, or a machine
driven. So whereas, you know, ahandful of years ago, we as a
marketer, you would have to gointo HubSpot, or go into
Salesforce and make filters andrule for every situation that
(21:22):
someone could fall into. Well,now we can use tools like
Einstein or the machine learningfrom HubSpot to identify intent
factors from their users historythat we wouldn't have access to
mixed with location data mixedwith previous activity within
the CRM to provide them a veryuniquely designed experience on
(21:44):
a page. And all of that can bedone with human and machine
hands, you know, collaboratingalong the way. You know, that's
one of the things that it's an,that's an interesting area for
me, as I see it coming togetheralmost in full fruition in some
of those technology platforms.
Because eventually, you know,HubSpot has a chatbot feature,
(22:05):
so does Salesforce andSalesforce marketing hub. So the
chats will be in there, theemail marketing will be in
there, you can connectadvertising platforms and
programs to those types of hubs.
So I see those as being acentral crux of how in the
future we're gonna use this,this universe of AI, if you
(22:27):
will.
Nels Jensen (22:29):
So let's, let's talk about the pros and the
cons.
Joey Strawn (22:33):
Yes, this is, there's a lot on both sides.
This is one of those where Ifeel like if I had been in a
debate team in high school, andthis was the conversation, you
would have to be able to argueboth sides, because there's
validity, I think everywhere.
Nels Jensen (22:50):
Yeah, so to me the most obvious from the benefits,
and I'm looking at it much morenarrowly than you are right, my
role is content, long formcontent. But it's definitely the
the research, and definitely,you know, time and speed and
efficiency. So research can bewhat I've found is the research
(23:18):
to me, tends to be the moremicro or macro you go, the
better, you know, it's not it'snot and that's not meant to be
general, it's meant to be getaway from as vague. At all
costs. Yeah. And then, but thetime and speed is, is it really
gets you started. And in somecases, it delivers great
(23:41):
efficiency, there are somethings, you know, if I'm like,
you know, what's the valueproposition for having XYZ for a
small manufacturer? And it'slike, okay, I know, I could sit
there, oh, yeah, you know,whatever. And it might take me
three minutes to write the twosentences. But this, you know,
might deliver something that'slike, Okay, I could use that.
(24:03):
And, but then that gets to thecons that we'll get to in a
minute. And about, should I usethat exactly there?
Joey Strawn (24:09):
Well, and that that, to me, is the immediate
benefit of all of this as thespeed and the amount that can be
analyzed at once. I mean, it'sjust there is no comparison to
AI and artificial intelligencecan do the work faster than we
could ever do it. So that to meis I think, is to us, the big
pro, and even in my world, likethe the social strategies or the
(24:33):
development code that can bewritten from Ai saves a lot of
time, you know, on an agencyside or you know, a time spent
on a salaried worker, the lesstime you spend doing doing
things you shouldn't have to themore time you can spend doing
things that are reallyprofitable to the company. If
you're, you know, an owner of acompany and you're looking at
your workforce. There are a lotof ways that are probably
(24:56):
appealing to say, well, I couldreally have this core team focus
on these profitable areas andhave ai do some of these menial
tasks and keep the gearsspinning, and I can use my human
capital, if you will, a lot moreeffectively, you know, I think
that is a huge Pro. But on theflip side of that is, to your
point nails on the con is? Well,should you use that, you know,
(25:19):
the is? Is it trustworthy? Wherewas it generated from the, you
know, images, output images, andit's a little bucket because the
that, like, there are artiststhat make art, and they put them
on the internet, and then itpulled together in a new thing.
And it's like, Well, where didthe all that come from? So
that's its own little world, butlike, with the information and
(25:41):
the accuracy, is it trustworthy?
Now, it's like, if you ask it aprompt, can you be sure that if
then you put your name on it andpost it, what it puts out there
isn't going to be ripped apartby the online? You know, truth
patrol, you know?
Nels Jensen (25:57):
Sure, sure,
Joey Strawn (25:58):
How much time should you have to spend to
validate what's in it, asopposed to the speed in which
the prompt was given? Yeah, youknow, that's a big question mark
of I don't know that answer. Butthat is kind of one of the cons
of, it's almost like whenWikipedia was the thing, you
know, was started to be a thing.
Colleges like, wow, that thingis super wrong. And Wikipedia is
way incorrect. And it's like,they're actually it's like, way
(26:22):
more correct. And so I don'tknow, it's out there.
Nels Jensen (26:26):
But there will be there will be a time when a
major brand sends out anautomated email that basically
misinterprets a term. Yep. Thatthat is vague. And there will be
the, you know, mini crisis PR,and all that it will happen. So
um, you know, I can't tell youwhen or where, but it will, so I
(26:48):
wouldn't, I would,
Joey Strawn (26:49):
I'm going to put my money I'm going to do like a
Babe Ruth point now is by thistime next year, you and I will
have a very soft, solid casestudy of a company who did that
within 2020. Yes. Or they sentout an email and mislike
misconstrued a social term thathas a slim meaning.
Nels Jensen (27:07):
Oh, 100 100%. So the trustworthiness to so you
when you talked about the theimage generators? So obviously,
you know, when we when theinternet really took off and
reached its tipping point, youknow, a lot of people talked
about the, oh, there's no morebarriers to publishing isn't
(27:28):
this great? And it's like, assomebody who basically grew up
in the editor, gatekeeper world,it's like, yeah, be careful what
you wish for. And maybe, maybeagree, sure enough, we've seen
lots of abuses of whether it'ssocial media, blogs, news, you
know, whole news channels, youknow, ever so I'm, I'm really
(27:50):
concerned about deep fakes, I'veseen a video that basically, you
know, is a deep fake of apolitician delivering a message
that they never delivered onthings that they would never
have said. And so it's pristine,it looks perfect. And you know,
(28:11):
so there's a whole, and that's,you know, that's less of a
concern in manufacturing andindustrial marketing, that it is
societal. But this at some pointdoes become related. If, if you
can't trust sources ofinformation, or platforms, then
(28:32):
that detracts from the effortsthat we're doing, you know, to
market for the manufacturers.
Joey Strawn (28:38):
And I'll put it in a really, you know, trust is
something that especially sincethe blooming of the internet,
that's become a very valuablecommodity. If people can trust
you, if they can feel like theycan believe what your company
puts forward, that gains thatgets you a lot of capital. So
(28:58):
with this kind of blow ofartificial intelligence, if
there is a layer of untrust, andjust the media in general, I'll
put it in a very real example.
Now, some of our, some of ourclients or some of the people
that are industries, work withaerospace, or work with defense
or work with airplanes, and haveto have very specific, very
pinpoint measurements andspecifications and realities to
(29:22):
their work. And if it's an, youknow, an automatically generated
or AI generated message that'snot true and tells an a layer of
falsity then those could havereal effects and real life
effects for companies that don'tget bids companies whose jobs
get refused on delivery of QA.
(29:46):
There's a lot that can go intobacking up the claims that the
speed in which you know the AIgenerated it sure correct was
given to you. Yeah, I you know,we've already talked about them
not being fine already. It'sVirtual Assistant world like
that, to me is a con is I don'tever want it to become a crutch
where people are just like threewords in a prompt, copy paste
(30:09):
pasted on a website. Like it'snot, it's never fine already.
He's got to have eyes on it.
Nels Jensen (30:14):
That's why my job is not in jeopardy in the near
Exactly. You know, I couldcertainly see where a small
content team might not havequite as many editors
gatekeepers AI strategists, andis relying on AI, especially as
it gets better, and it's gettingget a lot better faster, too.
That's the that's the otherthing. But yes, it's, um, you
(30:37):
know, so at a certain point, ifeverybody is using largely the
same sources, then isn't allthis going to be a certain
sameness? I mean, you know,that's, that's another downside,
that you, how do youdifferentiate your product or
service or your message? So, youknow, there's, there's still
(30:59):
always going to be value andcreativity, there's still always
going to be value and quality?
So, yes. Is AI going toeventually get better at that,
too? Yes, I mean, you can useeven Chet GPT. Now, write me an
800 word blog post on XYZ, andin two minutes, it's done. And
(31:19):
then you can say, make itfunnier, and it does. So I mean,
it's like, and I'm actuallylaughing. So it's, you know, I
realized this all isn't finalready. Now, it's going to
continue to get better. But, andwe'll talk a little bit more
about this coming on, but it'sreally super important to
(31:40):
understand how to use it. And,and where it needs to go into
that. It's not just, oh, let's,you know, let's get the
interface and go. It's like,what is it you really, really
want to ask it to do? Right?
Joey Strawn (31:58):
Yeah. And I think that's, that's, that's a good
point, too. And I think anotherpro that is good about this is
it's it's scalable, you know, itdoes become a layer of
assistance and a layer of helpto a team that, you know,
nowadays is mobile is spread outcan be small. So, you know, this
(32:20):
could be a good layer and helpsmaller teams become a lot more
scalable, and provide scalablepot, good, positive solutions to
their audiences and to their,their constituents. I mean, this
could be really, really good forsmall manufacturing teams, who
have just been struggling tocompete with the big boys for a
(32:41):
lot of years, because now youcan have some help, you can have
a layer of support andscalability that you didn't get
before. So there are ways forsmall industrials, small supply
chain manufacturers to reallytake advantage of the tools that
are out there. But you know, itreally is focusing, there's a
lot that's going to be poppingup. There's a lot of shiny
buzzword effects, fun thingsthat are going to be out there,
(33:04):
and really kind of focusing isgoing to be that I mean, that's
what we're gonna focus on oursecond section of the episode
Nelson, kind of on the shopfloor is really how we're
prioritizing.
Nels Jensen (33:14):
Yeah, well, I'm eager to hear how you're going
to use this. Yeah, I
Joey Strawn (33:19):
have asked you to put the how you're going to use
it together, which I'm eager tohear as well. The last thing I
want to touch on before we moveon this onto the shop floor is
just if, and I come in, I'm verycurious, on your opinion on
this, given your history injournalism, and as an editor and
having seen some of thesemachinations in the world, is,
(33:40):
what is what does plagiarismlook like in this sort of world?
Like if machines are justpulling from everything, and
it's all the same? Pooleventually, like what you said,
there's sameness that'sinevitable. There's that zoo for
something like this?
Nels Jensen (33:55):
Or maybe it is maybe it isn't? I don't know, I
don't know, even if I were toask, you know, a tool to do the
exact same thing. One day later,will I get the exact same
response? You know, that's aninteresting test showed up,
right? Should I is it or, youknow, yes, you can learn more.
So it should evolve. But yeah,and you know, plagiarism is, you
(34:20):
don't hear as much about it, orI don't anyway, in journalism
circles, because I think to someextent, there's an awful lot of
sort of link attribution anyway,so there's less incentive to
actually plagiarize. But wetalked about this offline in
terms of even academia whereit's like, okay, how do you how
(34:42):
do you make sure your studentactually wrote something as
opposed to just pull it off theinternet to begin with? So that
exists now? Will it be harder?
You know, I think it's there'sjust going to be a whole lot
less of authentic IQ, individualwork. I mean, it's just now it's
like keyword. There's so manytools you can use, what is the
(35:06):
keyword strategy for something,and there are tools already that
you can just go and type acouple of things in. And it
helps you do it. So, you know, Ithink we're as long as we're
viewing this tool sets, I don'tknow that plagiarism will become
more of an issue, maybe, maybeit will. But I just find it hard
to believe that a lot of peopleare going to fully automate
(35:28):
blogs, you know, right, or, Imean, it's, I think social
posts, you maybe will see a lotmore similarity in social posts,
maybe No, I don't know. What doyou think
Joey Strawn (35:41):
It's social as well as social kind of is all the
same already. It sort of soundsa lot the same. So that probably
is already happening. I I'minterested to see to see what
happens. I think I agree withyou. I don't I don't have much
more to add, than then what yousaid, I think it will become
kind of a not so much a stickierissue. But it'll be more on the
(36:03):
duplicate content side, likewhat is what is the search
engines? How did the searchengines differentiate between
content that's, you know,generated by the same types of
tools, you know, to generate thesame types of content? That'll
be that'll be an interesting,
Nels Jensen (36:18):
That's a that's a good point. Because that's,
that's already been in play.
Right? You know, exactly howdifferentiation and creativity,
those are always going to betenants of, of messaging, right?
Yeah. Well, okay. Let's,
Joey Strawn (36:34):
I'm excited to hear what you're prioritizing. We've
talked about a lot, there's toomany tools for one person to do
everything. So let's head ondown to the shop floor and else
and tell our listeners how we'reprioritizing things this year.
And maybe it can help them andget some direction on how they
want to prioritize. So let'shead on down to the shop floor.
Sounds good. All right. I don'twant to repeat a lot of what
(36:59):
we've said, this has been a veryin depth and good conversation,
I think. But you and I were bothlike, okay, there's so much
going on? How are we even goingto focus this. And so we each
came up with three ways thatwe're going to be focusing this
year. So I'll go through mythree, and then you can go
through your three, I'm excitedto hear what you've got. So I
mean, for me, we talked aboutit, my big one is just what can
(37:22):
it do in the areas of research,I am very interested to see how
good and helpful it can be inthe areas of like keyword
research and idea generation,especially for content
calendars, or keyword, you know,ranking opportunities. That is,
that's kind of the first thingthat I really want to dive into.
(37:42):
I'm a big fan of sem rushalready. And you know, there's a
lot of the AI tools and machinelearning that it's incorporated.
So just the research angle to meis super exciting. The next one,
the big area for me this year isgoing to be analytics and
feedback loops. You know, withGA four and us having to
transition and a lot of peopletransitioning to GA for that
(38:05):
system and platform this year,I'm really expected to see a lot
of good optimizations andmachine learning analysis to
come back from that. And that'skind of I'm excited to see what
it can do, whether it's, youknow, a big query integration
with GA four, or whether thereare some different sort of AI
(38:25):
software's that need to beincorporated into it? Or if it's
just a lot of combined apps thatwork together to output you
know, machine analytics. I'mjust curious what that what
those feedback loops could looklike. Because that conversation,
if it if speed is involved, andaccuracy is involved on the
data, that could really makethings interesting from a
(38:47):
marketing perspective. And thenlast for me is customer
interaction like engagementmetrics, like how is AI going to
be used in conversationalmarketing and chatbots? Like,
will all chatbots be mostly AIby the end of the year? Or will
still there be people, you know,typing in chats at a chat
(39:09):
center? You know, thepersonalization? How personal?
Can content on a website or inan email be and how relevant to
an audience and how perfectlytimed? Can it get those kind of
those three areas like how toresearch and generate the
directions, how to actuallyengage the customers and keep
(39:31):
them engaged? And then how toget feedback on all that
engagement like that. That is mykind of world of AI this year,
there's going to be a lot of funthings that happen in code
development and image generationand some others but I'm going to
let better qualified peoplefocus on
Nels Jensen (39:47):
You know, I, I love your list, and I especially love
feedback loops. And let me justgo on a short little tangent
here. We'll get to my list soonbut okay, the world of hiring It
is so broken, right? You hearabout these companies that do 10
rounds of interviews and, youknow, you hear about in
(40:07):
manufacturing, you know, we haverunners, the the usually brand
new people to the workforce whodon't even make it through their
first, you know, full day on thejob, the closing the feedback
loops on the recruiting, hiring,onboarding, retention, you just,
it's like you can, you canimprove your communication so
(40:29):
much, if you can close feedbackloops, you don't need to have
the third round of interviews,if you can adequately capture
what happened in the first tworounds of interviews, you know,
it's so to me, and I'm just sortof riffing off this, I hadn't
really thought about it. But AIshould help the recruitment
world, and the onboarding worldso so much, it's Oh, please,
(40:52):
please,
Joey Strawn (40:53):
I will add a layer to that, I will add a layer to
that, as well, as someone whohas been in the seat of hiring
and looking through resumes, Ihave seen how the quote unquote,
automated systems have worked inthe past. And it's a matter of
like, look for these sixkeywords in a resume. And if
those keywords aren't in thatresume, then put them in the Do
(41:15):
not contact folder and yeah, meas a hot that does those to me
eliminate so many goodcandidates, because it's not
learning. It's not. It's notintelligent. It's all just based
around arbitrary placeholdersand milestones. And that, to me,
is an area of just like yousaid, if there's a machine or if
(41:36):
there's an AI that canlegitimately scour through the
resumes and highlight the mostrelevant and helpful ones that
would one get people the rightjobs and help hiring directors
and decrease the amount ofinterviews people have to do it
with just that's a good,amazingly apt area for 2023
(41:58):
exploration for somebody. Allright, app developers get on
their AI. All right. All right.
Now, I want to know, what areyou focused on this year?
Nels Jensen (42:06):
So I what I will be exploring with AI in 2023. And
we've talked a little bit aboutit. So I my list of three, the
the first one is what I callframing and experimentation.
Right? So I'm basically lookingat it from a content strategy
perspective. So I'll beexperimenting with, you know,
(42:26):
because the better set ofinstructions you give, the
better result you get. So I'llbe trying to learn how to frame
the ask to get a better outcome.
And then the the depth part ofthat will be the applications in
the execution, right? Can youreally ask it for this sight
type of task? Or how good is it,you know, when you, you know,
(42:48):
are doing completely different
Joey Strawn (42:50):
Or, or maybe it's just sort of for someone who
tasks. So to me, that's the,that's sort of the big picture.
We've talked about research tome, I'm much more interested in
the micro research, I, I see alot of the macro, to me is not
nearly as helpful, at least inthe things that I've tried and
the things I do. So again, it'sit's really more in depth
(43:14):
research. But I think there'shuge potential from a content
perspective on that. And thethird one is saving time. And
that, for example, where ithelps at a macro is like writing
a summary of something. So itcould be as something as simple
as, you know, okay, I've come tothe point where I need to, you
(43:35):
know, talk about what are whatare the benefits of, you know,
adding 3d printing or additivemanufacturing, you know, just I
need a quick little summaryhere, whatever, that's been
written a million times, right.
So I could go pull somethingdirectly from the web, but it's
actually easier to put it in thetool and frame it a little bit
(43:57):
and get something that I want.
Similar for a list of benefitsor a list of hurdles or
challenges. It's like, thelittle the little list of goals,
if you will, it can be very goodat generating those. The big
mystery for me is the idea of ofnot using it as a research
(44:18):
assistant, but using it for arough draft. So a first draft,
and that's where I think I'll bespending considerable time
including next week, you know,looking at okay, so how
effective you know, can this beisn't, you know, if you're a
really slow writer, it mightsave more time. People, a lot of
(44:39):
people find it easier to editwhat is in front of them instead
of generate something that isn'tthere. Right.
needs a kickstart maybe givingseven prompts around a single
idea is what they need. It saves30 minutes of staring at a blank
screen. And didn't come up withthose seven prompts on their
(45:02):
own. One of the things that yousaid that I also want to loop
back to because I think it'sgoing to be interesting to see
where it goes in the future isthe framing. That to me is going
to be very interesting. Iremember when colleges started
to teach, like the differentBoolean search codes and the
(45:22):
different ways to search Google,I was like, Well remember, if
you put quotations around it, itwill search the exact phrase.
And if you put site colonquotations around a thing, it
will search just that web,there's all these specific ways
you can search. In Google, or insearch engines. We're gonna have
a whole new lexicon and, anddictionary of ways to frame
(45:46):
questions to AI, that pump outthe best answers, and we're
gonna have to as a society ofhumans, be trained in how to do
that. It's almost gonna be likelearning how to communicate with
the robots, but like, ya know,to your point analysis, like in
academia, I, it would notsurprise me in, you know, three
or four years that a part of amarketing class would be, hey,
(46:08):
how do you frame a search withinan AI platform to get the
correct content generated or toget the correct sources
referenced or whatever it maybe? That to me will be an
interesting new area ofexploration in the future.
Nels Jensen (46:26):
Yep. So yeah, it's it's a, it is a new world, even
though it's not totally newworld. But right. There's always
like one, yep, yep.
Joey Strawn (46:38):
Feels Well, it's a new year, a new you a new
Nell's, a new AI. It's a newuniverse of all of this
artificial intelligence. I mean,like, like you said at the
beginning, and I do want toreiterate, we we're not
developers on any of these,we're not experts on these
tools. A lot of this, we'reexploring at the same rate that
everybody else is. And, youknow, this is something that we
(47:00):
know is going to make adifference in all of our realms.
And so if you have tools thatyou've been using, or if you
have things that we didn't talkabout on this episode that you
think are worth exploring, letus know. You know, email us at
podcast and industrialmarketer.com. Follow us on
social media or comments on theblog posts on our website,
industrial marketer.com. Forthis episode, let us know what
(47:23):
you're exploring how you'refinding useful industrial
marketing uses for artificialintelligence, or even some of
the problems that you've runinto. A lot of this year, we're
going to be diving into thetechnology and the tools that
surround us as marketers and sowe want to know those hurdles
that you're running into. Followthe show, share with your
(47:44):
friends, let's pull together andand ask us those questions.
Because those are the thingsthat we want to be answering and
talking about this season on theshow. So again, if you haven't
subscribed to the podcast, whatare you doing? Come on, you know
you want to and email us yourquestions and your and your
comments at podcasts andindustrial market. or.com
Nelson, are you excited? Are youpumped about what the machines
(48:05):
are going to do for us?
Nels Jensen (48:07):
I am putting the machines to work hopefully I
will have let it outcome.
Joey Strawn (48:12):
Lets put them to work and hope that it's not like
a terminator output. But forright now. We're good. They're
going to be our great virtualassistants. And we're going to
put them to work and let's seewhat great marketing we can make
in 2023
Popular Podcasts
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
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.