Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
All right, episode
number seven AI-powered seller.
Let's go as the kids say.
So I'm excited for today'sepisode pretty big time here.
Sam Altman, if you don't know,released a video that kind of
broke parts of the internetaround B2B sales in particular.
So what I'm gonna do today isI'm gonna break down the video
for you.
I'm gonna give you my hot takeon what it means, basically that
(00:23):
, hey, they're alreadydeveloping B2B agents that can
enrich email and do a wholebunch of things in real time.
I'll also talk about othertypes of agents that you need to
know about.
I call them productivity agents, right?
You've heard me talk aboutthese discovery prep, research,
et cetera.
I'm going to keep talking aboutthem until every single seller
is using these things, right?
So today's episode is's gonnabe a quick one, but I promise
(00:48):
you're gonna get insight intowhere things are going and then,
at the end, I'm gonna get verytactical on some ways that you
can carve out time to make surethat you are.
You are not doing, you know,five to ten hours worth of work
a week, right?
You are instead using AI agentsto do the work for you.
So make sure to stick around tothe end and we'll get to that.
So, without further ado, let'sjump in and talk about AI agents
(01:08):
.
Okay, so many of you may haveseen it.
It is, we've been talking aboutit.
I've been talking about it.
You know, if you go to meetjourneyai, for example, we've
been building it.
We're talking about salesagents, but we've got some
landmark stuff happening, folks,and again, we'll play the video
for everybody as well too.
(01:29):
So if you're listening along,don't worry, you'll have the
audio or you can go check outthe video on YouTube.
If you are not aware, mr SamAltman, head of OpenAI in Tokyo,
released a video, and thisvideo highlighted a B2B sales
agent.
Out of all the use cases thathe chose to highlight, he
(01:53):
highlighted a B2B, a series ofB2B agents that are able to
handle inbound, and when I sayhandle, I mean handle.
We're talking all the stuffthat marketers and salespeople
hate quick, fast inbound.
And again it's email.
I'm going to talk about someother trends here for you, but
this is big.
It can basically look at thelead.
(02:15):
It can then use publiclyavailable information to enrich
the lead Baller, right, soyou're not paying for all those
credits and then it can engagewith the lead right, it can
answer questions back and forthand you basically are creating
these kind of series of agentsthat can do a lot of the mundane
tasks that especially a lot ofinbound reps or people are doing
(02:36):
today.
So today's episode, I'm goingto dive into this use case, but
I'm going to talk about othersthat you guys really need to be
aware of.
So stick around and we're goingto jump into it now.
So first let's listen to thisvideo live let's start with
something we know sales.
Speaker 2 (02:52):
Here we have a sales
contact form where someone is
trying to reach out to OpenAI tolearn about ChatGPT Enterprise.
Now we know how this usuallyworks.
Someone submits this form, itgoes into a system and then a
team of people.
Speaker 1 (03:06):
Look, I think we've
all seen this right your classic
form fill.
Somebody fills out a form as apart of this.
Speaker 2 (03:12):
And this can take
hours or days and feel like a
slow process for your buyers.
What could it look like if youhad agents inside of your
workforce working for you rightnow?
So first we'll submit the leadhere and we'll come over to our
system.
Here you can see what a virtualteammate, a virtual sales
(03:35):
associate.
There you go.
Speaker 1 (03:36):
So imagine this
there's no hey marketers out
there.
How much do you love leadsbeing followed up with
immediately?
Speaker 2 (03:43):
You'll see that the
agent is picking up work for you
already to try to help thislead.
The account email came in andalready it's doing what a sales
associate might do on their ownand researching that lead,
figuring out the industry, therevenue, title and other
information.
Speaker 1 (03:59):
There you go, no
costs, just all publicly
available information.
Watch out Zoom info.
Speaker 2 (04:07):
Now, in this case,
the agent was able to see that
this is a really good prospect.
Openai wants to buy OpenAI, sowe think they're a good customer
and so we're going to get thecalendar availability.
Speaker 1 (04:16):
Sam's got jokes.
Oh, look at this, boom, getcalendar availability.
Okay, imagine, okay, this isliterally an agent, my friends,
that goes and gets the calendaravailability too.
Boom, grab the lead.
Boom, enrich the lead.
Boom, look at calendaravailability and listen.
Speaker 2 (04:33):
No big deal.
It's right in Japanese.
All right, y'all.
Applications for this areendless.
If you start to look aroundyour organization, think about
(04:54):
all the small tasks that peopleon your teams do and the ways
those add up.
Speaker 1 (04:58):
All right, y'all, let
me know in the comments what
you think.
What do you think this meansfor sellers out there, right?
So you've got now an agent andwe're pairing this.
There's actually a company I'ma really big fan of called Mega
AI.
Shout out to those guys in thevoice world.
I'll do an episode.
I created a voice clone.
If I haven't already, I'll dropthe phone number.
(05:19):
You can call me if you want tolearn about the discovery
process.
That's the first one I picked.
But imagine a world where theseagents I've got this version of
OpenAI, then I've got my voiceclone agent that's doing the
phone follow-up, and let me tellyou what's so powerful about
these.
So a lot of you might look atthis, and so you know again,
throughout today's episode I'mgoing to talk about this use
(05:40):
case.
I'm going to talk about acouple others in terms of, you
know, if I'm in sales, what arethe things I want to be on top
of to make sure I'm, you know,staying ahead of the game.
So why this is such a big deal,like the idea of look, auto
enrichment we're doing thattoday.
Auto booking a calendar we'redoing that today.
Translate in real time we'regetting there, you know, doing
(06:02):
phone follow ups or some chatwe're getting there.
Let me tell you what the realbig deal is.
And this is a, this is an emailversion.
I'm going to talk about thevoice version when you can
imagine how you pair these twotogether.
Not only is the voice or theemail able to just do things in
real time right and kind ofengage in real time, they also
(06:24):
have an infinite knowledge base.
And let me tell you some weird,weird, weird, weird stuff.
Actually, you know what?
Next week's episode I'm goingto do a little mini episode on
creating my own voice agent.
So for those of you who don'tknow, I created a voice agent.
You can interact with it.
Ask it anything about thediscovery process.
But what blew my mind aboutthis type of inbound follow-up
(06:44):
in particular was this voiceagent knew everything about
every industry.
So my product leader he thoughtit would be funny.
You know he knows about allthese different you know other
technologies.
I don't.
He's like.
So you know, I'm reallyconcerned that you don't really
understand my industry.
You know, tell me about kind ofadvancements in minimal coding
(07:07):
and blah, blah, blah.
And I want you to get specificin, like what the discovery
process would look like and itjust started going.
It started saying stuff and youknow what.
We'll play the recording innext week's episode.
I'll maybe give a little.
We'll give a little tease to ithere, so you can hear it.
Speaker 4 (07:23):
That sounds great.
I'm wondering if you can helpme with something else, though,
and the thing I need help withis I'm trying to figure out how
to write the Fibonacci in aPythonic function.
So can you write me a Pythonfunction that recursively
generates the Fibonacci sequenceup to I don't know 30?
Speaker 3 (07:41):
digits Certainly.
To write a Python function thatrecursively generates the
Fibonacci sequence up to the30th term, you can use the
following code Python defFibonacci n if n e8, 1, return
an else.
Return Fibonacci in 1 plusFibonacci N If N E81,.
Return an else.
Return Fibonacci N1 plusFibonacci N2 to generate the
Fibonacci sequence up to the30th term for I in range 30.
Print Fibonacci I.
Speaker 1 (08:02):
I started talking I I
say I my voice clone started
talking about Fibonacci, numbersand like all this stuff, and
like it was smart and it wasable to take all my best
practices around what goes intogood discovery and marry it with
an infinite database, that isyou know.
You know chat GBTs, llm andjust really do a lot of cool
(08:24):
stuff in real time.
And so you know, for me, if youthink about inbound, think
about yourself as a buyer, youthink about you come inbound.
Well, what does that probablymean?
Well, it probably means Ialready kind of know what I want
.
It kind of already knows that I, you know, I probably know a
little bit about the problemthat you solve and might even
work with your competitor.
So when I come inbound, do youthink I want to get qualified?
(08:47):
Do you think I want to hop on acall with somebody who knows
absolutely nothing?
And when I ask any productrelated questions, they're like
that's a great question, let meget my AE on the phone.
The answer is hell, no.
Nobody wants to interact withsomebody who can't add value in
the equation.
My friends, this is the worldthat we're moving to.
You know we've got some otherepisodes coming up where we're
(09:08):
going to go deep on.
You know skills for tomorrow.
But I think for a lot of you,think about that.
Think about that concept ofinbound leads.
If you can't add tremendousvalue and you're just booking
appointments, literally, ifyou're an inbound rep, or if
you're a company and you haveinbound people that are just
doing this, gone, hit me up.
We're already doing this forcompanies.
We're implementing thesetechnologies, we're implementing
(09:30):
voice, we're implementing these.
People prefer these.
And then they can always say,oh, do you want to talk to
someone?
Live real, quick, great.
And then we got to make sure wepair them up with someone who
can add value.
So, my friends, these types ofagents, I want to kind of put
agents into two buckets foreverybody.
First I'll talk aboutproductivity agents, and this is
where we're going to spend abulk of today's episode.
(09:50):
Productivity agents are likewhat we've built at Scaled with
our custom GPTs or meetjourneyaiand what these productivity are
.
They're built for roles andthey go and do tasks for you.
They will do your accountresearch.
They will do call prep.
They will help you scenarioplan.
They will help you scenarioplan.
They'll help you to puttogether an account plan.
These things that might takefive, 10, 15 hours, they can do
(10:12):
in an hour and a half threehours right Of you interacting
with it.
So these productivity basedagents are made for people to
interact, to automate themundane part of their jobs.
It's just like how, when Googlecame about, maybe you know,
john used to like to go to thelibrary and so now John doesn't
have to go to the libraryanymore, john can just go Google
things.
Now that's what these agents dofor you.
(10:34):
And then we've got theseautomation agents, which is
really the next frontier, whichis like, okay, I get this part,
but like, should people just bedoing this at all?
Like, is there really a value?
And again, like an incrementalvalue.
You know, of course, maybe youcould say there's 5%, 10%, but
is there incremental value indoing things?
So, like, inbound leadengagement is a really great one
for me.
(10:54):
What about working with ourcustomers?
These automation agents?
What about the ability to run alive conversation on a new
product release with one of yourclients, at scale customized to
their use case?
There's an agent, we can buildan agent for that.
We can build an automationagent for that.
So there's a lot of thesethings where it's you know
(11:17):
update, you know basicengagement, where, yeah, we're
entering this new world, whereyou know we can start to take
some of this off of people'splate, which is fine, like who
wants to follow up with aninbound lead.
Anyway, it's not like there'sthis rich conversation that
anybody's expecting as a part ofit.
So, for a lot of you out there,I want you to think about these
(11:37):
two use cases.
Obviously, if you listen to theshow or you watch the show AI
Powered Seller now you'veprobably heard us talk me and KD
, or me and others talk aboutthe future and what's happening
as well too, but this is a bigdeal.
This, you know.
This is one of the first times,I think, that a company like
(11:58):
OpenAI has highlighted a B2Bsales use case, which is pretty
eye-opening.
So, for a lot of organizationsout there, again, what you want
to think about is step one, andI'll kind of walk you through
our process.
So we have a process at Scaled.
You know we really are leadingthe charge on AI-enabled revenue
(12:19):
transformation Art is whatwe're calling it and there's a
real simple process.
And for all of you sales reps,my AI-powered sellers, my
sellers out there you need to dothis for yourself, because it's
the exact same.
Step one is assess.
You need to break down thethings that you're doing in your
day-to-day and you need to sayshould I be doing this, could AI
(12:41):
do this for me?
And then, if you listen to thisshow, we give you 9,000
different things that it cantake off your plate.
But identify those areas.
They could be research, theycould be repetitive tasks that
you're doing over and over again.
But you have to assess for yourrole.
We're going to have a lot ofcontent come out around this on
the website.
When that does, I'll update theshow.
In the meantime, you can alwaysDM me on LinkedIn, drop a
(13:04):
comment if you're watching onYouTube like learn more, and
I'll send you over kind of thisself-assessment process that
we're working on when it's live.
So you have to just assess yourrole.
Then you have to score lift.
Okay, if I can get rid of thisthing, how big of an impact it's
gonna have.
Well, this thing looks big,this thing looks kind of big, et
cetera.
So, once you understand thatthis idea of assess, score,
(13:33):
execute right and that kind ofcomes in our AI impact report
that we do for companies, butyou can do it for yourself too.
You know what.
Maybe we'll publish a flavor ofthis for everybody in the
episode.
You know what we're going to do, that, we're going to do that.
So, team, make sure we geteveryone.
I'm going to I'll publish aflavor of the report to just get
your guys' juices flowing interms of what's possible and
(13:55):
what you can actually look topotentially review with some
recommendations.
So remind me, we'll put aversion of the impact report in
the show notes for everybody.
Take a look at that.
Let me know what you think.
I'd love to hear it in thecomments as well, of course.
If you're getting value, makesure to subscribe to your
favorite podcast channel.
Or if you're on YouTube, makesure to like and subscribe to
(14:16):
the channel.
So I think that this should beexciting.
So for a lot of sellers outthere, think about the
productivity stuff.
Gosh, if I could use customGPTs or Meet Journey AI to solve
these problems for me, likeheck, yeah, I can just go and
have, like I can spend my timegoing deep and learning this AI
stuff and being a betterindustry expert All deep and
learning this AI stuff and, youknow, being a better industry
(14:37):
expert, all those things rightas a part of this.
So that, to me, is a really bigpiece.
And then the next step I thinkabout are these automations.
You know, for you as a rep,there are actually things that
you can do right.
And again it goes back to likeonce you assess the use cases,
type those into Google, typethose into ChatGPT right, even
(15:00):
me, sometimes I still say Googleright, instead of saying like,
no, go to chat GPT first.
But even you can go and say,hey, I'm trying to find an
automated tool that can do this,this and this.
What should I do?
And I really feel like sellers.
I want you to think about this.
Your companies are behind.
They are behind.
We're working with companiesthat have 400 sellers and
they're like yeah, we deployedco-pilot.
It's like dude, come on.
Dude, come on, those are notcustom GPTs.
Just so you guys know, if youhave Copilot, you do not have
custom GPTs.
You've got like a generic AIthat's a replacement for chat
(15:22):
GPT, a mediocre replacement thatIT and security picked because
they think it's safe.
I'm sorry to burst anyone'sbubble, but that's the reality
of what it is.
It's not that it's bad, it'sbetter than nothing, but it's
not customized to your workflowsto save you significant time,
which is where custom GBTs comein.
So I would be investing timeright.
(15:42):
If I was a rep, I'd be startingto think about wow gosh, this
is so exciting, I can take thisoff my plate, that off my plate,
et cetera.
But you know these things arecoming and so, like all in other
episodes, you know, make sureto check out if you guys haven't
bought the book yet.
Innovative Seller right here.
You know, take a look at thebook.
I talk about a lot of theseconcepts and you know things
that you can do as a rep.
One of the things I talk aboutis called I call it the 80-15-5
(16:04):
principle, which is, you know,80% of your time you're invested
in that 15 five type work.
So for everyone out there, Iwant to make sure you know in
your calendar, make sure you'vegot time blocked to learn more
about this type of stuff, towatch videos like this, to
(16:27):
listen to the podcast or watchit on YouTube, because this is
the future and where we're going, so real cool video.
It's exciting to see OpenAIgoing down this route.
We're at the beginning of this.
So for a lot of you out there,productivity how's it going to
make me easier, my life easiertoday.
And then automation.
Are there ways that there'scertain technologies that can
automate it?
All of you, as sellers, can dothis stuff on your own.
(16:49):
You do not need your company'spermission.
You're not uploadingproprietary data.
Don't go and upload aSalesforce report and say Jake
said it was safe, it is probablynot that big of a deal, but I
don't want to get in troublefrom the compliance department
of somebody who said well, jakeDunlap said ABC.
So that's this week's episode.
I just wanted to do a quick, myhot take on automations and
(17:12):
also talk a little bit aboutproductivity agents too, which I
think are equally as important,and make sure that you guys are
staying on top of the what'sthe latest and greatest, as
always.
So thanks for tuning ineveryone.
Another episode, numero 7,siete in the books.
So thanks for joining.
Make sure, if you haven'talready, subscribe or follow on
your favorite podcast channels.
(17:32):
If you're on YouTube, make sureto like and subscribe as well
too, and we will see you on thenext episode of AI Powered S.