June 23, 2025 • 47 mins
James Bergl is join by Rosh Singh from Relevance AI, revealing how their "AI workforce" is empowring sales teams . Discover insights on building adaptable, resilient, and curious sales talent.
A must-watch for leaders aiming to redefine growth through intelligent automation, scaling with the quality of ideas, not just headcount.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
What we are is the home of AI workforce, creating
AI agents that autonomously areable to perform human dependent
tasks.
Speaker 2 (00:07):
Leveraging this technology for replacing SDIs.
Speaker 1 (00:12):
I wouldn't say it's necessarily for replacing SDRs,
but it's more for eliminatingmanual work that people do.
Speaker 2 (00:18):
From revolutionizing cloud sales to driving growth at
one of the most exciting AIcompanies in the world, Meet
Rosh Singh, the powerhouse headof sales at Relevance AI.
Speaker 1 (00:30):
He's not just selling the future he's helping build
it one AI workforce at a time.
The agent would go out booktons of meetings for me.
I would be doing 20 plus firstmeetings a week, like very
consistently, like, if you thinkof an SDR, a lot of what they
do, like you really want them tobe on the phone all day, right,
they can make a hundred callsand not know a single thing
(00:51):
about any of the clients they'rereaching out to or the
prospects.
There's so many things thatmanual effort is being spent on,
but with an ai agent, whatyou're essentially doing have
you ever done business inmalaysia?
Speaker 2 (01:00):
yes, and how do you find the difference in um
cultures, like doing business inasia?
Oh man, it's.
Speaker 1 (01:07):
It's so different and , honestly like, because I ran
um sales and marketing for acyber security startup in uh and
they were primarily based inmalaysia.
So, um, I used to go over to kl.
At that time I was going tokill a lot, um, but man doing
like like for log me in.
I did a lot of APAC stuff inSingapore, uh, malaysia,
(01:30):
thailand, indonesia, whateverbut when actually going and
being like, hey, let's go andclose deals with Malaysians in
Malaysia, not from aninternational context, oh man,
it was.
It was pretty different and, tobe honest, I did not enjoy it
at all.
Speaker 2 (01:45):
Yeah, it's funny you say that because when I was in
the MSP land at Enable and thenDatto and, to be honest, pax8 as
well, we had Asian branches andwe always struggled, and I
struggled, to get the teamhumming over there, and part of
it was culture, but part of itwas also just the business
models of how the ASEAN marketdo business when it comes to IT
(02:07):
support and services, and therecurring revenue model is
something that they're notacclimatized to.
But yeah, I didn't overly enjoyit.
It was hard work.
What do they say?
The juice isn't worth thesqueeze.
Speaker 1 (02:19):
A lot of the time, yeah, time, yeah, I think, like
you know when, when doing apacwork from sydney, uh, and you're
, you're targeting.
I mean, your core of yourbusiness is always going to be
australia, um, for for the mostpart.
But then singapore is a greatmarket, right, very like mature,
like more like a western market.
But when you look in into theother markets in southeast asia,
(02:40):
you've either got to have likea very, very defined ICP that
you can go after the cream ofthe crop or you just have to
only take what comes to you,like, yeah, juice is not worth
the squeeze a lot of time.
And I don't know, like inMalaysia there was there's this
one occasion where I went overthere and with my COO he was
like, hey, come to a meeting.
So I was like, great, um, we'llgo to this meeting, we'll do
(03:02):
this, this pitch and stuff.
And we got to the meeting atlike 10 am and no one was at the
office and the guy was like I'mdownstairs having breakfast.
I was like, okay, so we go tothe cafe downstairs and then
he's there and he's with hiswife and I'm like, so we're
gonna go up.
He's like, oh no, I'm just, I'mjust ordering, like you know,
I'm just waiting for a drink.
I was like, oh, okay, cool,let's have a coffee, whatever.
It's more of a relationship.
(03:22):
So and then, like 20 minuteslater, their food arrives and
they're just like eatingbreakfast and I'm like, dude,
our meeting started half an hourago.
Like what is going on?
And um, eventually we got up tothe office at like 1115.
He's there with a bunch ofpeople.
They're all on just on theirphones the whole time.
They're like, yeah, just goahead and present, and you know
we'll be kind of listening, Iguess.
Speaker 2 (03:42):
There's nothing harder than a salesperson
presenting to a room full ofpeople on their phones yeah, and
they just feel like packing upand saying all right guys.
Speaker 1 (03:48):
I know it's just like .
What am I doing here?
Are you getting any value outof this?
So it's very different.
I'm totally.
I love the Asian culture andall that, but I think you've,
you know, you've got tounderstand where.
Speaker 2 (04:01):
Anyway, I want to, I guess, firstly thank you for
coming along today, incrediblyexcited to hear your story, and,
rather than me, do anintroduction, I thought, tell me
a little bit about who you areand particularly Relevance AI,
in terms of who they are, whatthey're doing in the market.
Speaker 1 (04:18):
Yeah, cool man.
I'm super honored to be here aswell.
Thank you so much for having meon.
So I'm Rosh Rosh Singh.
I'm the head of sales atRelevance AI, and what we are is
the home of the AI workforce,right?
So we are a platform designedfor domain experts to be able to
create AI agents in a low-codesort of way, so that they can
start automating tasks that werepreviously very dependent on
(04:40):
humans and, by doing that,really enabling companies to
scale with the quality of theirideas rather than just with raw
headcount, which is whatcompanies have really been
limited to in the I don't knowfor the last few hundred years.
Right, there's been a verylinear relationship between
business growth and headcount,and that's really what we're
trying to decouple those twometrics and really by doing that
(05:02):
, by creating AI agents thatautonomously are able to perform
human-dependent tasks and beable to scale those up
infinitely.
So that's what we're doing.
Yeah, we've been doing it, forthis whole idea of the AI
workforce came about two yearsago.
Back then, nobody knew what anAI agent was, right?
And obviously now the wholeworld's being painted with
agents, so it's been a superexciting journey.
Speaker 2 (05:23):
Yeah, incredible.
And you've been with thebusiness for two and a half
years yeah, almost three years,almost three years.
And I guess pre that was almostpre chat GPT and we're living
in a post chat GPT world todayand I think that was
revolutionized, revolutionaryfor how everyone engages with AI
as a, as a generic topic.
But I want to go back to, youknow, pre chat gpt and the
(05:49):
concept of ai, and this was aconcept where, you know,
everyone was beginning to thinkabout it, talk about it, but
never really seen anything cometo fruition.
How did you make the decisionto go from a stable career,
stable, um, stable life, um, youknow, in a non-ai world, to
thinking that this is a, this isa cool, exciting opportunity?
And how did you assess whetherthis was the right career move
for you way back then?
Speaker 1 (06:12):
Yeah, I mean we sort of intersected in the, I guess,
rmm sort of remote supportservice desk world and back then
I've been doing sort ofenterprise sales for close to 15
years prior to, or about 12, 13years prior to relevance ai.
Um, at that time I was workingfor a company called solo wins.
So solo wins um, you know,network and infrastructure
(06:34):
monitoring company, veryenterprisey, very, you know,
stable and that sort of thing.
Um, and basically I was kind ofgearing up to get engaged and,
you know, kind of move into verymuch like next chapter of life.
And I thought to myself youknow, the reason I got into the
tech industry in the first placewas to build a company.
I actually did computer scienceat uni, so I was actually a
(06:55):
developer and then I kind offell into sales and I fell in
love with it.
But I always wanted to build acompany from the ground up and
really get very involved in thetech side of things while still
remaining in the customer facingside.
I thought, man, this is my lastchance to do something crazy,
you know.
So let's do it, let's getinvolved in a smaller company.
So I went looking for jobs andbasically got two great offers
(07:19):
and one tiny company with thesetwo co-founders sort of went hey
, we'll give you like half ofwhat you're getting elsewhere.
Um, and I thought, am I crazyfor even entertaining this?
But anyway, got in discussionswith them and what they were
doing was really exciting.
They were really inspirationalwith what they wanted to do and
their vision and, um, yeah, Ijust thought, man, one last shot
(07:41):
at doing this before I've got alot more responsibility in my
life.
Speaker 2 (07:44):
So let's give it a, let's give it a crack, you know
yeah, and was the valueproposition in the solution
three years ago the same as whatit is today?
Or is that evolved?
Oh man, it's completelydifferent completely.
Speaker 1 (07:53):
Yeah, it's, it's complete.
Like back then it was very muchum, companies have.
80 of all data in companies isunstructured and we were using
vectors and clusteringvectorized data in order for
them to get value out of theirunstructured data.
There was some LLM magic inthere and actually vectors are
underlying tech of AI as we knowit today.
(08:15):
So a vector, in very simplifiedterms, is basically taking
unstructured data and turning itinto something that can be more
structured and compared withother unstructured data so that
you can essentially analyze itand draw similarities, and you
do what's called clustering ofvectors, so you can take
unstructured data, group theminto bits that are more similar
(08:39):
to each other and thenessentially get more insights
out of that unstructured data.
Speaker 2 (08:44):
And AI.
It was AI back then.
Speaker 1 (08:46):
Yeah, so it was like that.
I mean, that is a form ofmachine learning and AI LLMs,
like generative AI.
What we know of today was beingapplied on that platform More
to take that, look at that dataand turn it into something
insightful.
But yeah, that was pre-chat GPT, like you said.
And then all of a sudden wewere using GPT-3 at that time to
(09:08):
do stuff Wasn't that great, youknow.
And then chat GPT comes out.
You know, openai goes, here'sGPT 3.5 and here's a way that
you can converse with it.
And all of a sudden people werelike, oh, we don't need to
necessarily be exposed tovectors and clusters and that
sort of stuff anymore, we canjust communicate with these
generative models.
And that completely flipped iton its head.
That was about six months afterI joined.
Speaker 2 (09:31):
Yeah, well, when you joined, how big was the team?
Speaker 1 (09:34):
So we were about 20 to 22 people when I joined.
Okay, yeah.
Speaker 2 (09:38):
Okay, and you came in as a sales rep or sales leader.
I actually came in as a salesrep, yeah, so I got hired under
a sales rep or sales leaderactually came as a sales rep,
yeah.
Speaker 1 (09:44):
So I got hired under a sales leader at that time, uh,
who had a very traditionalenterprise sales background, and
he was like, hey, let's likeyou've got an enterprise sales
background, let's take thisthing and let's crack into the,
the sort of um, you know, the,uh, the big boys you know of
australia.
And I thought, okay, let's,let's do it.
You know, and I mean that timethere wasn't like the product
(10:04):
was barely figured out and weended up selling to like
insights departments, likeconsumer insights and market
research departments andagencies and stuff that wanted
to make sense of all thatunstructured data that they have
.
But, yeah, after about sixmonths I transitioned into the
leader role.
Speaker 2 (10:24):
Yeah, and what was that journey like, going from
individual contributor to salesleadership, and what was the
change in the team through thatprocess?
Speaker 1 (10:32):
So the team shrunk right.
So we actually in those sixmonths we went from 22 people to
about, like I want to say,about 11 or so, and from there
we actually raised our first,our series A, pretty soon after
that and then we started justexploding right.
But, yeah, so in that changeover to leadership, it was
(10:56):
basically just me and one or twopeople on the team and we were
basically just trying to figureout, like, what's the
go-to-market, how do we actuallytake this product that we have,
morph it into something that'sactually very useful?
And then the idea of creatingAI tools like mini automations
using LLMs came about, and thenagents to orchestrate those
(11:16):
automations and then an AIworkforce to orchestrate those
agents.
And that's where we are today.
Speaker 2 (11:21):
Yeah, incredible.
And when we talk about AIagents and workforce, what are
some of the typical use casesthat you're seeing today?
Speaker 1 (11:29):
So a really good agentic use case, and we're not
talking about, first of all,just sort of separate what a
co-pilot is and what anautopilot is right.
A co-pilot is very much likechat, gpt, right, use it.
It's like humans with a bit ofai in the loop.
Right, you're conversing withit, it's helping you do
something, but you're constantlykind of going back and forth
with it.
An autopilot is much more likehiring an ai worker that is
(11:52):
trained up on a process and thenyou delegate work to it.
It's got an objective, it's gotcapabilities, it's got a
process and it does stuff onautopilot right.
So when we talk about a reallygood agentic autopilot agentic
process, it's something that'straditionally very high volume,
something that's very repetitiveand something that, most
importantly, is not able to betraditionally automated with
(12:13):
just rules-based automation.
We see a lot of that happeningin go-to-market and in sales.
Yeah, okay, right, sales is.
I mean, you would know this.
There's so much manual workthat happens and that manual
work cannot be automated right,like very effectively.
You can automate it.
It's not going to be as good asa person.
So that's why you have juniorsales reps who will do lead
(12:34):
qualification and research andlike general SDR kind of work.
That is where most of ourconversations are, because it's
a really good agentic use case,but also because a lot of
businesses want to invest inthat because it's directly
linked to revenue.
So I would say more than 80% ofthe conversations we have are
(12:55):
in the go-to-market space, likesales space specific.
Speaker 2 (12:57):
So we're talking about leveraging this technology
for replacing SDRs or some ofthe maybe onboarding experiences
of bringing a new client onboard or engaging with them in a
pre-sales engagement, justeliminating the need for a
person in the first place, andit's working well I.
Speaker 1 (13:15):
I wouldn't say it's necessarily for replacing sdrs,
but it's more for, uh,eliminating manual work that
people do and kind ofreinventing that role.
Like, if you think of anDR, alot of what they do, like you
really want them to be on thephone all day, right, but they
could.
They could make a hundred callsand not know a single thing
about any of the clients they'rereaching out to or the
(13:36):
prospects, right.
Or they could do researchbefore every single call and end
up making 25 calls in a day,right.
How do you, like most salesleaders, want to hit a balance
between those and they do thingslike three by three research,
right, only do three bits ofresearch for three minutes and
then make your call?
Right.
But with an AI agent, whatyou're essentially doing is
you're abstracting away thatpart of their job that takes so
(13:58):
much time.
You're creating a new role inyour company of like an SDR
agent or like a researcher agent, and that agent is completely
taking that away so that the SDRcan focus purely on what only
they can do and what they dobest, which a lot of the time,
is making calls, buildingrelationships, booking meetings.
Speaker 2 (14:14):
I mean, this is really progressive technology
and I met with you probably sixmonths ago and I followed the
success and the growth of theteam and it looks like you're
going global and havingtremendous growth.
Um, I also think on the on theflip side, whilst there's
massive growth, there'spotentially and correct me if
I'm wrong here like a certaintype of buyer that is going to
(14:34):
invest in this as a, as a, Iguess, first mover.
Um, are you seeing that interms of the persona or the type
of buyer that's engaging inthis sort of technology?
Speaker 1 (14:42):
Yeah, I mean it's generally.
We find that, like, high growthcompanies are the ones who,
like they, they invest so muchin growth and, to them,
exponential growth anything theycan do to accelerate that is
going to be really worthwhile.
Yeah, with a lot of theseagentic use cases, the only way
to really scale them up is byhiring more people, and that's
(15:04):
what they're doing, right, buthow do you actually shortcut
that?
How do you do that moreintelligently?
You can't just do it throughrules-based automation or
deterministic stuff.
You have to do it a bit moreintelligently.
And they're the ones who aregoing.
So much of this manual work isholding us back.
Like this is a this is a realobstacle to our scaling.
So let's solve this problem.
Speaker 2 (15:22):
So we're really finding that high growth
companies, like traditionally,tech has been a pretty good one
for us, um, and we understandthat space, obviously, yeah,
yeah and when you're you'relooking at building out a
go-to-market strategy, a planand a process, people that you
need to go out and representyour business as well.
But how do you think about that?
(15:43):
Um, because, yeah, it soundslike when you started there was
20, then at half.
Then you've grown up again.
What was the reason?
Was that a people problem thatyou had to change?
Was it products?
Speaker 1 (15:53):
I think that on the product and engineering side, I
probably can't really speak toomuch to that, but it was
definitely on the go-to-marketside.
It was probably growing theteam before we really had a a
real concrete plan of like thisis what we want to go after,
right, and this is exactly whatour icp is and how much we can
(16:13):
actually get from this.
Um, once we actually landed onsomething that was like the time
that we really startedexploding was when we went, okay
, well, here's the idea of an AIagent.
Nobody really knows what thisAI agent is, or an AI agent
builder.
You can't go up to someone andbe like, hey, build AI agents.
I'll be like, what even is that?
(16:34):
I don't even know what it is.
How am I supposed to buildsomething?
So we sort of came up withwhat's a good use case, what's a
good application of this.
Let's build that on ourplatform and sell that as almost
like a pre-packaged product.
And that's when we said what'sthe first vertical we want to go
after?
And it was well, sales.
Right, we understand it.
We know it's a problem.
(16:55):
That's when one of the world'sfirst AI BDRs came into play and
then forming a go-to-marketstrategy around.
That was really around who'strying to grow fast, who's
hiring lots of people to grow.
Can we actually give them abetter way to do that?
And that's how we had our first10x quarter 10x quarter wow.
Speaker 2 (17:14):
So I want to double click on this because this is
really important.
In a piece I see so manystartup scale-up businesses
struggling with.
Is that chasm and crossing thatfrom?
I need to grow.
Do I just hire people that aregoing to solve my problems?
Yeah, what was the the mainproblem?
I mean, you talked about icp.
Maybe product market fit alittle bit as well.
(17:34):
Was there something that standsout that you look back and say
this is this is what we would dodifferently if we had our time
again?
Speaker 1 (17:41):
um, sort of like right, would you say, like right
now or back then, yeah, okay,yeah, um, what would we have
done differently if we were tostart again?
Um, yeah, that's an interestingone, I think.
Um, to be honest, I think thatwe should have scaled up the
sales team a little earlier thanwe are, okay, um, so we were
(18:05):
probably a little bit tooconservative with like hey,
we've got a great idea here andit's working really well.
Um, rather than just go hey,let's, let's see if we can, like
, keep this repetitive motion upor this um, repeatability up a
bit more and when you say scalethe sales team up, what was the
point where you say we shouldhave scaled?
Speaker 2 (18:22):
was that when you had a product, market fit and you
actually getting traction?
Definitely yeah.
And and the the piece beforethat would.
I think you mentioned that youmay be hard too early.
Speaker 1 (18:33):
Um, yeah, yeah, so it's a balance right like yeah
so I think, like um, in someways, it was like trying not to
repeat the mistakes of the past,right, um, but also, yeah, but
also like recognizing that thesituation was very different.
You know, um, like we were, um,we had an absolute gangbusters
(18:54):
quarter and I brought on a salesrep to be like, hey, I'm, I'm
doing a lot of this myself Likelet's actually start to scale
this up.
But at that point, I think,like really knowing that it's
going to take a while to ramppeople up, especially into a
concept that's extremely new.
Yeah, and you've got to have avery specific kind of person
that wants to take on thatchallenge, that's going to be
(19:15):
resilient, adaptable and curiousenough to take on that
challenge.
It's not something that happensovernight, right.
So, probably, if I were to doit again, I think getting like
really being on the lookout forthe right people, being able to
take our time, look for theright people, very slowly and
deliberately, is what I wouldhave given myself the time for.
Speaker 2 (19:35):
Yeah, completely agree, and and the people and I
like what you said is the, theadaptability and, um, the, the
sort of the drive to go outthere and stick with it as well.
Yeah, that journey.
When you look back at some ofthe mishires or those that maybe
weren't a fit for the business,was there anything that stands
out for I know why we hired thatperson, but actually in
hindsight, that that profile orthose characteristics don't fit
(19:58):
well in our type of business, Ithink, to look at it from a
different lens, it's more whenhiring right now, what are some
of the things that traditionallythe sales leader may look for
and go?
Speaker 1 (20:10):
that's a complete slam dunk.
But maybe wouldn't work in thissituation, right, um?
And so I typically look forthree things, right, as I was
saying before.
The first is adaptability,because, man, this is the
fastest growing space in tech,right, and you're at a company
that was basically talking aboutthe thing that everybody is
talking about right now.
We were talking about beforethat happened, right, um, it's,
(20:34):
it's evolving every day.
Um, there are new competitors,there's new technology coming
out.
This, it's just this huge goldrush, um, and the moment that we
stop innovating is the momentthat we our results start
slowing down, and you feel thatin real time.
Oh yeah, absolutely.
And so any salesperson thatcomes into this has to be able
(20:55):
to adapt very quickly,especially when they're not
given a playbook that is likethis is what works and this is
what will continue working.
So they need to be able to.
And how do you test for?
Speaker 2 (21:05):
adaptability.
Speaker 1 (21:07):
It's tough.
I would say that that's like youknow, as far as I'm sure you
want to know about hiringchallenges like that's probably
one of the ones that's thehardest.
But looking out for someonelike you know, someone who may
have worked at Salesforce andMicrosoft and Oracle all their
life and being extremelysuccessful, is potentially not
the kind of person that will besuccessful in this kind of
environment.
(21:27):
For sure, you know and I Ithink that that's not
necessarily rocket science, but,um, there are a lot of people
who think they they might be umand maybe won't, won't be
successful.
So, yeah, just talking abouttrying to understand how, uh,
let's see at a previous companythat we're at trying to
understand if there was likesome kind of crazy pivot moment
in that company or if there waslike a um, if there was like
(21:51):
some real negative event thathappened, you know how did they
actually deal with that um, andseeing how they sort of reacted
in situations like that is kindof one of the things that you
can do to test adaptability yeah, I completely agree.
Speaker 2 (22:03):
And, uh, what?
Some of the things that we talkabout and we use the same
example.
If you come from a bigcorporate Salesforce and SAP and
Microsoft you typically aregoing in selling something that
everyone has heard of.
They know who you are, theyprobably got a need, you've got
a team and an army of resources,you've got pre-sales, you've
got customer success, you've gotsales process, you've got an
(22:24):
effective comp plan.
Those people often strugglewhen you put them in great and
you put them in an environmentthat they for sure, um, that
they haven't been in before, andthey they're not able to, you
know, create a bit of demand ina situation, um, where the
person doesn't know who you'retalking about, um yeah, 100 so,
um yeah, adaptability, um, and Ilike that about, um yeah,
(22:49):
someone that's had an adverseevent in their career and how
they've dealt with it.
That's a great tip.
Right there, we also talk aboutan entrepreneurial mindset and
that's someone that continuallybuilds on the fly build the
plane as they're flying it andwe will often be looking for
similar sort of sort of scale upbusinesses people that have
have worked in a scale up beforeas well Like that's potentially
(23:11):
an indicator.
Speaker 1 (23:11):
I mean, yeah, definitely, you know it's our,
our team definitely look forpeople who have been at
companies that have scaled like,ideally, from you know,
somewhere around the 10 millionto a hundred million mark.
That's.
That's a really good indicator.
You indicator, you know thatbecause companies change so much
between those revenue ranges,um, and if they're able to
continue consistently beingsuccessful as things change so
(23:33):
rapidly, then that's a greatindicator as well when you look
at motivating a sales rep atthis sort of business where the
future is unknown, like you'reoff piste, yeah, and you're
carving this out for the firsttime.
Speaker 2 (23:46):
How do you create a compensation plan that does
drive behavior, that is going toput them in a position that
they can earn but also notbankrupt the business at the
same time?
Speaker 1 (23:55):
Yeah, that's a great question, I mean.
I think the first thing is justreally reasonable targets.
I'm very much a believer of.
I want everyone to overperform.
I don't know how long I'm goingto be able to keep that up,
right, but as far as I can, Ijust want everyone's targets to
be very reasonable so that theycan overperform and when you're
(24:17):
not hitting your target, it'sunacceptable, right, because
they are reasonable.
There's lots of businesses thatwill just go, yeah, you're
getting some crazy OT and likethe target is just, it's just
not achievable at all.
Right, I want the company thatyou know, if you're, if you're
going to work at a companythat's early stage, like series
A, series B or even sooner, um,you really want that ability to
(24:38):
be able to 200, 300, 400% yourannual or your quarterly numbers
, um, and creating anenvironment where they can do
that is definitely you, isdefinitely one of the core
pillars of that, because youwon't be able to do that at a
Salesforce.
Or I mean, maybe on your bestyear you might get close to 200
or whatever, but you'll never beable to go past that.
Speaker 2 (24:59):
Yeah, those teams have got experience and data on
their side and they know exactlywhat your maximum potential can
be and they know exactly what'sjust maximum potential can be,
and it's very exactly what'sjust a little bit out of reach,
and that's where you're going tobe right.
Speaker 1 (25:11):
Yeah, so I'd say that's one thing keeping targets
reasonable and having goodaccelerators so that, like
people really do feelincentivized to go past that um.
Another is just equity as well.
Being motivated by that is itis the hallmark of someone who
is potentially going to bereally successful in an
environment like this, becauseyou need to feel ownership over
(25:31):
the company's entire success.
Salesperson who works at acompany that's early stage does
not just need to log in, dotheir sales and then log out at
5 pm.
Right, the best ones.
Get involved in the productconversations, really feed stuff
back to the product team, getinvolved in the marketing, even
in the product conversations,really feed stuff back to the
product team, get involved inthe marketing, even in the
hiring efforts, and really feellike they own the overall
(25:53):
results of the company.
And that comes from equity.
So I'm very much a believer ofgive generous equity and always
make it known that you can getmore and more ownership of a
company if you do consistentlywell.
Speaker 2 (26:07):
When you look at lead flow and bringing on a new
sales rep, have you got aphilosophy or strategy around
the amount of inbound leadsversus self-generated?
Is there a magic number thatyou try to get to make sure
everyone's well-fed?
Speaker 1 (26:22):
Yeah, actually that's a really good question.
Um, when, when I was kind oflike doing this very solo, you
know the the table wasoverflowing with inbound right
and it was very much like talkto everyone very much biased for
numbers right, even ifsomeone's not super qualified
let's go for it.
Speaker 2 (26:41):
you know, let's talk.
And I mean, when you sayoverflowing with leads, how I
mean it sounds like everyone'sdream, I know Actually it's a
utopic environment man, I mean.
Speaker 1 (26:52):
So our agent would go out and do outbound as well for
us.
Like I've never had SDRs atthis company.
Right, we did sort of in theearly days, but when we started
doing agents we never had BDRsor SDRs and the agent would go
out, book tons of meetings forme.
I would be doing 20 plus firstmeetings a week like very
(27:14):
consistently, and that would bea combination All agent driven.
Yeah, all agent driven.
So not all outbound driven, butlike they would come to us and
the agent would pick them upstraight away, negotiate a time,
qualify them whatever and thenbook it in my calendar.
That's very, very, very real.
Speaker 2 (27:28):
Like it it was happening and, from a marketing
perspective, how were theylearning about the brands
initially?
Was that digital marketingspend?
Was it reddit?
Speaker 1 (27:37):
yeah, no, it's um, we have a really strong growth guy
um you know our, our head ofmarketing.
he's he's excellent at that um,so brought a lot of the right
people in and and I mean it wasalso like a lot of tailwinds at
that time, right, so a lot oflike there wasn't a lot of
people doing AI BDR, right, itwas a very, very hot topic, so
we're getting the right peoplefor sure.
Like, the market eventually gotflooded with a lot of vertical
(28:00):
players who are like we only doAI, bdr and it's easier to set
up and all that sort of stuffand very much a flash in the pan
.
Now the conversation is a lotwider, like what manual tasks
can we actually automate?
And we very much dominate thatspace.
But coming back to youroriginal question, what's a
magic number?
I think that a sales rep shouldideally have somewhere in
(28:22):
between 25 and 50% of their topof funnel coming from inbound 25
and 50% of their top of funnelcoming from inbound.
But it's really like I thinkthat someone having a hundred
percent of stuff coming frominbound is not necessarily the
best thing, cause it indicatesone of, maybe even both of these
two things.
One is that the quality thatyou're getting is too varied and
(28:43):
you're not speaking to, you'renot like speaking to the ideal
people with all your meetings,right, um?
Or if you are, then you shouldbe having another rep in there,
right, and you should bespreading this out across a few
reps because you ideally want to.
Let's say, you know, we, we tryand balance around 10 first
meetings a week, which is it'spretty high, honestly, like it's
(29:04):
quite high.
But, as in, you want each ofyour reps to be doing 10 first
meetings, yeah, that's right,that's how we balance.
Like, um, that's a good sort ofmetric, like an indicator of
good top of funnel and you knowyour metrics in terms of close.
Speaker 2 (29:15):
If I have this many meetings, I'm going to be
getting this many opportunities.
Speaker 1 (29:18):
Yeah, yeah I mean you have to be metrics driven.
Yeah for sure.
Yeah, um, but 10 first meetingsa week, biasing towards like
just having more conversations,because you never know when
there's going to be somethingoutside of ICP.
That turns out to be a reallygood use case, especially when
you've got a horizontal product.
But if I'm serving up 10 idealpersonas via inbound, then
(29:42):
they're not driven to do anysort of outbound generation to
the ideal people that theyshould be speaking to, and that
just means that I've got to getmore people to spread that
inbound out so that there isactually capacity to reach out
to the ideal people.
So I definitely think that 50%of pipelines should be
self-generated in an ideal worldand then the rest of it should
(30:03):
come inbound.
Speaker 2 (30:04):
That's the number I typically say as well 50%.
I think that's the rightbalance and for those that are,
I've seen hiring teams say, yeah, bring a sales rep on and it's
a hundred percent self-generated.
We've got marketing.
It doesn't really deliver, soyou've got a really expensive
SDR or BDR.
Yeah, yeah, for sure.
If you're going to just getthem to cold call all day, just
hire a resource at a quarter ofthe price.
Speaker 1 (30:33):
Right source at a quarter of the price, right,
exactly.
So I think it is importantbecause you know, like, out of
all the people you want, like,if you've got a rock solid ICP,
let's say I want to talk tosales VPs of you know high
growth tech companies If I justfilter down, if I'm getting so
much inbound of that exactperson, right, I want my reps to
to.
I want to spread that outbetween like a few different
reps, right, um, if I'm, uh, ifI'm getting a lot of inbound but
(30:53):
like, let's say, only 20 ofthat person, then I want, like
the rest of the activity to gobe, like, go find that person.
Right, not every single one ofthem is going to come to us, so
there should be effort dedicatedinto going to find them as well
, right, um, and if there isn't,then you know that something's
kind of off balance in the team.
Yeah, yeah, yeah.
Speaker 2 (31:13):
And going back to your, your indexing on reps that
you're hiring for.
You talked about adaptability.
What are the other things thatyou score when you're trying to
decide if you bring someone on?
Speaker 1 (31:24):
So there's, there's three main like there's a lot of
them, right, but the three mainones that I always talk about
are adaptability.
The second one is resilience.
So in a space that's so quicklychanging, there's going to be
real triumphant moments and thenthere's going to be times where
things kind of shift in a waythat makes it harder for you,
(31:46):
and that's just the way it goes,right.
If you can't be resilientduring the times where you've
got a bit of headwinds, thenyou're not going to have the
right space for when the timesthat things have gone well for
you, right.
So anyone in this environmentdoes need to be resilient.
If one bad quarter gets youdown, you're not going to stick
(32:08):
around, you're not going to bein the right space for the good
quarters.
You know, and it's reallyimportant that you are.
So resilience is definitely onething, because not everything's
always going to go your way.
And the third thing iscuriosity, and this is probably
one that I index on.
Like a great deal, a productlike ours can do kind of
(32:30):
anything.
It's crazy to say, I think,like a lot of people feel as if
their product is worthy of aCEO's attention, but the reality
of it is that very few productsare because they cater to a
particular persona and theysolve a particular problem.
And really, like you may be,like take, for example, an RMM
solution, right, like veryfamiliar with that.
Like, should a ceo really careabout that?
(32:52):
Of course, maybe, but like it'smostly the cios realm, um,
something that really can doanything, like our, our ai
workforce solution.
You can create agents that do,um, you know, top of funnel
sales stuff, marketing stuff,talent acquisition, h like any
other stuff in hr, customersupport, general operations,
finance.
There's so many things thatmanual effort is being spent on
(33:16):
that can be tackled with this.
And a good rep really cannot benarrow-minded enough to be like
here's one use case that I'mtalking to this person about.
This is where I'm going tosteer the conversation.
The rep truly has to be likehere's one use case that I'm
talking to this person about.
This is where I'm going tosteer the conversation.
The rep truly has to be curiousenough during a discovery to be
able to discover use cases thatmaybe even the buyer didn't
(33:38):
even think about at the start ofthe call and really get to the
root of what is the underlyingproblem that we're trying to
solve and what are the multipleways in which we can get there,
and that's really how youdifferentiate a horizontal
solution from a vertical one.
Speaker 2 (33:50):
Do you look at the sales motion so, whether it's
been an enterprise or complex orplatform or point solution like
, is that important for you?
Does that demonstrate theskills?
Are you looking for someonethat's sold that complex,
multi-stakeholder solution?
Yeah, definitely yeah.
Speaker 1 (34:09):
And it's interesting, though, because a lot of the
time you get the really seniorguys who have sold these complex
solutions to huge companies andstuff, but they really rely on
a lot of support to do it Right.
Yes, like, if you've got a toplevel enterprise guy, like
you're throwing enterpriseresources at them.
There's someone to fill outRFPs, there's someone to do
demos, there's product experts,overlays, all sorts of stuff and
(34:31):
you don't get that in a smallercompany.
So the balance really is tofind somebody who can navigate
complex multi-stakeholderprocesses but also be agile
enough and resourceful enough todo a lot of stuff on their own.
Speaker 2 (34:47):
So, yeah, that's the balance you've got to strike and
have you deployed a salesmethodology like a medic or spin
or challenger type sales?
Speaker 1 (34:54):
Yeah, so we have started doing MedPick.
I'm a fan of MedPick becauseit's not necessarily something
that you kind of tell yourpeople to do.
It's more something that youcan look back and be like what
am I missing here?
And you can very easilyidentify gaps or things that you
(35:19):
need to focus on with that deal.
It's very obvious sore thumbsin your pipeline.
Should this deal be here?
Why is this deal here when wedon't understand this?
So that's kind of why I likeMedpick a lot.
As far as sales methodologiesgo, my early days were at
LogMeIn and LogMeIn was very bigon Sandler.
Now Sandler you can see it as avery kind of insurance salesman
(35:41):
sort of methodology.
But I think a lot of the rootsof Sandler are very solid in
terms of that consultativereverse kind of selling very
much.
You know sort of not pushingand more pulling.
Yeah, I really like that.
But yeah, in terms of salesmethodologies that I'm putting
in place, like MedPick isprobably the one.
Speaker 2 (36:01):
Yeah.
Another question, just toreflect on where you are today
from a go-to-market build theworld that we're living in today
, versus, say, five I was goingto say 10, but even maybe five
years ago have you seen anyshifts?
And, whether it's withinrelevance or the industry which
(36:21):
you're potentially even drivingin go-to-market motion, are
there any new roles that you'reseeing that are starting to pop
up?
Going oh, that's a new role.
I've heard of a go-to-marketmotion.
Are there any new roles thatyou're seeing that are starting
to pop up?
Going oh, that's a new role.
I've heard of a go-to-marketengineer.
Speaker 1 (36:31):
I was going to say, yeah, gtm engineer is the one
that's popping up all the timeright now.
Yeah, I think the whole conceptof GTM engineer, I don't know.
Do you know where it came fromby the?
Speaker 2 (36:42):
way Not offhand.
I Not offhand.
I did read about it when Iheard it for the first time.
Speaker 1 (36:47):
One of the first times I ever heard GTM engineers
was a company called Clay.
You know Clay.
They started calling theirsalespeople GTM engineers
because they would be like, jumpon a call and come with a very
specific enrichment challengethat you've got and they kind of
said I'm a GTM engineer, I'mnot a salesperson.
I don't know if that was justan effort to rebrand an account
(37:07):
executive, right.
But now you have gcm engineerswho, um, they are like gcm
engineering agencies and stufflike that.
I think it's just about, um, itcomes from the fact that sales
is so much about your stack,your tech stack.
Now, if you're doing stuff witha weak tech stack, then you're
wasting time, you're notaccelerating as much as you
(37:29):
should be and you're not hittingexpectations really.
So yeah, I think that'sprobably have you guys got a GTM
engineer?
I have a RevOps guy.
That is like amazing.
Speaker 2 (37:40):
I was going to say.
My interpretation is GTMengineer is the new RevOps.
Speaker 1 (37:44):
Yeah.
So that's another way to put it, I think, somebody who
basically orchestrates yourstack, I think, and really takes
your strategy and goes like,how do we, how do we, you know,
create a tech stack thatachieves this or aligns with
this?
Crazy thing is that with, like,we actually work with a lot of
RevOps teams and one of thethings that we find is that they
(38:08):
go in going, oh, I want toautomate this thing because I
want more pipeline, I want toautomate our outbound or our
research, but then they realizethat what we really want is
pipeline productivity revenue.
But there's all sorts of otheruse cases that feed into that.
Right, it could be people arespending time figuring out who
to upsell, or people arespending time figuring out who
to upsell, or people arespending time figuring out who
has churn potential or whatever,and they go out and they find
(38:29):
other products that they add totheir stack to do that that
never really fit fully theirrequirements and don't integrate
fully and kind of fit withintheir budget.
But now our customers andRevOps are able to just create
an agent to do that within hoursand does exactly what they want
, and it's almost like theyhired a role into their company
to do that exact task in exactlythe way that they were trained
them.
So that's kind of one of thethings that you know shameless
(38:52):
plug, you know AI workforceplatform, but yeah, that's
that's kind of what GTMengineers, revops, people are
doing with, like creating agents.
Speaker 2 (39:00):
Yeah, okay, interesting.
On the topic of tech stacks,have you got anything, any new
tech stack or any technologiesthat you guys are using
internally, that is kind offront of mind, or that you're
really enjoying at the moment?
Speaker 1 (39:13):
um, one of the things that we started using one of my
guys um kind of just took afree trial of and started using
was like a deal room.
Have you?
Have you heard what a deal roomis?
Yeah, yeah so I don't know,like am I?
Am I really behind on this?
Like I, I never really used adeal room in the past.
Speaker 2 (39:28):
When I say, I know what a deal room is.
I've worked with some deal roomcompanies and maybe my
interpretation is wrong, butit's effectively a safe
environment to upload.
Exactly that's exactly what itis.
It's a secure box.
Speaker 1 (39:43):
Yeah.
So I think uploading stuffsecurely is definitely one
element of it, but it's alsojust a customized interactive
workspace between you and thecustomer, uploading stuff.
That's really relevant to yourconversation with them and it
allows them to easily circulatearound Interesting.
So you come into an environmentwhere, okay, speak to the you
(40:07):
know, um, whatever sales offmanager and you're just like,
hey, this is ai agents and thisis what you can do and all that
sort of stuff.
And then they go, oh, I've gota loop in, I gotta loop in this
person, I gotta loop in thatperson, or my cio wants to talk
to you about security orwhatever it might be.
Yeah, and so instead of, oh,here's, you know, can you send
me this and can you send me that?
And and okay, well, here's abunch of documents flying around
(40:27):
and a million different emails.
You just upload it into thisdeal room and you go, yeah, send
them the link to this deal room, or here it is, or you can
connect, you can multi-threadsomeone on LinkedIn and just be
like, oh, you're the CIO.
Hey, I'm talking to your VP ofsales over in Australia.
He's a CIO in the US.
Like, just in case you want tobe kept up to date with the
conversation.
Here's the deal room Right andthey can log in into it.
(40:49):
You can see who's been engaged.
You can really see how manypeople have been invited, which
resources they're going to.
It allows you to monitor thehealth of the deal really well
as a as a sales leader.
Speaker 2 (41:00):
And so you can set it up and customize it as like
almost a project management tool.
Yeah, exactly, yeah.
So what's?
Speaker 1 (41:05):
the one that you guys use.
So we use Aligned Aligned, yeah, and their CEOs are super
active on LinkedIn.
I think they use us as areference as well.
But, yeah, that goes reallywell.
I think that's a part of thetech stack that I hadn't really
used before, but it's been superhelpful.
Speaker 2 (41:21):
Yeah, interesting.
I'm just as you're talking.
I'm thinking that would besuper useful for the recruiting
side of our business, going backand forth with candidates and
CVs and multiple stakeholders,and it's either on Slack or it's
an email, exactly, and you cancontrol that conversation.
Speaker 1 (41:35):
Just seize control of all the stuff that is floating
around in another company, getvalue out of it, but also make
it really easy for them to shareit around.
So that's been really good andfor me in our sales methodology
it's become a very standard,like after you have a first
meeting set up in a line room.
Here's a template send them thelink rather than be, like you
(41:55):
know, at the end of every calloh, send me the deck and send me
the recording and send mewhatever other materials that
you've got.
It's a pretty typical thingthat a prospect is going to say,
um, you just send them this,this room, and it's like
everything's in there, allowsyou to move very quickly, allows
them to have all the stuff thatthey want, allows you to be
able to see how engaged they areas well.
Speaker 2 (42:12):
Super cool.
Yeah, I like that a lot.
I'm going to check that out aswell.
Anything else in the otheroutreach tools?
Speaker 1 (42:19):
Yeah, I mean, I think a lot of what, a lot of what we
would do, is just created byagents.
To honest, like, if we, if wewant a tool, we just create an
agent for it.
Um, it's just like people youknow, if you, if you want, uh,
oh, I need something to, I don'tknow, enrich my database, uh,
or enrich my crm, like, or do abunch of research or qualify,
(42:40):
like these 10 000 companies thathave come into the very top of
funnel, you're not going to hirea person to do it.
You'd look for a tool to do it.
But now we can essentially havethe quality of hiring a person
without the cost or the timetaken to do it, just like create
an agent to do it.
You just do that with ourplatform.
So we've actually eliminated alot of our tech stack just by
using agents.
Speaker 2 (43:01):
Yeah, wow, incredible , incredible.
So just forward-looking.
Uh, the business is is grown,you guys are um global.
The team is substantiallylarger than it was.
What does the next 12 monthslook like for you?
Speaker 1 (43:14):
um, yeah, we're, we're growing super rapidly.
So I think mentioned earlierthat yeah, uh, what was it at
the about 18 months ago?
We were, you know, between sortof 10 and 20 people, 12 and 20
people.
Now we're coming up on 80,which is very cool.
We're doing all this pr nextweek for what we call agent drop
.
Um, so we're releasing, youknow, lots of announcements,
(43:36):
releasing, uh, new functionalityevery day on our platform.
Um, what's the next 12 months?
Look like like I'm I'm going tobe growing out the go-to-market
team in san francisco.
We just opened our office there.
We're going to have a truckload, we're going to have a nice
(44:00):
small but mighty sales teamthere, along with CSM solutions
engineers, any sort ofgo-to-market and customer-facing
folks there.
And, yeah, I mean our wholegoal as Relevance is that we
want to eventually have abillion agents live for a
million users around the world.
And yeah, that's kind of thedirection that we're trending
towards.
Speaker 2 (44:12):
Yeah, incredible and for anyone that is on a similar
journey to yourself, is buildingout a go-to-market team,
similar stage SaaS business.
Any final tip that you wouldgive them?
Speaker 1 (44:23):
Yeah, wow, I mean there's a lot of lessons that
we've kind of learned along theway.
Um, yeah, Wow, I mean there's alot of lessons that we've kind
of learned along the way.
Um, any tips that I would givesomeone building like at what
stage?
Like the similar stage to us, Iguess?
Speaker 2 (44:34):
Yeah, so you know, series a probably that sort of
stage where you've um, yeah,similar size, maybe 40, 60
people in the business, yep, um,thinking about the future.
Speaker 1 (44:52):
Yep, I would say, um, what's the most important thing
they should be thinking aboutnow?
Yeah, um, it sort of depends onon the kind of solution that
they have.
But I would say, definitelyreally try and understand.
Uh, try and do one thing reallywell um, definitely really try
and understand the icp for thatone thing that you're doing and
really do that extremely well.
Um, that's you know when, whenwe had a platform that can build
agents and they can kind of doa bit of anything, we went, well
(45:13):
, we could just try and geteverybody to use this in exactly
the way that they want.
But really the best approachwas to go what's one problem
that this solves?
That is really like a goodproblem to solve and people are
definitely looking to solve.
Get deep into that, reallyunderstand your buyer for that.
Um, you know, create yourmaterials around that, create
(45:33):
your whole presence around thatand then build out from there.
You know, um, that's really whathelped us do, like, like I said
, that first 10x quarter isreally getting deep into what do
sales leaders want and where'sall that manual work happening?
And and let's create somethingthat's actually going to address
that.
Or you've probably createdsomething already, but how do
you frame your entire messagearound addressing that need?
(45:55):
And I guess the other thingwould be build your team on very
solid foundations.
Make sure that you've got verymuch the right people.
Our new head of people is veryadamant on this.
Get exceptional people thatdon't settle for very good,
settle for exceptional.
They're going to end up doingtwice the output of a very good
(46:15):
person and you really want tobuild your foundation like that
because this stage of thejourney it's extremely important
.
Speaker 2 (46:22):
Yeah, love that.
And if anyone listening wantsto get in touch with you or
learn more about relevance, howcan they do so?
Speaker 1 (46:27):
Oh man, connect with me on LinkedIn.
That's the best way to connectwith me on LinkedIn.
Just jump onto relevance, takea trial, request a demo, mention
me.
That's totally fine.
We are going to be making a lotof noise in the next few days.
So, yeah, definitely getinvolved and see how you can get
involved in this crazy world ofAI agents right now.
(46:48):
Really, really simply, I wouldsay that that's the best way to
go.
Speaker 2 (46:52):
Amazing, super, super interesting, valuable
conversation.
Thank you so much.
Speaker 1 (46:56):
Thank you.
Thank you so much, james.
This is really really good,super honored to be part of it
and, yeah, I can't wait tofollow the rest of the journey
of the company and the greatwork you guys are doing.
What we are is the home of AI workforce, creating
AI agents that autonomously areable to perform human dependent
tasks.
Speaker 2 (00:07):
Leveraging this technology for replacing SDIs.
Speaker 1 (00:12):
I wouldn't say it's necessarily for replacing SDRs,
but it's more for eliminatingmanual work that people do.
Speaker 2 (00:18):
From revolutionizing cloud sales to driving growth at
one of the most exciting AIcompanies in the world, Meet
Rosh Singh, the powerhouse headof sales at Relevance AI.
Speaker 1 (00:30):
He's not just selling the future he's helping build
it one AI workforce at a time.
The agent would go out booktons of meetings for me.
I would be doing 20 plus firstmeetings a week, like very
consistently, like, if you thinkof an SDR, a lot of what they
do, like you really want them tobe on the phone all day, right,
they can make a hundred callsand not know a single thing
(00:51):
about any of the clients they'rereaching out to or the
prospects.
There's so many things thatmanual effort is being spent on,
but with an ai agent, whatyou're essentially doing have
you ever done business inmalaysia?
Speaker 2 (01:00):
yes, and how do you find the difference in um
cultures, like doing business inasia?
Oh man, it's.
Speaker 1 (01:07):
It's so different and , honestly like, because I ran
um sales and marketing for acyber security startup in uh and
they were primarily based inmalaysia.
So, um, I used to go over to kl.
At that time I was going tokill a lot, um, but man doing
like like for log me in.
I did a lot of APAC stuff inSingapore, uh, malaysia,
(01:30):
thailand, indonesia, whateverbut when actually going and
being like, hey, let's go andclose deals with Malaysians in
Malaysia, not from aninternational context, oh man,
it was.
It was pretty different and, tobe honest, I did not enjoy it
at all.
Speaker 2 (01:45):
Yeah, it's funny you say that because when I was in
the MSP land at Enable and thenDatto and, to be honest, pax8 as
well, we had Asian branches andwe always struggled, and I
struggled, to get the teamhumming over there, and part of
it was culture, but part of itwas also just the business
models of how the ASEAN marketdo business when it comes to IT
(02:07):
support and services, and therecurring revenue model is
something that they're notacclimatized to.
But yeah, I didn't overly enjoyit.
It was hard work.
What do they say?
The juice isn't worth thesqueeze.
Speaker 1 (02:19):
A lot of the time, yeah, time, yeah, I think, like
you know when, when doing apacwork from sydney, uh, and you're
, you're targeting.
I mean, your core of yourbusiness is always going to be
australia, um, for for the mostpart.
But then singapore is a greatmarket, right, very like mature,
like more like a western market.
But when you look in into theother markets in southeast asia,
(02:40):
you've either got to have likea very, very defined ICP that
you can go after the cream ofthe crop or you just have to
only take what comes to you,like, yeah, juice is not worth
the squeeze a lot of time.
And I don't know, like inMalaysia there was there's this
one occasion where I went overthere and with my COO he was
like, hey, come to a meeting.
So I was like, great, um, we'llgo to this meeting, we'll do
(03:02):
this, this pitch and stuff.
And we got to the meeting atlike 10 am and no one was at the
office and the guy was like I'mdownstairs having breakfast.
I was like, okay, so we go tothe cafe downstairs and then
he's there and he's with hiswife and I'm like, so we're
gonna go up.
He's like, oh no, I'm just, I'mjust ordering, like you know,
I'm just waiting for a drink.
I was like, oh, okay, cool,let's have a coffee, whatever.
It's more of a relationship.
(03:22):
So and then, like 20 minuteslater, their food arrives and
they're just like eatingbreakfast and I'm like, dude,
our meeting started half an hourago.
Like what is going on?
And um, eventually we got up tothe office at like 1115.
He's there with a bunch ofpeople.
They're all on just on theirphones the whole time.
They're like, yeah, just goahead and present, and you know
we'll be kind of listening, Iguess.
Speaker 2 (03:42):
There's nothing harder than a salesperson
presenting to a room full ofpeople on their phones yeah, and
they just feel like packing upand saying all right guys.
Speaker 1 (03:48):
I know it's just like .
What am I doing here?
Are you getting any value outof this?
So it's very different.
I'm totally.
I love the Asian culture andall that, but I think you've,
you know, you've got tounderstand where.
Speaker 2 (04:01):
Anyway, I want to, I guess, firstly thank you for
coming along today, incrediblyexcited to hear your story, and,
rather than me, do anintroduction, I thought, tell me
a little bit about who you areand particularly Relevance AI,
in terms of who they are, whatthey're doing in the market.
Speaker 1 (04:18):
Yeah, cool man.
I'm super honored to be here aswell.
Thank you so much for having meon.
So I'm Rosh Rosh Singh.
I'm the head of sales atRelevance AI, and what we are is
the home of the AI workforce,right?
So we are a platform designedfor domain experts to be able to
create AI agents in a low-codesort of way, so that they can
start automating tasks that werepreviously very dependent on
(04:40):
humans and, by doing that,really enabling companies to
scale with the quality of theirideas rather than just with raw
headcount, which is whatcompanies have really been
limited to in the I don't knowfor the last few hundred years.
Right, there's been a verylinear relationship between
business growth and headcount,and that's really what we're
trying to decouple those twometrics and really by doing that
(05:02):
, by creating AI agents thatautonomously are able to perform
human-dependent tasks and beable to scale those up
infinitely.
So that's what we're doing.
Yeah, we've been doing it, forthis whole idea of the AI
workforce came about two yearsago.
Back then, nobody knew what anAI agent was, right?
And obviously now the wholeworld's being painted with
agents, so it's been a superexciting journey.
Speaker 2 (05:23):
Yeah, incredible.
And you've been with thebusiness for two and a half
years yeah, almost three years,almost three years.
And I guess pre that was almostpre chat GPT and we're living
in a post chat GPT world todayand I think that was
revolutionized, revolutionaryfor how everyone engages with AI
as a, as a generic topic.
But I want to go back to, youknow, pre chat gpt and the
(05:49):
concept of ai, and this was aconcept where, you know,
everyone was beginning to thinkabout it, talk about it, but
never really seen anything cometo fruition.
How did you make the decisionto go from a stable career,
stable, um, stable life, um, youknow, in a non-ai world, to
thinking that this is a, this isa cool, exciting opportunity?
And how did you assess whetherthis was the right career move
for you way back then?
Speaker 1 (06:12):
Yeah, I mean we sort of intersected in the, I guess,
rmm sort of remote supportservice desk world and back then
I've been doing sort ofenterprise sales for close to 15
years prior to, or about 12, 13years prior to relevance ai.
Um, at that time I was workingfor a company called solo wins.
So solo wins um, you know,network and infrastructure
(06:34):
monitoring company, veryenterprisey, very, you know,
stable and that sort of thing.
Um, and basically I was kind ofgearing up to get engaged and,
you know, kind of move into verymuch like next chapter of life.
And I thought to myself youknow, the reason I got into the
tech industry in the first placewas to build a company.
I actually did computer scienceat uni, so I was actually a
(06:55):
developer and then I kind offell into sales and I fell in
love with it.
But I always wanted to build acompany from the ground up and
really get very involved in thetech side of things while still
remaining in the customer facingside.
I thought, man, this is my lastchance to do something crazy,
you know.
So let's do it, let's getinvolved in a smaller company.
So I went looking for jobs andbasically got two great offers
(07:19):
and one tiny company with thesetwo co-founders sort of went hey
, we'll give you like half ofwhat you're getting elsewhere.
Um, and I thought, am I crazyfor even entertaining this?
But anyway, got in discussionswith them and what they were
doing was really exciting.
They were really inspirationalwith what they wanted to do and
their vision and, um, yeah, Ijust thought, man, one last shot
(07:41):
at doing this before I've got alot more responsibility in my
life.
Speaker 2 (07:44):
So let's give it a, let's give it a crack, you know
yeah, and was the valueproposition in the solution
three years ago the same as whatit is today?
Or is that evolved?
Oh man, it's completelydifferent completely.
Speaker 1 (07:53):
Yeah, it's, it's complete.
Like back then it was very muchum, companies have.
80 of all data in companies isunstructured and we were using
vectors and clusteringvectorized data in order for
them to get value out of theirunstructured data.
There was some LLM magic inthere and actually vectors are
underlying tech of AI as we knowit today.
(08:15):
So a vector, in very simplifiedterms, is basically taking
unstructured data and turning itinto something that can be more
structured and compared withother unstructured data so that
you can essentially analyze itand draw similarities, and you
do what's called clustering ofvectors, so you can take
unstructured data, group theminto bits that are more similar
(08:39):
to each other and thenessentially get more insights
out of that unstructured data.
Speaker 2 (08:44):
And AI.
It was AI back then.
Speaker 1 (08:46):
Yeah, so it was like that.
I mean, that is a form ofmachine learning and AI LLMs,
like generative AI.
What we know of today was beingapplied on that platform More
to take that, look at that dataand turn it into something
insightful.
But yeah, that was pre-chat GPT, like you said.
And then all of a sudden wewere using GPT-3 at that time to
(09:08):
do stuff Wasn't that great, youknow.
And then chat GPT comes out.
You know, openai goes, here'sGPT 3.5 and here's a way that
you can converse with it.
And all of a sudden people werelike, oh, we don't need to
necessarily be exposed tovectors and clusters and that
sort of stuff anymore, we canjust communicate with these
generative models.
And that completely flipped iton its head.
That was about six months afterI joined.
Speaker 2 (09:31):
Yeah, well, when you joined, how big was the team?
Speaker 1 (09:34):
So we were about 20 to 22 people when I joined.
Okay, yeah.
Speaker 2 (09:38):
Okay, and you came in as a sales rep or sales leader.
I actually came in as a salesrep, yeah, so I got hired under
a sales rep or sales leaderactually came as a sales rep,
yeah.
Speaker 1 (09:44):
So I got hired under a sales leader at that time, uh,
who had a very traditionalenterprise sales background, and
he was like, hey, let's likeyou've got an enterprise sales
background, let's take thisthing and let's crack into the,
the sort of um, you know, the,uh, the big boys you know of
australia.
And I thought, okay, let's,let's do it.
You know, and I mean that timethere wasn't like the product
(10:04):
was barely figured out and weended up selling to like
insights departments, likeconsumer insights and market
research departments andagencies and stuff that wanted
to make sense of all thatunstructured data that they have
.
But, yeah, after about sixmonths I transitioned into the
leader role.
Speaker 2 (10:24):
Yeah, and what was that journey like, going from
individual contributor to salesleadership, and what was the
change in the team through thatprocess?
Speaker 1 (10:32):
So the team shrunk right.
So we actually in those sixmonths we went from 22 people to
about, like I want to say,about 11 or so, and from there
we actually raised our first,our series A, pretty soon after
that and then we started justexploding right.
But, yeah, so in that changeover to leadership, it was
(10:56):
basically just me and one or twopeople on the team and we were
basically just trying to figureout, like, what's the
go-to-market, how do we actuallytake this product that we have,
morph it into something that'sactually very useful?
And then the idea of creatingAI tools like mini automations
using LLMs came about, and thenagents to orchestrate those
(11:16):
automations and then an AIworkforce to orchestrate those
agents.
And that's where we are today.
Speaker 2 (11:21):
Yeah, incredible.
And when we talk about AIagents and workforce, what are
some of the typical use casesthat you're seeing today?
Speaker 1 (11:29):
So a really good agentic use case, and we're not
talking about, first of all,just sort of separate what a
co-pilot is and what anautopilot is right.
A co-pilot is very much likechat, gpt, right, use it.
It's like humans with a bit ofai in the loop.
Right, you're conversing withit, it's helping you do
something, but you're constantlykind of going back and forth
with it.
An autopilot is much more likehiring an ai worker that is
(11:52):
trained up on a process and thenyou delegate work to it.
It's got an objective, it's gotcapabilities, it's got a
process and it does stuff onautopilot right.
So when we talk about a reallygood agentic autopilot agentic
process, it's something that'straditionally very high volume,
something that's very repetitiveand something that, most
importantly, is not able to betraditionally automated with
(12:13):
just rules-based automation.
We see a lot of that happeningin go-to-market and in sales.
Yeah, okay, right, sales is.
I mean, you would know this.
There's so much manual workthat happens and that manual
work cannot be automated right,like very effectively.
You can automate it.
It's not going to be as good asa person.
So that's why you have juniorsales reps who will do lead
(12:34):
qualification and research andlike general SDR kind of work.
That is where most of ourconversations are, because it's
a really good agentic use case,but also because a lot of
businesses want to invest inthat because it's directly
linked to revenue.
So I would say more than 80% ofthe conversations we have are
(12:55):
in the go-to-market space, likesales space specific.
Speaker 2 (12:57):
So we're talking about leveraging this technology
for replacing SDRs or some ofthe maybe onboarding experiences
of bringing a new client onboard or engaging with them in a
pre-sales engagement, justeliminating the need for a
person in the first place, andit's working well I.
Speaker 1 (13:15):
I wouldn't say it's necessarily for replacing sdrs,
but it's more for, uh,eliminating manual work that
people do and kind ofreinventing that role.
Like, if you think of anDR, alot of what they do, like you
really want them to be on thephone all day, right, but they
could.
They could make a hundred callsand not know a single thing
about any of the clients they'rereaching out to or the
(13:36):
prospects, right.
Or they could do researchbefore every single call and end
up making 25 calls in a day,right.
How do you, like most salesleaders, want to hit a balance
between those and they do thingslike three by three research,
right, only do three bits ofresearch for three minutes and
then make your call?
Right.
But with an AI agent, whatyou're essentially doing is
you're abstracting away thatpart of their job that takes so
(13:58):
much time.
You're creating a new role inyour company of like an SDR
agent or like a researcher agent, and that agent is completely
taking that away so that the SDRcan focus purely on what only
they can do and what they dobest, which a lot of the time,
is making calls, buildingrelationships, booking meetings.
Speaker 2 (14:14):
I mean, this is really progressive technology
and I met with you probably sixmonths ago and I followed the
success and the growth of theteam and it looks like you're
going global and havingtremendous growth.
Um, I also think on the on theflip side, whilst there's
massive growth, there'spotentially and correct me if
I'm wrong here like a certaintype of buyer that is going to
(14:34):
invest in this as a, as a, Iguess, first mover.
Um, are you seeing that interms of the persona or the type
of buyer that's engaging inthis sort of technology?
Speaker 1 (14:42):
Yeah, I mean it's generally.
We find that, like, high growthcompanies are the ones who,
like they, they invest so muchin growth and, to them,
exponential growth anything theycan do to accelerate that is
going to be really worthwhile.
Yeah, with a lot of theseagentic use cases, the only way
to really scale them up is byhiring more people, and that's
(15:04):
what they're doing, right, buthow do you actually shortcut
that?
How do you do that moreintelligently?
You can't just do it throughrules-based automation or
deterministic stuff.
You have to do it a bit moreintelligently.
And they're the ones who aregoing.
So much of this manual work isholding us back.
Like this is a this is a realobstacle to our scaling.
So let's solve this problem.
Speaker 2 (15:22):
So we're really finding that high growth
companies, like traditionally,tech has been a pretty good one
for us, um, and we understandthat space, obviously, yeah,
yeah and when you're you'relooking at building out a
go-to-market strategy, a planand a process, people that you
need to go out and representyour business as well.
But how do you think about that?
(15:43):
Um, because, yeah, it soundslike when you started there was
20, then at half.
Then you've grown up again.
What was the reason?
Was that a people problem thatyou had to change?
Was it products?
Speaker 1 (15:53):
I think that on the product and engineering side, I
probably can't really speak toomuch to that, but it was
definitely on the go-to-marketside.
It was probably growing theteam before we really had a a
real concrete plan of like thisis what we want to go after,
right, and this is exactly whatour icp is and how much we can
(16:13):
actually get from this.
Um, once we actually landed onsomething that was like the time
that we really startedexploding was when we went, okay
, well, here's the idea of an AIagent.
Nobody really knows what thisAI agent is, or an AI agent
builder.
You can't go up to someone andbe like, hey, build AI agents.
I'll be like, what even is that?
(16:34):
I don't even know what it is.
How am I supposed to buildsomething?
So we sort of came up withwhat's a good use case, what's a
good application of this.
Let's build that on ourplatform and sell that as almost
like a pre-packaged product.
And that's when we said what'sthe first vertical we want to go
after?
And it was well, sales.
Right, we understand it.
We know it's a problem.
(16:55):
That's when one of the world'sfirst AI BDRs came into play and
then forming a go-to-marketstrategy around.
That was really around who'strying to grow fast, who's
hiring lots of people to grow.
Can we actually give them abetter way to do that?
And that's how we had our first10x quarter 10x quarter wow.
Speaker 2 (17:14):
So I want to double click on this because this is
really important.
In a piece I see so manystartup scale-up businesses
struggling with.
Is that chasm and crossing thatfrom?
I need to grow.
Do I just hire people that aregoing to solve my problems?
Yeah, what was the the mainproblem?
I mean, you talked about icp.
Maybe product market fit alittle bit as well.
(17:34):
Was there something that standsout that you look back and say
this is this is what we would dodifferently if we had our time
again?
Speaker 1 (17:41):
um, sort of like right, would you say, like right
now or back then, yeah, okay,yeah, um, what would we have
done differently if we were tostart again?
Um, yeah, that's an interestingone, I think.
Um, to be honest, I think thatwe should have scaled up the
sales team a little earlier thanwe are, okay, um, so we were
(18:05):
probably a little bit tooconservative with like hey,
we've got a great idea here andit's working really well.
Um, rather than just go hey,let's, let's see if we can, like
, keep this repetitive motion upor this um, repeatability up a
bit more and when you say scalethe sales team up, what was the
point where you say we shouldhave scaled?
Speaker 2 (18:22):
was that when you had a product, market fit and you
actually getting traction?
Definitely yeah.
And and the the piece beforethat would.
I think you mentioned that youmay be hard too early.
Speaker 1 (18:33):
Um, yeah, yeah, so it's a balance right like yeah
so I think, like um, in someways, it was like trying not to
repeat the mistakes of the past,right, um, but also, yeah, but
also like recognizing that thesituation was very different.
You know, um, like we were, um,we had an absolute gangbusters
(18:54):
quarter and I brought on a salesrep to be like, hey, I'm, I'm
doing a lot of this myself Likelet's actually start to scale
this up.
But at that point, I think,like really knowing that it's
going to take a while to ramppeople up, especially into a
concept that's extremely new.
Yeah, and you've got to have avery specific kind of person
that wants to take on thatchallenge, that's going to be
(19:15):
resilient, adaptable and curiousenough to take on that
challenge.
It's not something that happensovernight, right.
So, probably, if I were to doit again, I think getting like
really being on the lookout forthe right people, being able to
take our time, look for theright people, very slowly and
deliberately, is what I wouldhave given myself the time for.
Speaker 2 (19:35):
Yeah, completely agree, and and the people and I
like what you said is the, theadaptability and, um, the, the
sort of the drive to go outthere and stick with it as well.
Yeah, that journey.
When you look back at some ofthe mishires or those that maybe
weren't a fit for the business,was there anything that stands
out for I know why we hired thatperson, but actually in
hindsight, that that profile orthose characteristics don't fit
(19:58):
well in our type of business, Ithink, to look at it from a
different lens, it's more whenhiring right now, what are some
of the things that traditionallythe sales leader may look for
and go?
Speaker 1 (20:10):
that's a complete slam dunk.
But maybe wouldn't work in thissituation, right, um?
And so I typically look forthree things, right, as I was
saying before.
The first is adaptability,because, man, this is the
fastest growing space in tech,right, and you're at a company
that was basically talking aboutthe thing that everybody is
talking about right now.
We were talking about beforethat happened, right, um, it's,
(20:34):
it's evolving every day.
Um, there are new competitors,there's new technology coming
out.
This, it's just this huge goldrush, um, and the moment that we
stop innovating is the momentthat we our results start
slowing down, and you feel thatin real time.
Oh yeah, absolutely.
And so any salesperson thatcomes into this has to be able
(20:55):
to adapt very quickly,especially when they're not
given a playbook that is likethis is what works and this is
what will continue working.
So they need to be able to.
And how do you test for?
Speaker 2 (21:05):
adaptability.
Speaker 1 (21:07):
It's tough.
I would say that that's like youknow, as far as I'm sure you
want to know about hiringchallenges like that's probably
one of the ones that's thehardest.
But looking out for someonelike you know, someone who may
have worked at Salesforce andMicrosoft and Oracle all their
life and being extremelysuccessful, is potentially not
the kind of person that will besuccessful in this kind of
environment.
(21:27):
For sure, you know and I Ithink that that's not
necessarily rocket science, but,um, there are a lot of people
who think they they might be umand maybe won't, won't be
successful.
So, yeah, just talking abouttrying to understand how, uh,
let's see at a previous companythat we're at trying to
understand if there was likesome kind of crazy pivot moment
in that company or if there waslike a um, if there was like
(21:51):
some real negative event thathappened, you know how did they
actually deal with that um, andseeing how they sort of reacted
in situations like that is kindof one of the things that you
can do to test adaptability yeah, I completely agree.
Speaker 2 (22:03):
And, uh, what?
Some of the things that we talkabout and we use the same
example.
If you come from a bigcorporate Salesforce and SAP and
Microsoft you typically aregoing in selling something that
everyone has heard of.
They know who you are, theyprobably got a need, you've got
a team and an army of resources,you've got pre-sales, you've
got customer success, you've gotsales process, you've got an
(22:24):
effective comp plan.
Those people often strugglewhen you put them in great and
you put them in an environmentthat they for sure, um, that
they haven't been in before, andthey they're not able to, you
know, create a bit of demand ina situation, um, where the
person doesn't know who you'retalking about, um yeah, 100 so,
um yeah, adaptability, um, and Ilike that about, um yeah,
(22:49):
someone that's had an adverseevent in their career and how
they've dealt with it.
That's a great tip.
Right there, we also talk aboutan entrepreneurial mindset and
that's someone that continuallybuilds on the fly build the
plane as they're flying it andwe will often be looking for
similar sort of sort of scale upbusinesses people that have
have worked in a scale up beforeas well Like that's potentially
(23:11):
an indicator.
Speaker 1 (23:11):
I mean, yeah, definitely, you know it's our,
our team definitely look forpeople who have been at
companies that have scaled like,ideally, from you know,
somewhere around the 10 millionto a hundred million mark.
That's.
That's a really good indicator.
You indicator, you know thatbecause companies change so much
between those revenue ranges,um, and if they're able to
continue consistently beingsuccessful as things change so
(23:33):
rapidly, then that's a greatindicator as well when you look
at motivating a sales rep atthis sort of business where the
future is unknown, like you'reoff piste, yeah, and you're
carving this out for the firsttime.
Speaker 2 (23:46):
How do you create a compensation plan that does
drive behavior, that is going toput them in a position that
they can earn but also notbankrupt the business at the
same time?
Speaker 1 (23:55):
Yeah, that's a great question, I mean.
I think the first thing is justreally reasonable targets.
I'm very much a believer of.
I want everyone to overperform.
I don't know how long I'm goingto be able to keep that up,
right, but as far as I can, Ijust want everyone's targets to
be very reasonable so that theycan overperform and when you're
(24:17):
not hitting your target, it'sunacceptable, right, because
they are reasonable.
There's lots of businesses thatwill just go, yeah, you're
getting some crazy OT and likethe target is just, it's just
not achievable at all.
Right, I want the company thatyou know, if you're, if you're
going to work at a companythat's early stage, like series
A, series B or even sooner, um,you really want that ability to
(24:38):
be able to 200, 300, 400% yourannual or your quarterly numbers
, um, and creating anenvironment where they can do
that is definitely you, isdefinitely one of the core
pillars of that, because youwon't be able to do that at a
Salesforce.
Or I mean, maybe on your bestyear you might get close to 200
or whatever, but you'll never beable to go past that.
Speaker 2 (24:59):
Yeah, those teams have got experience and data on
their side and they know exactlywhat your maximum potential can
be and they know exactly what'sjust maximum potential can be,
and it's very exactly what'sjust a little bit out of reach,
and that's where you're going tobe right.
Speaker 1 (25:11):
Yeah, so I'd say that's one thing keeping targets
reasonable and having goodaccelerators so that, like
people really do feelincentivized to go past that um.
Another is just equity as well.
Being motivated by that is itis the hallmark of someone who
is potentially going to bereally successful in an
environment like this, becauseyou need to feel ownership over
(25:31):
the company's entire success.
Salesperson who works at acompany that's early stage does
not just need to log in, dotheir sales and then log out at
5 pm.
Right, the best ones.
Get involved in the productconversations, really feed stuff
back to the product team, getinvolved in the marketing, even
in the product conversations,really feed stuff back to the
product team, get involved inthe marketing, even in the
hiring efforts, and really feellike they own the overall
(25:53):
results of the company.
And that comes from equity.
So I'm very much a believer ofgive generous equity and always
make it known that you can getmore and more ownership of a
company if you do consistentlywell.
Speaker 2 (26:07):
When you look at lead flow and bringing on a new
sales rep, have you got aphilosophy or strategy around
the amount of inbound leadsversus self-generated?
Is there a magic number thatyou try to get to make sure
everyone's well-fed?
Speaker 1 (26:22):
Yeah, actually that's a really good question.
Um, when, when I was kind oflike doing this very solo, you
know the the table wasoverflowing with inbound right
and it was very much like talkto everyone very much biased for
numbers right, even ifsomeone's not super qualified
let's go for it.
Speaker 2 (26:41):
you know, let's talk.
And I mean, when you sayoverflowing with leads, how I
mean it sounds like everyone'sdream, I know Actually it's a
utopic environment man, I mean.
Speaker 1 (26:52):
So our agent would go out and do outbound as well for
us.
Like I've never had SDRs atthis company.
Right, we did sort of in theearly days, but when we started
doing agents we never had BDRsor SDRs and the agent would go
out, book tons of meetings forme.
I would be doing 20 plus firstmeetings a week like very
(27:14):
consistently, and that would bea combination All agent driven.
Yeah, all agent driven.
So not all outbound driven, butlike they would come to us and
the agent would pick them upstraight away, negotiate a time,
qualify them whatever and thenbook it in my calendar.
That's very, very, very real.
Speaker 2 (27:28):
Like it it was happening and, from a marketing
perspective, how were theylearning about the brands
initially?
Was that digital marketingspend?
Was it reddit?
Speaker 1 (27:37):
yeah, no, it's um, we have a really strong growth guy
um you know our, our head ofmarketing.
he's he's excellent at that um,so brought a lot of the right
people in and and I mean it wasalso like a lot of tailwinds at
that time, right, so a lot oflike there wasn't a lot of
people doing AI BDR, right, itwas a very, very hot topic, so
we're getting the right peoplefor sure.
Like, the market eventually gotflooded with a lot of vertical
(28:00):
players who are like we only doAI, bdr and it's easier to set
up and all that sort of stuffand very much a flash in the pan
.
Now the conversation is a lotwider, like what manual tasks
can we actually automate?
And we very much dominate thatspace.
But coming back to youroriginal question, what's a
magic number?
I think that a sales rep shouldideally have somewhere in
(28:22):
between 25 and 50% of their topof funnel coming from inbound 25
and 50% of their top of funnelcoming from inbound.
But it's really like I thinkthat someone having a hundred
percent of stuff coming frominbound is not necessarily the
best thing, cause it indicatesone of, maybe even both of these
two things.
One is that the quality thatyou're getting is too varied and
(28:43):
you're not speaking to, you'renot like speaking to the ideal
people with all your meetings,right, um?
Or if you are, then you shouldbe having another rep in there,
right, and you should bespreading this out across a few
reps because you ideally want to.
Let's say, you know, we, we tryand balance around 10 first
meetings a week, which is it'spretty high, honestly, like it's
(29:04):
quite high.
But, as in, you want each ofyour reps to be doing 10 first
meetings, yeah, that's right,that's how we balance.
Like, um, that's a good sort ofmetric, like an indicator of
good top of funnel and you knowyour metrics in terms of close.
Speaker 2 (29:15):
If I have this many meetings, I'm going to be
getting this many opportunities.
Speaker 1 (29:18):
Yeah, yeah I mean you have to be metrics driven.
Yeah for sure.
Yeah, um, but 10 first meetingsa week, biasing towards like
just having more conversations,because you never know when
there's going to be somethingoutside of ICP.
That turns out to be a reallygood use case, especially when
you've got a horizontal product.
But if I'm serving up 10 idealpersonas via inbound, then
(29:42):
they're not driven to do anysort of outbound generation to
the ideal people that theyshould be speaking to, and that
just means that I've got to getmore people to spread that
inbound out so that there isactually capacity to reach out
to the ideal people.
So I definitely think that 50%of pipelines should be
self-generated in an ideal worldand then the rest of it should
(30:03):
come inbound.
Speaker 2 (30:04):
That's the number I typically say as well 50%.
I think that's the rightbalance and for those that are,
I've seen hiring teams say, yeah, bring a sales rep on and it's
a hundred percent self-generated.
We've got marketing.
It doesn't really deliver, soyou've got a really expensive
SDR or BDR.
Yeah, yeah, for sure.
If you're going to just getthem to cold call all day, just
hire a resource at a quarter ofthe price.
Speaker 1 (30:33):
Right source at a quarter of the price, right,
exactly.
So I think it is importantbecause you know, like, out of
all the people you want, like,if you've got a rock solid ICP,
let's say I want to talk tosales VPs of you know high
growth tech companies If I justfilter down, if I'm getting so
much inbound of that exactperson, right, I want my reps to
to.
I want to spread that outbetween like a few different
reps, right, um, if I'm, uh, ifI'm getting a lot of inbound but
(30:53):
like, let's say, only 20 ofthat person, then I want, like
the rest of the activity to gobe, like, go find that person.
Right, not every single one ofthem is going to come to us, so
there should be effort dedicatedinto going to find them as well
, right, um, and if there isn't,then you know that something's
kind of off balance in the team.
Yeah, yeah, yeah.
Speaker 2 (31:13):
And going back to your, your indexing on reps that
you're hiring for.
You talked about adaptability.
What are the other things thatyou score when you're trying to
decide if you bring someone on?
Speaker 1 (31:24):
So there's, there's three main like there's a lot of
them, right, but the three mainones that I always talk about
are adaptability.
The second one is resilience.
So in a space that's so quicklychanging, there's going to be
real triumphant moments and thenthere's going to be times where
things kind of shift in a waythat makes it harder for you,
(31:46):
and that's just the way it goes,right.
If you can't be resilientduring the times where you've
got a bit of headwinds, thenyou're not going to have the
right space for when the timesthat things have gone well for
you, right.
So anyone in this environmentdoes need to be resilient.
If one bad quarter gets youdown, you're not going to stick
(32:08):
around, you're not going to bein the right space for the good
quarters.
You know, and it's reallyimportant that you are.
So resilience is definitely onething, because not everything's
always going to go your way.
And the third thing iscuriosity, and this is probably
one that I index on.
Like a great deal, a productlike ours can do kind of
(32:30):
anything.
It's crazy to say, I think,like a lot of people feel as if
their product is worthy of aCEO's attention, but the reality
of it is that very few productsare because they cater to a
particular persona and theysolve a particular problem.
And really, like you may be,like take, for example, an RMM
solution, right, like veryfamiliar with that.
Like, should a ceo really careabout that?
(32:52):
Of course, maybe, but like it'smostly the cios realm, um,
something that really can doanything, like our, our ai
workforce solution.
You can create agents that do,um, you know, top of funnel
sales stuff, marketing stuff,talent acquisition, h like any
other stuff in hr, customersupport, general operations,
finance.
There's so many things thatmanual effort is being spent on
(33:16):
that can be tackled with this.
And a good rep really cannot benarrow-minded enough to be like
here's one use case that I'mtalking to this person about.
This is where I'm going tosteer the conversation.
The rep truly has to be likehere's one use case that I'm
talking to this person about.
This is where I'm going tosteer the conversation.
The rep truly has to be curiousenough during a discovery to be
able to discover use cases thatmaybe even the buyer didn't
(33:38):
even think about at the start ofthe call and really get to the
root of what is the underlyingproblem that we're trying to
solve and what are the multipleways in which we can get there,
and that's really how youdifferentiate a horizontal
solution from a vertical one.
Speaker 2 (33:50):
Do you look at the sales motion so, whether it's
been an enterprise or complex orplatform or point solution like
, is that important for you?
Does that demonstrate theskills?
Are you looking for someonethat's sold that complex,
multi-stakeholder solution?
Yeah, definitely yeah.
Speaker 1 (34:09):
And it's interesting, though, because a lot of the
time you get the really seniorguys who have sold these complex
solutions to huge companies andstuff, but they really rely on
a lot of support to do it Right.
Yes, like, if you've got a toplevel enterprise guy, like
you're throwing enterpriseresources at them.
There's someone to fill outRFPs, there's someone to do
demos, there's product experts,overlays, all sorts of stuff and
(34:31):
you don't get that in a smallercompany.
So the balance really is tofind somebody who can navigate
complex multi-stakeholderprocesses but also be agile
enough and resourceful enough todo a lot of stuff on their own.
Speaker 2 (34:47):
So, yeah, that's the balance you've got to strike and
have you deployed a salesmethodology like a medic or spin
or challenger type sales?
Speaker 1 (34:54):
Yeah, so we have started doing MedPick.
I'm a fan of MedPick becauseit's not necessarily something
that you kind of tell yourpeople to do.
It's more something that youcan look back and be like what
am I missing here?
And you can very easilyidentify gaps or things that you
(35:19):
need to focus on with that deal.
It's very obvious sore thumbsin your pipeline.
Should this deal be here?
Why is this deal here when wedon't understand this?
So that's kind of why I likeMedpick a lot.
As far as sales methodologiesgo, my early days were at
LogMeIn and LogMeIn was very bigon Sandler.
Now Sandler you can see it as avery kind of insurance salesman
(35:41):
sort of methodology.
But I think a lot of the rootsof Sandler are very solid in
terms of that consultativereverse kind of selling very
much.
You know sort of not pushingand more pulling.
Yeah, I really like that.
But yeah, in terms of salesmethodologies that I'm putting
in place, like MedPick isprobably the one.
Speaker 2 (36:01):
Yeah.
Another question, just toreflect on where you are today
from a go-to-market build theworld that we're living in today
, versus, say, five I was goingto say 10, but even maybe five
years ago have you seen anyshifts?
And, whether it's withinrelevance or the industry which
(36:21):
you're potentially even drivingin go-to-market motion, are
there any new roles that you'reseeing that are starting to pop
up?
Going oh, that's a new role.
I've heard of a go-to-marketmotion.
Are there any new roles thatyou're seeing that are starting
to pop up?
Going oh, that's a new role.
I've heard of a go-to-marketengineer.
Speaker 1 (36:31):
I was going to say, yeah, gtm engineer is the one
that's popping up all the timeright now.
Yeah, I think the whole conceptof GTM engineer, I don't know.
Do you know where it came fromby the?
Speaker 2 (36:42):
way Not offhand.
I Not offhand.
I did read about it when Iheard it for the first time.
Speaker 1 (36:47):
One of the first times I ever heard GTM engineers
was a company called Clay.
You know Clay.
They started calling theirsalespeople GTM engineers
because they would be like, jumpon a call and come with a very
specific enrichment challengethat you've got and they kind of
said I'm a GTM engineer, I'mnot a salesperson.
I don't know if that was justan effort to rebrand an account
(37:07):
executive, right.
But now you have gcm engineerswho, um, they are like gcm
engineering agencies and stufflike that.
I think it's just about, um, itcomes from the fact that sales
is so much about your stack,your tech stack.
Now, if you're doing stuff witha weak tech stack, then you're
wasting time, you're notaccelerating as much as you
(37:29):
should be and you're not hittingexpectations really.
So yeah, I think that'sprobably have you guys got a GTM
engineer?
I have a RevOps guy.
That is like amazing.
Speaker 2 (37:40):
I was going to say.
My interpretation is GTMengineer is the new RevOps.
Speaker 1 (37:44):
Yeah.
So that's another way to put it, I think, somebody who
basically orchestrates yourstack, I think, and really takes
your strategy and goes like,how do we, how do we, you know,
create a tech stack thatachieves this or aligns with
this?
Crazy thing is that with, like,we actually work with a lot of
RevOps teams and one of thethings that we find is that they
(38:08):
go in going, oh, I want toautomate this thing because I
want more pipeline, I want toautomate our outbound or our
research, but then they realizethat what we really want is
pipeline productivity revenue.
But there's all sorts of otheruse cases that feed into that.
Right, it could be people arespending time figuring out who
to upsell, or people arespending time figuring out who
to upsell, or people arespending time figuring out who
has churn potential or whatever,and they go out and they find
(38:29):
other products that they add totheir stack to do that that
never really fit fully theirrequirements and don't integrate
fully and kind of fit withintheir budget.
But now our customers andRevOps are able to just create
an agent to do that within hoursand does exactly what they want
, and it's almost like theyhired a role into their company
to do that exact task in exactlythe way that they were trained
them.
So that's kind of one of thethings that you know shameless
(38:52):
plug, you know AI workforceplatform, but yeah, that's
that's kind of what GTMengineers, revops, people are
doing with, like creating agents.
Speaker 2 (39:00):
Yeah, okay, interesting.
On the topic of tech stacks,have you got anything, any new
tech stack or any technologiesthat you guys are using
internally, that is kind offront of mind, or that you're
really enjoying at the moment?
Speaker 1 (39:13):
um, one of the things that we started using one of my
guys um kind of just took afree trial of and started using
was like a deal room.
Have you?
Have you heard what a deal roomis?
Yeah, yeah so I don't know,like am I?
Am I really behind on this?
Like I, I never really used adeal room in the past.
Speaker 2 (39:28):
When I say, I know what a deal room is.
I've worked with some deal roomcompanies and maybe my
interpretation is wrong, butit's effectively a safe
environment to upload.
Exactly that's exactly what itis.
It's a secure box.
Speaker 1 (39:43):
Yeah.
So I think uploading stuffsecurely is definitely one
element of it, but it's alsojust a customized interactive
workspace between you and thecustomer, uploading stuff.
That's really relevant to yourconversation with them and it
allows them to easily circulatearound Interesting.
So you come into an environmentwhere, okay, speak to the you
(40:07):
know, um, whatever sales offmanager and you're just like,
hey, this is ai agents and thisis what you can do and all that
sort of stuff.
And then they go, oh, I've gota loop in, I gotta loop in this
person, I gotta loop in thatperson, or my cio wants to talk
to you about security orwhatever it might be.
Yeah, and so instead of, oh,here's, you know, can you send
me this and can you send me that?
And and okay, well, here's abunch of documents flying around
(40:27):
and a million different emails.
You just upload it into thisdeal room and you go, yeah, send
them the link to this deal room, or here it is, or you can
connect, you can multi-threadsomeone on LinkedIn and just be
like, oh, you're the CIO.
Hey, I'm talking to your VP ofsales over in Australia.
He's a CIO in the US.
Like, just in case you want tobe kept up to date with the
conversation.
Here's the deal room Right andthey can log in into it.
(40:49):
You can see who's been engaged.
You can really see how manypeople have been invited, which
resources they're going to.
It allows you to monitor thehealth of the deal really well
as a as a sales leader.
Speaker 2 (41:00):
And so you can set it up and customize it as like
almost a project management tool.
Yeah, exactly, yeah.
So what's?
Speaker 1 (41:05):
the one that you guys use.
So we use Aligned Aligned, yeah, and their CEOs are super
active on LinkedIn.
I think they use us as areference as well.
But, yeah, that goes reallywell.
I think that's a part of thetech stack that I hadn't really
used before, but it's been superhelpful.
Speaker 2 (41:21):
Yeah, interesting.
I'm just as you're talking.
I'm thinking that would besuper useful for the recruiting
side of our business, going backand forth with candidates and
CVs and multiple stakeholders,and it's either on Slack or it's
an email, exactly, and you cancontrol that conversation.
Speaker 1 (41:35):
Just seize control of all the stuff that is floating
around in another company, getvalue out of it, but also make
it really easy for them to shareit around.
So that's been really good andfor me in our sales methodology
it's become a very standard,like after you have a first
meeting set up in a line room.
Here's a template send them thelink rather than be, like you
(41:55):
know, at the end of every calloh, send me the deck and send me
the recording and send mewhatever other materials that
you've got.
It's a pretty typical thingthat a prospect is going to say,
um, you just send them this,this room, and it's like
everything's in there, allowsyou to move very quickly, allows
them to have all the stuff thatthey want, allows you to be
able to see how engaged they areas well.
Speaker 2 (42:12):
Super cool.
Yeah, I like that a lot.
I'm going to check that out aswell.
Anything else in the otheroutreach tools?
Speaker 1 (42:19):
Yeah, I mean, I think a lot of what, a lot of what we
would do, is just created byagents.
To honest, like, if we, if wewant a tool, we just create an
agent for it.
Um, it's just like people youknow, if you, if you want, uh,
oh, I need something to, I don'tknow, enrich my database, uh,
or enrich my crm, like, or do abunch of research or qualify,
(42:40):
like these 10 000 companies thathave come into the very top of
funnel, you're not going to hirea person to do it.
You'd look for a tool to do it.
But now we can essentially havethe quality of hiring a person
without the cost or the timetaken to do it, just like create
an agent to do it.
You just do that with ourplatform.
So we've actually eliminated alot of our tech stack just by
using agents.
Speaker 2 (43:01):
Yeah, wow, incredible , incredible.
So just forward-looking.
Uh, the business is is grown,you guys are um global.
The team is substantiallylarger than it was.
What does the next 12 monthslook like for you?
Speaker 1 (43:14):
um, yeah, we're, we're growing super rapidly.
So I think mentioned earlierthat yeah, uh, what was it at
the about 18 months ago?
We were, you know, between sortof 10 and 20 people, 12 and 20
people.
Now we're coming up on 80,which is very cool.
We're doing all this pr nextweek for what we call agent drop
.
Um, so we're releasing, youknow, lots of announcements,
(43:36):
releasing, uh, new functionalityevery day on our platform.
Um, what's the next 12 months?
Look like like I'm I'm going tobe growing out the go-to-market
team in san francisco.
We just opened our office there.
We're going to have a truckload, we're going to have a nice
(44:00):
small but mighty sales teamthere, along with CSM solutions
engineers, any sort ofgo-to-market and customer-facing
folks there.
And, yeah, I mean our wholegoal as Relevance is that we
want to eventually have abillion agents live for a
million users around the world.
And yeah, that's kind of thedirection that we're trending
towards.
Speaker 2 (44:12):
Yeah, incredible and for anyone that is on a similar
journey to yourself, is buildingout a go-to-market team,
similar stage SaaS business.
Any final tip that you wouldgive them?
Speaker 1 (44:23):
Yeah, wow, I mean there's a lot of lessons that
we've kind of learned along theway.
Um, yeah, Wow, I mean there's alot of lessons that we've kind
of learned along the way.
Um, any tips that I would givesomeone building like at what
stage?
Like the similar stage to us, Iguess?
Speaker 2 (44:34):
Yeah, so you know, series a probably that sort of
stage where you've um, yeah,similar size, maybe 40, 60
people in the business, yep, um,thinking about the future.
Speaker 1 (44:52):
Yep, I would say, um, what's the most important thing
they should be thinking aboutnow?
Yeah, um, it sort of depends onon the kind of solution that
they have.
But I would say, definitelyreally try and understand.
Uh, try and do one thing reallywell um, definitely really try
and understand the icp for thatone thing that you're doing and
really do that extremely well.
Um, that's you know when, whenwe had a platform that can build
agents and they can kind of doa bit of anything, we went, well
(45:13):
, we could just try and geteverybody to use this in exactly
the way that they want.
But really the best approachwas to go what's one problem
that this solves?
That is really like a goodproblem to solve and people are
definitely looking to solve.
Get deep into that, reallyunderstand your buyer for that.
Um, you know, create yourmaterials around that, create
(45:33):
your whole presence around thatand then build out from there.
You know, um, that's really whathelped us do, like, like I said
, that first 10x quarter isreally getting deep into what do
sales leaders want and where'sall that manual work happening?
And and let's create somethingthat's actually going to address
that.
Or you've probably createdsomething already, but how do
you frame your entire messagearound addressing that need?
(45:55):
And I guess the other thingwould be build your team on very
solid foundations.
Make sure that you've got verymuch the right people.
Our new head of people is veryadamant on this.
Get exceptional people thatdon't settle for very good,
settle for exceptional.
They're going to end up doingtwice the output of a very good
(46:15):
person and you really want tobuild your foundation like that
because this stage of thejourney it's extremely important
.
Speaker 2 (46:22):
Yeah, love that.
And if anyone listening wantsto get in touch with you or
learn more about relevance, howcan they do so?
Speaker 1 (46:27):
Oh man, connect with me on LinkedIn.
That's the best way to connectwith me on LinkedIn.
Just jump onto relevance, takea trial, request a demo, mention
me.
That's totally fine.
We are going to be making a lotof noise in the next few days.
So, yeah, definitely getinvolved and see how you can get
involved in this crazy world ofAI agents right now.
(46:48):
Really, really simply, I wouldsay that that's the best way to
go.
Speaker 2 (46:52):
Amazing, super, super interesting, valuable
conversation.
Thank you so much.
Speaker 1 (46:56):
Thank you.
Thank you so much, james.
This is really really good,super honored to be part of it
and, yeah, I can't wait tofollow the rest of the journey
of the company and the greatwork you guys are doing.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
New Heights with Jason & Travis Kelce
Football’s funniest family duo — Jason Kelce of the Philadelphia Eagles and Travis Kelce of the Kansas City Chiefs — team up to provide next-level access to life in the league as it unfolds. The two brothers and Super Bowl champions drop weekly insights about the weekly slate of games and share their INSIDE perspectives on trending NFL news and sports headlines. They also endlessly rag on each other as brothers do, chat the latest in pop culture and welcome some very popular and well-known friends to chat with them. Check out new episodes every Wednesday. Follow New Heights on the Wondery App, YouTube or wherever you get your podcasts. You can listen to new episodes early and ad-free, and get exclusive content on Wondery+. Join Wondery+ in the Wondery App, Apple Podcasts or Spotify. And join our new membership for a unique fan experience by going to the New Heights YouTube channel now!