#17: RevOps Automation & AI with Justin Crawford, CEO of The Automation Company

Welcome back to another episode of the Junction, y'all.

This episode I'm super pumped for.

We actually have a guest.

Justin Crawford from the automation

company is joining us today.

He's the founder and CEO.

Justin reached out to me on LinkedIn and said

he kind of liked what we were doing here.

And we said, well, why don't you

come join us on the podcast?

You have a lot of experience in this department, so

I really just kind of want to jump into it.

Chase, what do you say? Let's do it.

All right, Justin, give us the background.

Where did you come from?

How'd you get to be in this space?

What should our audience know about you

before we jump into talking about specific

use cases around AI and cool.

Happy to be here also, but originally

from California, grew up in Texas, made

my way back to Southern California.

I went to school at Pepperdine and when I graduated, I

just knew I was going to be a marketing mogul.

Fast forward three months.

I was unpacking and packing trade show booth

for a company no longer in business.

But weren't we all, though?

I mean, that's how most of my

early marketing days were, seminars, right? Oh, man.

Yeah, we were in our salesforce.

You'll appreciate this, Justin.

We were in our Salesforce environment the other day

and we were updating our lead sources and there

were these categories for seminar internal, seminar external.

And I said, first off, when's the last time have

we ever gotten a lead from a quote unquote seminar?

Aren't they called Webinar?

We're going to get emails from people and they'll be

like, oh, I used to do seminars all the time.

I think that's probably more of

a word that's used in education.

Or I can see like a professor being like, I

give a seminar on this, but I don't know.

So I just got to say, I feel you on that one.

Lots of trade shows and you're getting it

and you're like, this doesn't feel like you

said, packing, unpacking trade show booths.

Totally get that. Yeah.

Surprisingly like 30, 40% are just like big parties.

So there's worst case scenarios.

But some of those trades, I've

actually ended up being pretty fluff.

I wouldn't trade it.

But the good thing about that role is

I actually got the opportunity to get exposed.

We're talking 1012 years ago, right?

So I got the opportunity to get exposed to Salesforce.

At the time, the company was setting

up Salesforce and coupling with Eloqua.

So I got a chance to get exposed to those tools.

So in a sense, I've really been doing

marketing automation since the start, but was kind

of accidentally thrown into it, if you will.

But from there, left that role, started working

at a pretty major three PL that was

just in the process of acquiring dozens of

companies and inheriting their tech stack.

So was able to peek behind the scenes of like Marketo

and eloque and act on, so on and so forth.

So just realized I kind of like this, and

it doesn't seem like a lot of people are

talking about it, especially not how they are today.

It wasn't such a hot topic.

Fast forward realized that there was a

gap, so I wanted to start consulting.

People were moving around that specific industry,

so that's where I really started out

in the transportation management industry.

But when the work or the request started to outgrow me,

I started to become known as like the automation guy.

Hey, this guy helped us stand

up, helps by the role company. I met a new company.

I know someone who could do that

because it was foreign language back then. Right.

So anyway, when the work started to outgrow me, I wanted

to start building a team that could support those different functions

and kind of that's where we are today, just able to

do it at a much larger scale and at a much

more rapid pace and just love it.

And AI just really changed it up completely.

So it's really exciting to see where

it's come in the last 1012 years.

I can tell you from my little experience, kind

of first getting into when you would hear Marketo.

Right.

To me at the time, ten years ago,

it was like, what does it do?

It sends emails for you,

kind of like scheduling, right?

And there's like some automation and we can

do some drip email stuff, but we weren't

talking about the whole lifecycle of a customer.

It was just kind of these one off campaigns.

Even as of five years ago or three years ago,

actually working for a company, we didn't have connected systems

that helped us kind of connect all the dots around.

Well, how did that impact something later on

in the sales cycle, if at all?

So I think that it's really interesting when you say

people weren't talking about it the way they are today.

Now it's like we have all these tools that

like HubSpot that have wraparound capabilities, and it's really

changed the way that we talk about marketing and

really they're getting into what service and sales and

all these other aspects of business.

I really like the way you described it, Justin.

It's like I didn't know a whole lot.

I started to learn more, got more

involved, got more and more involved.

Learned it to the point where I

couldn't do it all by myself anymore. Right.

And you're kind of perfecting this

whole idea along the way.

And then you are now roping in AI basically

to scale to some extent what you're doing.

And I think that's what everybody's

been trying to do, right?

You just have been doing it with more people.

It sounds like now with AI, with all the

things that are going on, there's more opportunity to

not necessarily do it with more people, but to

do it with AI to help you scale further.

Justin, where do you see that coming

into play with what you're doing now?

Well, one just kind of being

a marketing professional, that's my origin.

It's so good to not have to explain

that we do more than just like coloring

things in and social media posts now.

There's the tech side to it that's exciting.

So I feel like it gives us a bit more credibility.

But with AI, the best way I can describe it

is it really levels the playing field for us.

It touches all functions of our business, relatively

speaking, we're a pretty small agency, right?

But we're now able to have access to,

I'll say, let's use marketing as an example.

Within our marketing, we're now able to have access to

industry expertise, if you will, that we wouldn't have.

So we're able to put out

content that really bridges the gap.

So I'll give a specific example.

One issue or a challenge for us before

was how do we close this knowledge gap

between our marketing team and our service team?

Service team, we're specialists, right?

They know HubSpot like the back of their hand.

How do I communicate it to marketing?

So they can now promote or put out thought leadership just

by just doing enough to be dangerous, if you will.

Now, with AI, right, simply outlining unique needs.

What I'm trying to accomplish now, say it in

the 10th grade level and help someone who's never

understood it, understand it for the first time.

It's now able to kind of build out that

full fledged documentation for our marketing team, if you

will, and they can repurpose it and kind of

have access to knowledge that they wouldn't have had.

We no longer have to do

an hour deep dive knowledge transfer.

So I think that's really where, and I can

use that example really for all the departments.

I'll use another one for our sales team.

Previously we would do a demo, and after the

demo, we had to have a deep dive.

No matter how good the prospect

was, it was just protocol.

You had to send your proposal.

Well, now what happens is, because we're transcribing those conversations,

our service team is just able to get a nice,

clean summary of those major pain points, those unique needs,

and in addition to them now having the information they

need without actually having to be on every proposal call

or discovery call, they're also able to use AI to

put together that proposal.

And it's touching all the unique needs

that were mentioned on the call.

So instead of a generic kind of cookie cutter

proposal, each one we send out is very unique

and specific and tailored to the things that you

mentioned were your major pain points and the challenges

you wanted to solve on the call.

So just a couple of examples of how AI is

kind of used as essentially another member of the team.

Yeah, I love that thought.

I want to come back to the statement of work

that you're talking about, but I love the thought that

everybody's kind of on the same level playing field, right?

Like everybody has the same

access to some extent, right.

Generically speaking, to all of this AI stuff. Right.

And everybody can build out, and this

is your opportunity to seize it.

But then I really like your thought of like, well, I

don't know as much as you guys do about marketing.

I can go in and ask questions right in

the chat bot, be able to ask the really

dumb questions like, what does CTA stand for?

Because Mel would be like, oh,

my gosh, I work with this.

Do I think I know what that stands for?

But it gives that person that ability to ask

questions that they probably wouldn't have otherwise asked and

they would just pretend like they know.

I really love that thought.

I want to dig into what you're doing

on building out the statement of works.

Walk us through that process, because from what I've

heard, people are doing different levels of that.

We're doing some level of that.

Do you have the actual step by step?

Here are the things we're going to do.

Are you writing user stories? I don't know.

Break it down for us? I don't know. For the audience.

Right.

To understand what that looks like. Sure. Yeah.

So obviously it starts with that discovery.

It's crucial to really dig in and

just what are you wanting to solve?

But also reading in between the lines. Right.

The client won't always know what exactly it is.

They think they know what they need.

In many cases, they may know exactly what they need.

But they also are looking for that kind of

advisory on, can you point us in the right

direction based on what we think our needs are?

And from that discovery, again, it's getting

that information to the service team so

they can really scope it out.

What does this look like if

we're going to build it out?

And I would say our proposals are pretty hyper

personalized in that we really want to touch on

specifically the problems you're wanting to solve.

So in addition to layering out each of the, hey,

here are the major objectives, we also include a description

that kind of gives away the recipe, if you will.

I'm happy to do that because I want

the problem to get solved, whether it's with

us or you're solving it on your own.

So we essentially say exactly what we would

do to solve the problem in our sows. Right.

One thing I really like to do, and if you

guys haven't done it, I would recommend trying it, is

when you're leveraging the chat, CPTs of the world ask

it for justifications like, here's the problem, but justify why

this route is the right way to go and then

even provide an example of how it could be used

and the effect on the bottom line.

And I think we've seen a much better response rate

when our proposals kind of have that overkill in there.

It just gives that level of confidence that not

only do we understand, and this is a big

one, not only do we understand your problem and

we're telling you exactly how we're going to fix

it, it allows them to be the champion, because

as you guys probably know, there's several decision makers.

Right? It can be.

And it helps them go back to the team as

a champion and say, look, I found a solution and

I know exactly how we're going to solve it.

Here's the breakdown right here.

So it closes up some of

the ambiguity that you'll often get.

If it's like, optimize your workflows, what

are you going to do in there?

Are you going to break something? What does that mean?

Yeah, what does that even look like? Yeah.

So you're including the justification

in the sow itself?

Yes, we do the justification.

What tools are you using?

You mentioned chachupts of the world, but

what tool or tools are you using

to build some of this capability out? Sure. Yeah.

So we try to consolidate as much as we can.

So obviously we really like to leverage hubspot,

like every nook and cranny of it.

So that is what we use even for

our billing in a lot of cases.

But I would also say one thing that,

and I foresee a very strong integration coming,

but process modeling is another tool we use.

Lucidchart is kind of an example of one,

but that visual really helps even in the

early phases of still communicating the project.

That's one thing that I felt like we were missing out

on is we can explain it to her blue in the

face, but can you explain it to me in a way?

I can go take this to somebody who isn't

as familiar, intimately familiar with HubSpot and with Lucidchart.

I mean, we can essentially have AI platform.

I use Chat DBT and I have

another example here in a second.

But we can use Chat GBT now to build the flow

for us, build this flow for me in a process modeling

tool, and it's going to lay it out in a way,

step by step that's going to help convey the process.

And I know that's possible to do before,

but it would have taken a lot longer.

And I noticed that a lot of time,

AI just gives you food for thought.

Even if you would have had the bones

there, it's going to give you an idea

like, oh, I didn't think about that.

Right, just to help enhance that process.

And the other thing just kind of like

a separate side note, which is important because

we do kind of deal with global companies

as well, is just a translation, right.

So a tool like notion we found

to be really helpful in helping to

visualize those translations for marketing teams.

One example is you're creating an email in

English, but for the approval process, they need

to see it in three different languages.

Well, notion is a really nice

way to lay it out visually.

So you can see the three emails side by side.

So for compliance to review it, they're

now reviewing them all in one place.

But then also for the developer, he's not having to

jump around, he or she isn't having to jump around.

It's just all in one place

for them and it's visually appeasing.

So in that you're seeing the full translation right

in front of you with the images and whatnot.

So I think it just helps from

a process improvement standpoint as well.

We've been using blueprints ever from the beginning

where I have been diagramming out manually. Right.

What you're describing.

I have to check out the tool that you're

talking about, but you're actually describing something that we've

been reading in a book, the Jolt Effect.

I don't know if you've read it yet, but

one of the jolt is an acronym for what

you're supposed to do in the sales cycle.

The L is to limit the knowledge, right?

Or like, well, that sounds great, Justin, but

let me go check out this other solution.

I love that idea of the AI is

providing justification and giving you that story that

you can either use to convince yourself or

your peers that, hey, this is the solution.

We don't need to go find anything else because

that's what really lacks, and I'll be frank, in

a lot of our proposals, is why.

Well, yeah, it works and it's good,

but why should we do it?

So I really love that thought.

Curious to know, while you're building this out, is the

sow just printed for you and you guys go in

and kind of modify it here and there?

Or is it more of like a collaborative

effort where I write something, AI write something,

I write something back and forth?

What's the approach on that?

I think we've really gotten it down to a science

where it's pretty one to one in that we've found

a way to really map out, hey, here's the conversation.

Here are these main pain points

that we're trying to capture.

Of course, as we're building out sows, we're also

building out a product library, if you will.

There's times where you're just repurposing a

solution that you may have prescribed before,

but obviously it needs to be unique.

But you still have the

bones from that previous project.

And for us, it's mapping out, hey, here's the

problem, here's the need, and the beauty of it.

If you guys also haven't done this, is you can

kind of create a profile with your AI tool.

Several of them have this feature where, who are you?

So it knows our previous proposals, it knows how

we like the proposals to be laid out.

Now, we're really to a point where as long

as we kind of stick to that same format,

our proposals are going to be pretty consistent.

It knows it needs to provide justifications.

It knows to provide examples.

It knows to provide as much detail as possible.

But remember, we need to say at an

10th to 11th rate level, because I want

them to actually be able to digest it.

But I'd say it's one to one.

Of course, you have to maybe make some refinements.

It's not going to be perfect, but one

servicer usually just needs to review it.

And we're pretty confident to the point where

we're kind of confident in sending it off.

How much time has that saved you?

Have you quantified that from what it used

to take to put together that sow?

I don't have much hair left, but I think

it saved me hair in addition to time.

So a lot more grays coming in right now if I didn't.

I love that we haven't been measuring any

of our, or benchmarking any of our time

savings on AI to gray or nonexistent hair.

Hang on to what I got.

That's fantastic.

Yeah, I think it's a great call out.

People ask all the time, how do I

know that ROI of a good, just AI?

How do I measure it in ROI?

And I think time efficiency and just processes, right.

And that's really, like, the holy Trinity of it for me.

The time saving has been.

I can't even quantify it because we're able to

do it so we can get it done within

an hour now, which it would take days before.

The part where I felt it really hurt us is you

guys can imagine with how accessible data is now, clients are

a lot more empowered to go look at other alternatives.

And where we were falling short is we were having these

demo meetings, but taking three to four days to get back

with the proposal, the interest level just wasn't the same as

when we had them on that discovery call.

Talking about all these points.

You kind of lose some of the ump when it's a

week later and you're getting back to schedule that call.

Get it going. Right.

Get the SW out the door, because they would sign

it, like, right now, but then you have a three

or four day delay where that level of doubt, fear.

Right.

Am I making the right decision?

They've slept on it.

They've slept on it. Right.

I'm curious, talking about quantification, we actually

really don't do a lot of that.

In fact, I think beyond calculating

the ROI, it's probably even more

difficult to calculate the scalability of.

Well, now I did this thing, now I can

do it in an hour, which means I can

do 10,000 more sows than I would have before. Right.

Anyway, I'm curious if the way that you're

doing it now has increased your win rate.

Have you kept track of, we did it old way, human hand.

Right.

And now we're doing it new way with AI.

Do you see your win rate going up?

What other metrics are you tracking

now that you're doing it?

Maybe in a totally different way?

Yeah, that's a great question.

One thing that jumps out right away that we do notice

is, I think you were kind of alluding to it before,

is just the sample size is so much bigger now, we're

able to send out so many more sows.

And if we make a tweak, that second kind

of revised version can come so much faster.

So definitely I would attribute

it to two things, though.

What I've noticed just kind of to be the

one who actually used to have to create the

sow, which don't bring me back to that, please.

So the hyper personalization of it, how personalized

they're going to be that and how fast

we're able to turn them around, we've seen

just overall a better conversion rate.

Not only yes to the customer signing the project proposal, but

a lot of times what we're getting now is we usually

start with an audit and we always keep that in our

back pocket for those people who are like, I don't really

know what I want, but I know we need to optimize.

That actually has been what has been converting a

lot better because our audits are able to be

a lot more specific to the client's needs.

So typically what we'll start with is an audit.

And what we do is when we perform our audit,

if you decide to go with the automation company, you

can put those audit hours towards the initial project.

So there are variables that have changed now

that we do have access to AI.

So yes, our conversion rate is up, but

we're also enabled to do things and provide

offerings that we wouldn't have been able to.

I would have been a lot more hesitant in

the past to provide an audit to get started.

Right, because those can be

pretty involved, pretty manual.

Well, now what we need to do is just kind

of scan the platform, put those findings into a prompt,

and then we can have a full audit documentation built

out, which is also a value add because we're now

providing a tangible, almost 2030 page guide that's breaking down

your entire platform based on these 30 bullet points.

Our servicer has to just kind of

go through there, notate these bullet points.

You know, how audit, our tool knows

how our audit wants to be.

It's going to build out a full fledged audit, let us

know where to enter in screenshots, and if they decide to

go with us, we can put that towards a project plan.

If not, we still provided them a tangible

audit and in a way still provided some

value and essentially closed one a deal, even

though it was just for a project audit.

When you describe the audit, are you connecting your

internal platform that everybody is buying to HubSpot, right,

or to Salesforce to run a bunch of queries.

And then you've got 10,000

contacts, you've sent 300 emails.

Are you doing that level of assessment and then

generating an automatic output from AI and the data?

Yes, that's right.

We use a tool called Encycle.

It will connect your HubSpot platform.

Click of one button free.

It's going to scan your entire platform.

The beauty of it is if you say, okay, we

have 60K duplicates, you want to move forward with us.

That same tool can also just click one button

and it's going to merge all those duplicates.

It's going to standardize all the information.

So yes, we're connecting it.

So again, huge time saver.

I remember having to go through every

kind of configuration inside your platform.

Oh, this is wrong. This could be wrong.

I don't have the backstory now.

Run it through a scan, take the

scan findings, put that into an audit.

The manual piece is maybe adding in some of

the screenshots to give it that level of personalization.

But adding in 20 screenshots is a lot better than

having to build out a full 20 page document.

Totally, 100%.

We're kind of transitioning into what

you do for the clients.

Where do you see a lot of the

asks like are folks coming to you guys

and saying, hey, love this chat GPT stuff.

I don't know what it means, but I want to

use it and we have these types of problems.

Do you see a lot of that or are you

guys more so going to folks with kind of this

productized, hey, we can really do this thing for you.

As it relates to AI, I think it's usually us

bringing it to the table for clients, a big need

for our clients and where we start to leverage it.

So I would say where we have had

clients bring it to us, it's usually when

we're dealing with the sales function, they're wanting

to be able to prospect better, personalize better.

So one, if you guys are familiar with kind

of chatspot, that's the thing Hudspot's been toying with

and helping you just kind of summarize the company

really quickly to get that snapshot.

But I think when you take it a

step further, when you're able to kind of.

Here are all the properties I have for contact.

Develop a super personalized three email series coming

from a BDR goal of scheduling a meeting.

And what you're able to do now is really

have a very personal, as you guys can imagine.

And it's actually what I think is going

to be the next major shift that we

see is just hyper personalization with messaging.

But that's kind of what we're experimenting with now with a

client is, hey, gather these data points in a way.

Right?

We already have all of our properties in

HubSpot, and we have them labeled correctly.

Drop this in.

And now craft up a personalized three email

sequence for me for this particular contact.

And especially when you're doing target accounts and account

based marketing, it just helps you be able to

churn out those messages a lot faster.

Batch and Blast is kind of,

those days are slowly going away.

Even if it is batch and blast, there's

still so many ways to personalize those messages.

And I think the proof is in the pudding.

Even if you imagine yourselves right, you're more inclined to

respond to a message when it seems like it was

genuine and kind of written by an individual versus like,

hey, I can get you a $10,000 in ten minutes. Right?

Those are just kind of more

generic messages you may see.

So, yeah, it's definitely helped

from a sales function side.

I definitely think if you're leading with, maybe if you're

leading your sales conversations with, we have AI, or we

can do this, we have some sort of, like, you're

talking about more of a feature, but what people are

coming to you with and us is with business problems,

like, they want to do this thing better.

I want to have more personalized outreach.

I want to have better

marketing attribution and understand that.

And then you, as the consultant can say, well,

let's take a look at what you have today.

Do you have any automation built

into your existing tech stack?

And that might mean that the first

tactic is deploying some of that automation.

And, oh, by the way, there's this AI thing that we

can layer on top of it, but I think if companies

are leading with we have AI, that's not enough.

We need to demonstrate what is

the challenge that we're solving.

And if it happens to be utilizing AI and it

gets you a five X ROI on it, great.

But I've had some seen a lot of

SaaS companies coming out with this AI.

We've got AI, okay, but what we did, it didn't

replace the fact that I still need you to articulate

what problem you're solving or how you're making me, enabling

me to do my job better, faster.

Yeah, I actually saw something.

I think it's Artisan Aiartison Co.

I don't remember the exact website,

but found it on Reddit.

I'm sure it was a marketing post of sorts, but

they had talked about how they had just raised 1.5

million in the seed round on 15 million valuation and

they hadn't launched anything, which, I don't know, depending on

where you are in that space, maybe that's not a

big deal, or maybe it is, but what they're driving

toward is this idea of an agent.

So we've talked about agents before

in previous podcasts for the audience.

Like, an agent is like, well, I want Melbot

to come in and do my marketing for me. Right?

And now you're kind of talking to the bot,

which then goes in the background and does all

these things and then kind of reports on it.

Well, the first thing that they're doing, Justin, is

the same thing that you're talking about, right.

Is you're chatting with this

marketing BDR type bot, right?

Hey, I want to target this industry and

I'm going to focus on selling these things.

It will then go find the people that

it needs to email, it'll write the sequences

and then book meetings for you.

And I think the intent, right.

Is that you only kind of see

the results you're not really involved in.

Like, yeah, I would change this word.

It's doing it all on its own, Justin.

It sounds like that's kind of

the direction that it's going.

Do you kind of feel the same way or

is this platform going to crash and burn?

Well, I think it depending.

I need to see the first

few conversations to make a decision.

But I will say that concept of what

you're describing, we've seen success using it, short

of AI, in that we're automating. Right.

We have kind of some of the

predefined prompts limiting what people can respond.

I think if they're able to open that up

to where it's just an open kind of text,

and it has the knowledge to be able to

respond in a way that represents the company's tone.

I think it could be very successful because we've seen

success in kind of doing that same thing with SDRs.

But just like I said, with a predefined kind

of conversation, and then it allows the SDRs and

the AES to just focus on the engaged, kind

of warmed prospects versus all the cold outreach.

So I definitely think it could

be something that could work.

And A common misconception, too, is, as you were

saying it, I could just see some of our

clients or maybe some of the specialists that are

more in the weeds for our clients thinking like,

well, this is going to replace me.

And that's a huge common misconception that it's

like putting on an Iron man suit, right?

It just helps you do what you do

better, you can still generate more results.

And just to kind of take a step back when we

talk about the main problem we're seeing, and if somebody can

solve this AI, they're really going to be golden.

But what most of our clients

are really looking for is attribution.

I mean, they really want to know how can

I show my worth as how can marketing show

its worth, especially to get more budget sales?

The same thing. Preach.

It's that battle of like, wait,

no, I sourced this lead.

No, well, this lead came from a webinar,

so it's that ongoing back and forth that's

never happened here, like last week.

Yeah, it's one that's going to happen this

week, but that's the most common thing.

We deal with attribution and we do leverage automation

to kind of give the confidence in where things

are sourcing from that influence and attribution.

But if AI is eventually able to really concretely define

a proper lead source, I think that's something that would

really be a great solve for a lot of companies.

It reminds me of the multi,

it's new functionality in HubSpot.

Y'all might beat me to it.

The multi revenue, multi touch attribution.

Yeah, that thing. Right?

It's like the idea that.

Well, it's not just one potentially. Right.

It might be multiple because they came to the

conference last year, Justin had a phone call with

them and then Mel sent them an email. Right.

So it's that balance of, well, it's

not just one place, but it's multiple.

It's the things that I did for years in spreadsheets.

Right.

And you're looking at all the different touches and

you're trying to show, well, we got in front

of this person 28 times before they raised their

hand, or we closed the sale.

And then being able to automate that or bring it

into a technology platform that at least you've got it

in one spot, but then surface it in a timely

manner in a way that the team trusts and doesn't

go, no, I think they came from this other source.

I mean, that's a gut shot.

It's like we all did our part and it's not

about like, well, this is a marketing did this thing

at the end of the day, even though.

Even if it came through an organic web search, our sales

team had to crush know before it actually got closed.

So it's a one team effort all

around, but absolutely, that resonates with me.

Justin, what you're saying about the attribution stuff,

for sure, Justin, where else do you see

people asking for things, or you guys are

seeing some success in utilizing AI? Sure. Yeah.

Another kind of, another note on that one is

we actually have a webinar coming up just around

event attribution, because that's been another common one.

Just how do I attribute the success of an event?

Because events, they're still around, right?

And even if, whether it's virtual or in

person, I think that's something people are still

struggling to grasp, the success of an event.

What's the ROI from that event?

So I just wanted to throw that out there too.

As we talk about attribution, that's a very

hot topic for us in terms of AI.

I think some other benefits

I see is really documentation.

I think the two things that are value

adds the agency should maybe consider leveraging AI

more for is documentation is one.

I mean documentation, documentation, documentation.

If you can overdocument.

I say overdocument because it's probably more than

the majority of us did before AI.

But if you can really document out everything

that you're doing, what I've seen is it

really puts our clients at ease with having

the confidence to take on these new initiatives.

I would notice before we make recommendations and maybe

be really thorough, we talk about it, everyone's excited,

but they'd kind of fall to the wayside.

It's because they weren't clearly documented in a way

that the company can kind of take, because the

biggest pitfall to any optimization or any processes is

like not training the team on it.

I would liken it to giving a

toddler the keys to a lambo. Right.

You can build out this great automation, but if

the team isn't trained on how to use it,

you're going to have this really expensive platform, but

you're not really getting the most out of it.

So leverage AI to speak to the unique needs

and tailor it to the business model of the

client you're dealing with, or your own internal business.

But document, document, document everything.

And then even if it's something that, hey, we

may refer back to this, we may not.

It's a huge value add for the client.

I think alternatively, another thing we

leveraged AI for is compliance.

We work in some heavily regulated industries and

they really need to be compliant in terms

of their emails in terms of their messaging.

So that's another thing you can actually leverage

AI for is having it scanned communications.

Is this canspan compliant, is this CCPA compliant?

And it'll actually call out red flags and

help you just, and even if it's something

that's not on your client's radar, putting it

on their radar does add value for them.

Then it may be something that you help get

on their radar if it was never there.

So I think those are two.

And then the low hanging fruit obviously is, we

all know, is content and SEO optimization, right?

You can run a full site through AI and say,

hey, this is the keyword I want to focus on,

tailor this content to focus on that particular key phrase.

So that's like another area of opportunity we really

leverage AI for is just enhancing content and SEO.

I love that you bring up documentation

from the customer facing side of things.

We've had a ton of conversation internally recently

around using it to document our processes.

We actually did an episode recently on

the days of companies keeping their internal

process in like a three ring binder.

At least I remember those days.

And by the time you get that thing printed,

hole punched and on the shelf, it's already changed.

Obviously, that's a long time ago.

But the fact that there are things that we do,

especially growing scaling companies, if there's even just at its

simplest form, if, you know, if it's up in someone's

head, you can get a Zoom transcript or something of

this is how the process works and run that through

and at least have some way, that's one step.

But then you start to actually automate some of

these things and then it learns off of it

and it adapts to your process changes. That's huge.

And what we're working on right now is trying to

bring multiple data sources together on our own AI chatbot,

so that employees can basically query from one user interface,

but it's pulling from Salesforce and the intranet and HubSpot

based on their levels of access, right?

So they don't want to ask the boss how many days

of vacation they get or what are the company holidays?

Because they should have known that because they learned

that in onboarding, but they forgot where the link

is to the Internet to access the policy.

Now they just open up the chat bot and

say, how many days of vacation do I have?

And then on the back end, we're hoping to at

some point, again with the right level of access, give

insight to managers or some kind of director level folks

around what types of questions are people even asking, and

how can we surface that information earlier in the process

or make it more accessible to employees if those are

things that they should know? Right. Yeah, that's great.

Justin, I'm going to put on my audience hat

here for a second and I'd love for you

to explain what you were talking about.

Like, I'm five.

Well, probably because I don't understand it,

but I imagine that the audience is

like, well, that documentation sounds great.

I love documentation.

It's so great.

But what does that look like in five

year old terms or ten year old terms?

First you have a meeting where

you talk about the process.

Then you take the transcript, right.

And you pump it into AI.

And then the AI does the documentation.

Walk us through what that looks like, because I

imagine most of the people that listen to this

are going to be like, hey, that sounds wonderful.

I'm going to call Justin up and have him do it for me.

But some people are going to be like, well, I want to

do that myself, but I don't even know where to begin.

What does that look like? Sure. Yeah.

I think Step one is build your document library.

And the reason I say that is because as

we talk about solving problems for clients, right.

It's probably more likely than not that you're going

to come across similar problems more often than not.

And what that documentation helps, it helps you

ensure that you're delivering in a consistent way.

So I'll give one example, lead scoring, right.

That's something you're all

familiar with, lead scoring.

How do you show how interested

someone is in your brand?

Well, if we use documentation to kind of

build out a lead scoring table, again, show

the justifications of why it's important.

Well, as that example, when you come across

another client that needs lead scoring, you have

that foundation to work from and all you're

doing is tweaking it to their particular needs.

And what it looks like from a

client perspective is now we shared drives. Right?

So you have a shared drive where your

client can access all of these documentations that

really, one, in a way outlines. Wow.

The automation company has done this much for

us because in the other justification, it's like,

hey, we want you guys to know this.

We want to empower you.

We don't want to be so hooked into your business

that you need to call us to change everything.

Our ultimate goal is we want to train

you to be able to be better.

And I think step one of any good

automation strategy is identifying those bottlenecks and training.

Right.

So one, we have to identify what are the bottlenecks.

Let's automate those processes.

But we need to document it so that we don't

revert back to doing it the way we were before.

But then secondly, again, it's a value add.

I'm the client.

I now have a full document library

of all the things we're proposing.

I'm just going to feel a lot more confident.

And as the agency that's developing that

documentation, it's as simple as detailing out

what you're going to do.

What is your recommended solution? Lay that out.

And again, it can be shorthand, it

doesn't need to be paragraph form.

Lay it out and just deliver that prompt

and ask for it in a form of

a documentation that an enterprise client would expect.

This needs to leave no room for question.

Personally, when I use AI tools, I give it those kind

of prompts just to kind of explore the different ways.

And I've noticed that again, we're a smaller agency just

in the term of our language and the way our

documentation is laid out, it feels much more enterprise now,

which would give me confidence in the agency that I'm

working with that they really do this all the time.

Right?

And even if it's your first time, maybe coming

up with it, you have the solution, but how

do you articulate it in a way that gives

the client confidence that, hey, we have already thought

about all these hiccups along the way.

Here are the common challenges, and that's

where AI comes into play is because

it's adapting in real time, right?

So simply do what you were going to do, but lay it

out in a way and then provide us AI and just, hey,

layer in anything else you think that would be valuable here.

And that's essentially what you're

getting with your document library.

And the major takeaway for you as an agency is

you now have this repository of documentation, Wiki if you

will, but you can always refer back to it.

Mel, I love what you were describing in that.

I would find it great.

I was kind of thinking of it as

you described it for our service team, right?

They come across these common

questions around attribution, lifecycle stages.

It would be great if they could just kind of type

it into a prompt and just here are huge FAQ document,

bring up everything that has lifecycle stages and they can kind

of find the right answer in real time that way.

So I think that's just another benefit of documentation,

is just being able to detail out what exactly

you could potentially do for a client.

And it just goes back to confidence again.

You want to just give the client confidence.

You know exactly what you're doing.

That's a great takeaway.

A nugget here that I think with a lot

of what people are doing is mostly chatbot related.

Right?

Like I haven't figured out how to do the

API stuff yet, but I'm still typing in prompts.

One thing you hit on that you kind of maybe

glazed over a little bit, but I think is really

key is this idea that you can pre prompt it. Right?

Like I can take the template, highlight it all,

copy it, paste it, and depending on what chat

bot you're doing, some call it system context.

You could just, hey, here's the template.

Here's the things that I picked

up in my meeting, right?

Fill out the template that you pasted in with

the additional data that you think goes in here,

and then it will fill out the template.

Right, right.

And I love, it's such a simple, I

can't believe that we hadn't talked about it.

Maybe you're doing this, but Justin's recommendation around

leave no room for questions justify this.

I haven't done that in any

of the stuff I've been doing. So I love that.

That's a huge takeaway for me.

Big light bulb moment.

What you're talking about sounds really enticing.

Like, I imagine that whoever listens to this

is probably going to look you up.

Curious to tell us about your

company, the services that you offer.

How can I get in touch with you?

How do I get in touch with people that

can sell me what you guys are doing?

Tell us all about that, because

I'm actually kind of interested.

Yeah, me personally, not on Instagram,

not on TikTok or anything.

I think I still got some moves, but

LinkedIn is kind of where I live, so

feel free to connect with me on LinkedIn.

I'm always happy to have the conversation.

I truly love it.

I love talking about it all day.

So this was great to just get to talk

about it and nerd out a little bit.

So thank you guys for that too.

But the automation company,

our website is automationco.com.

We put out a lot of thought leadership

around automation, specifically around the different functions of

Rev ops, marketing, sales and service.

At our core, we're a HubSpot shop, but

we do quite a bit with Salesforce.

We really strongly believe in both of those platforms,

but in a sense we're platform agnostic in that

if it's a problem, we're happy to solve it.

Integrate into your embed into your flow, your

process, the automation company, who we are and

what we do is we help companies, simply

put, streamline their marketing, sales and service and

then optimize and automate those processes.

So whether you're a company that's a million

in revenue to a billion in revenue, we

like to look at automating your Rev ops.

Like Wifi.

You don't need to be a Fortune

500 to need good internet connection.

So again, it's leveling the playing field.

It's putting your marketing, sales and

service on steroids, the legal kind.

But I say that because it's almost an unfair playing

field when you have a company using automation versus a

company that's not, and they're both trying to scale.

Automation is just that Iron man too for you.

So we help you just get equipped to really be

able to compete at the best of your abilities.

I love it.

I do too.

I'm going to call you up and

see what we can do together.

But I love that thought of like this is a

level playing field, but you need to take advantage of

the tools and the platforms and that automation because if

you don't, you're probably already getting behind and at some

point it's going to be too late.

I think we talk about that.

We've talked about.

Well, Justin, thank you so much for reaching out.

This has been an awesome conversation.

We've learned a lot.

We're going to keep this going offline.

Otherwise, y'all get in touch with Justin, like

you said, connect with him on LinkedIn.

Justin Crawford, the automation company,

go to their website.

We'll drop some information in the

episode show notes as well.

But thank you so much for your time, Justin.

Really look forward to continuing the conversation offline

and in the meantime, keep it automated.