I Started an AI Consulting Business

Jon Law
434 followers
Setting up an AI consulting business in this video!
Other channels:
@EmprendiendoconJonLaw
@jonlawclips
@EmpreendendocomJonLaw
Note: I do not sell or advertise any products of my own. Watch out for scams!
#makemoneyonline #aibusiness #passiveincome
Check out the channel for more videos on investing, economics, & entrepreneurship!
Queries: [email protected]
#business #productivity #money #economics #entrepreneurship
So recently I made a video about trends I was seeing post one month of 2025.
And I laid out a specific business model in that video that, in my opinion, on every level
works as one of or the single best side hustle or business or space to get into this year.
And that model is AI consulting.
Now what I mean by AI consulting is helping businesses retool or switch out their old
software processes and strategies.
With the new generation of AI powered tech that we're seeing cut costs and increase productivity
across the board.
So after making that video I thought, well, if it's that good, if I say it's that good,
why don't I just actually do it?
That is this video.
As an experiment, just like these other experiments on the channel recently, and you should subscribe
to see more, I'm going to actually start an AI consulting business in the real world
and walk you through the whole process to take you through the failures, the setup and
the success.
If we have some, I don't know yet.
So in this video we'll go through my additional approach, the actual specific setup and everything
involved in getting the business rolling.
And then the real world initial results from our first outreach campaign.
Now delivery or doing the service here is the easy part, at least in my case.
I'm an AI nerd and I've worked in and with a lot of different businesses to identify
areas for AI driven improvements and also really just identify outdated workflows, processes
and areas for further automation and efficiency improvement.
And to scale delivery I think is similarly simple, not necessarily easy, but it is simple
in that you go out there, you find AI savvy young people and you pay them well to consult
for other businesses.
So instead the hard part about this business will critically be the sales and the sales
approach.
And that is the key differentiator that we're going to employ off the bat.
Instead of charging up front for the service we're providing, what I'm going to do is offer
to work completely for free for the client.
And I'll go into their business, I'll talk to employees, look at ad accounts and CRMs
and the like and benchmark out a list of areas and specific workflows where I think
AI can improve the business and also specifically how and what AI tools or systems or scripts
could remedy those areas for improvement.
And I will deliver that report again entirely free to the client.
They can go work with someone else to go about actually doing it.
But the key is I will then offer to do that, to do all those things for them for a certain
price.
And so I'm actually charging just to materialize the improvements that myself and the client
after digging through the business know are there.
And this has a few benefits.
First it just gets us in the door.
It's an easier way to get people on to sales meetings.
And there's only upside for the client because they could just not buy the report or buy
the actual implementation and just take the report and work on it internally.
And it also very clearly sells itself because reading the report is a list of their specific
pain points and very material ways in which they could make more money or solve their
own just problems and save themselves time and effort in very specific segmented out
X, Y, Z kind of ways.
And so I'm betting that enough people will choose to pay to actually get and materialize
those benefits to make the time that I spend with other clients building reports worthwhile.
And this also ties into the demographic that we're targeting because we're focusing on
businesses with more money that can pay more to justify the cost of some of those free
reports.
So that is the approach.
Now let's get into the actual infrastructure or the components of the business.
Now for any business you need a digital home, a website like the one I'm purchasing literally
right now for this business.
Now a great domain extension choice for any business, but especially digital and AI consulting
businesses like the one we're building here is dot online.
Dot online is the world's second largest new domain extension and searches with the word
dot online are searched on browsers over half a billion times every single month.
And it's used by thousands of popular brands and celebrities and entrepreneurs like these.
Like I said, dot online domains are perfect for all kinds of online businesses and what's
great about them is that while dot com domains are super saturated and let's be honest here,
you can never find the one you actually want.
There's a much larger inventory of dot online domains available that make it easy to find
the perfect one that fits your brand and business.
So for that reason, head over to get dot online or click the link in the description and use
code JOHNLAW, all caps, to claim a limited time offer of just 99 cents for the first
year for that dot online domain for this AI consulting business.
That said, the major infrastructure component of this business is instantly.ai, which is
an email platform that I'm entirely not affiliated with, but I do love and use all the time.
I'll use this to cold email potential clients and book them into sales meetings where hopefully
some of those will move forward into real deals.
So to set it all up on instantly, I bought these two primary domains for the business
and then created a email addresses to make sure emails don't go to spam.
You don't want to send roughly more than 30 emails per day per address.
So 30 times eight gives us a sending potential of 240 emails per day or roughly 5000 per
work month.
If we can close even one out of those 5000 per month at 10 or 20,000 per, then that's
great money right there.
But as always, it is a lot easier said than done with these things.
So then instantly I created a campaign and wrote this email sequence with four emails
sent over one and a half weeks or so, along with some headline variations to test what
works better.
After we get an initial performance baseline, I will go hardcore into testing all kinds
of scripts and copy.
But for now, we're less worried about optimization and more about validating the concept through
initial replies to the emails and booking.
So the last element here is leads.
And other than that, we are pretty much all set up as far as the infrastructure work goes.
I'll use Google Calendar to send out Google Meet invites for the sales meetings, and that's
easy and free.
And I did spend two weeks also on instantly warming these addresses up, which again, make
sure they don't go to spend.
So like I said, the last component here is leads.
And I did that with Apollo.
Now, I honestly have no clue what industry or demographic is going to work best as far
as sales conversions here goes.
And also AI retooling and consulting really works across most businesses that have some
kind of digital presence.
So I'm starting by throwing all kinds of leads into the mix.
And we'll just see what works and iterate based off those initial conversions.
If I was really doing it properly, I'd set up 50 or 60 different campaigns, each with
their own customer lead list and demographic, and then split test based off that.
But we're just going to take the 80-20 approach here.
So generally, I'm targeting C-suite executives of businesses doing at least $10 million a
year in pretty much every industry.
And again, in some, we'll see what works and what doesn't as far as leads go.
So that said, I've been waiting a few weeks for this.
Now I get to go and launch those campaigns.
So let's jump into instantly and do that.
So I have all my leads right here, and then I'll go into options.
And so we have our account selected here.
So this is where people will get emails from.
Those are the accounts we just warmed up over a period of two weeks.
Then we're sending emails only as HTML to help with deliverability.
And there are some changes we can make here.
So I can set the minimum time to seven minutes and three minutes as a variable component.
And otherwise, we are pretty good to go here.
And we're using pretty straightforward subject lines here as well as scripts.
And I do use some social proof from the last few years of work that I've done here as well.
So now I just click save and then launch.
And boom, we are live.
So now I'll wait a few days.
And then in the analytics tab here, we'll go over initial results.
Okay, so we're back for the update.
It's been a few days, I think four days of the campaigns now.
And we have 33 responses in the inbox.
Unfortunately, it turns out this is a really, really bad week.
Or this was a really bad week to send out emails because a lot of people are on spring break.
So I think we're going to get a lot more data next week.
And I'm also creating an out of office, an OOO sequence that will automatically put people
who respond with an out of office notification into that sequence so we can follow up later.
Now we got about five or six negative responses here.
And then one, it looks like at the top here, positive response.
I'm going to have to filter blur a lot of this out just because it is actual,
you know, user data.
But this response is basically just I want a meeting.
So that's the first one.
That's a really good sign and confirming indicator.
So we'll get on a meeting with this lead.
And I'll see how that goes.
And as for these others, they're all basically saying,
hey, if we do this in house, which is to be expected, a lot of people are retooling.
And if we go over to the Unibox here, you can see some of those responses.
Although again, I apologize, I will have to blur this.
So then here are the analytics.
This is not at all a great reply rate or anything.
So there is a lot of optimization work to do here.
But I think as well, next week will be better,
especially considering, again, people won't be out of office as much.
Potentially the week after as well, just whenever kind of the spring break
season for traveling with kids ends.
So looking good so far.
Again, you know, validation is the hardest part.
And we got one.
And if we get one, then we can just scale it from there,
assuming that contributes or converts well on the back end.
So that's this video.
In the next video, we'll go over results,
you know, a few weeks out as we really look to scale this funnel
and business to the first $110,000, $50,000 of revenue.
So thanks for watching and I'll see you in the next one.
Cheers.
