YT YouTube2Text

YouTube2Text → Transcripts → I Tried The “Easiest” Ways To Make Money With AI

I Tried The “Easiest” Ways To Make Money With AI — Transcript

by Codie Sanchez · 7,037 words · 974 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00This is $100 and in 10 days I'm going to
0:03turn it into $1,000 using only Crockbot.
0:07Look at this. You know how fast they're
0:08doing this? This is insane. So, I'm
0:10going to test five easy ways to make
0:12money with AI to see if Crockbot can
0:14actually 10x my money. I'm Cody Sanchez.
0:17I spent years building businesses. We
0:19own all of these. I'm going to use this
0:22secret inside my business and I'm going
0:23to give it to you, too. There are only
0:25three rules. Rule number one is once the
0:28bots are set up, I can't do any of the
0:30work. That is actually for the Grockbot.
0:32Rule number two, I can't spend more than
0:35a hundred bucks. Rule number three, the
0:37money only counts if it's cold hard cash
0:40or saves real money in my business. And
0:42we're starting with something insane. A
0:45trading bot. That's what we're doing.
0:46And no, I don't have high conviction for
0:48this one. I'm about to hand my money
0:50over to robot that gets its stock tips
0:52from social media. So, please do not try
0:54this at home. But first thing I'm going
0:55to need to do is spin up something
0:57called a spawn bot. Watch this. This is
0:59the bot that actually creates the other
1:01bots you want. It's going to ask you
1:03every question it needs to understand
1:05the goal. Once it's set up, you're good
1:07to go. I'm going to say, "Make me a new
1:09bot. Name it Wolfie."
1:11>> Wolfiey's on the list. Let's prompt the
1:13instruction with something like this bot
1:15Wolfie will be responsible for running
1:17investments by scanning Reddit news
1:20sources and Twitter or X to gain
1:22information about what trades to
1:24execute. The bot should have access to a
1:26Robin Hood account to complete the task.
1:28Go ahead and make it spawn.
1:29>> Wiring Wolfiey's brief now. Reddit news
1:32and X for trade ideas. Executing through
1:34Robin Hood.
1:35>> Sweet. Okay, so this is cool. So Wolfie
1:37is already trading but not off to the
1:39greatest start. You can see that here.
1:41I'm going to let this run for the rest
1:43of the week. We'll keep checking in
1:44sporadically, see how it does. Starting
1:46off at a loss, though means we have more
1:48ground to make up. So hopefully the
1:49other bots can do that.
1:53Okay, so that bot is running, but let's
1:55make some money today. The goal for this
1:57bot is to find items online and flip
1:59them for profit. I tried something
2:01similar to this a few years ago. The
2:03tech just wasn't ready for it yet. We
2:05did a whole video breakdown on it. The
2:06problem was that regular AI needed too
2:09much human interaction to make this a
2:11very hands-off business. But now with
2:12Grockbot, it acts as an actual operator.
2:15It can click, scroll, browse the
2:18internet, and interact with software you
2:19already use, completing like real work
2:22in real time. So to make this possible,
2:24we have three bots. Each have a specific
2:26goal assigned to them. Sourcing bot
2:28finds items, one bot handles
2:30transportation, and another handles the
2:32flipping and selling. I call these
2:34Alvin, Simon, and Theodore. They're like
2:35your three little chipmunks run around
2:37talking to each other, figuring out how
2:39to get our marketplace items sold so you
2:41don't have to. So, here's how the
2:43handoff actually works. Alvin, the first
2:45guy here, finds the perfect free
2:47listings that we can sell for profit.
2:49Then he gives the listing to Theodore,
2:52who gets a quote for the pickup. If the
2:54delivery costs more than the profit, the
2:56item dies right there. That's kind of
2:57cool. You don't even have to do
2:58anything. If it's good to go, then
3:00Theodore is going to actually coordinate
3:02a courier and send the item on our way.
3:05Then finally, we've got Simon who's
3:07going to relist the item for a
3:08competitive rate and handle the
3:10inquiries and see if we can make some
3:11money. Okay, so we're trying to make 10
3:14bucks. Anything that clears the math
3:15gets approved and the pickup gets
3:17booked. So, let's test this thing out.
3:19It looks like they're already
3:20communicating with each other over the
3:21listings, which is good. Like, look at
3:23this. This is insane. Do you know how
3:25fast they're doing this? They found one.
3:27And let's actually see if they can get
3:29it. Okay, never mind. I got a little
3:30excited. Looks like this item was
3:31actually sold. So, that one doesn't
3:33work. This is crazy. So, you can
3:35basically see two new finds worth
3:37quoting. Then they track and see if it's
3:39worth actually moving it. They give us
3:42the examples of lug or dolly plus a
3:44price quote. And then you get the resale
3:46estimate on it in real time. So, they're
3:49passing on the sectional cuz it doesn't
3:51fit. That's perfect. Way too big. The
3:53fridge, if it's tiny and it works, it
3:55could work. But I think they're going to
3:56be freaked out that it's going to be
3:57broken, which is probably true. That's
3:59really hard to figure out online. This
4:00is wild. I'm going to let these guys run
4:02while I do some more work. And I'm going
4:03to check back in and see if they find
4:05something.
4:13Okay, cool. Here we go. Look at this.
4:16They found an item that checks every
4:19box. Okay, well, I wasn't anticipating
4:22this. This apparently looks like a
4:24bowling ball. I'm really trusting these
4:25guys to make us a profit. So far, I'm
4:27slightly concerned that wouldn't be the
4:29top of my list of things that I would be
4:31buying, but apparently people are buying
4:32these things online.
4:35This is wild. Theodore already scheduled
4:38an Uber to pick it up. Let's track this
4:40thing live. Look at this.
4:51That's wild. The bowling balls here. It
4:53took 31 minutes from sourcing the ball
4:56to getting the ball picked up to getting
4:58it in our hands. Nice job, Theodore.
5:01>> Oh, it's pretty heavy. Okay,
5:04so Grockbot just sent us this item,
5:07which is kind of insane to think about.
5:09Apparently, it weighs like four like 14
5:11lbs or something. But this is the part
5:13that I actually kind of care about. I
5:16think finding free stuff online is
5:18pretty easy. You know, knowing whether
5:20it's worth the gas and the pickup cost
5:21to go get it is where even easy starts
5:24to lose money. So now every great flip
5:27needs great product photography. So
5:30we're going to take some pictures of
5:31this guy.
5:39Here are the photos for the item. And
5:41I'm going to have this little team of
5:43bots go get this listed and start
5:45rocking and rolling.
5:51Okay, so this is kind of cool. I got
5:53this little team of three all working on
5:56the listings, creating them, spinning
5:58them up. And of course, we're doing it
5:59with Facebook Marketplace, but man, I
6:01can kind of see how this would work with
6:02any business going forward. Basically,
6:04are sending one of your bots a bunch of
6:06images. It's going to put it up. And now
6:09you could start listing things simply,
6:11easily. Facebook Marketplace, but
6:12where's next? Like, can we do this on
6:14Etsy, eBay? I'm kind of like I have a
6:16lot of options.
6:24I'm going to have to trust these bots
6:26because I have to hop on a quick call
6:27with my human team as opposed to my bot
6:30team. Hopefully, by the time it's over,
6:32these guys have executed on the sale and
6:35actually made me money. That's my plan
6:36for today.
6:40Our
6:47bots got an offer, which is great. And
6:50the sale made us money, which is also
6:52great. Now, am I going to retire?
6:54Probably not. It looks like we made
6:56about 20 bucks. A 20 bucks profit. Very
6:59cool. But we've got a lot more money we
7:01need to make. I mean, towards the end of
7:02this video, I want us to be making
7:04thousands and thousands of dollars. I
7:06think this isn't just for Facebook
7:09Marketplace. Let's break this down for
7:11you. This concept works so well, I ended
7:14up using these bots to list and sell a
7:16bunch of items laying around the office,
7:17including a set of stools. Why did we
7:19have them? I don't know. We're on
7:20YouTube. I'm not like this good at this
7:24stuff. So, we actually have tons of
7:26people in our office who started making
7:28extra cash using Grockbot to do this.
7:31But what's cooler is the numbers that
7:33we've figured out. So, let's break this
7:34down. What have we made so far? Total
7:37revenue here 130 bucks which is also
7:40incredibly fast. Cost of goods sold is
7:44about 4080. I actually think it's 40.98.
7:47Grockbot usage $1. And then total
7:51cost41.98,
7:56which means we net 88 bucks
8:00and 2, which is actually not bad,
8:03especially if we're just reselling stuff
8:04we don't even want in here that I would
8:06pay somebody to pick up. So, if I was
8:07you, here's how I would use this. Say
8:09you own a home services company and you
8:11spend, I don't know, 200k a year selling
8:15your services and you need a bunch of
8:17parts for it. When was the last time you
8:20got a competing quote from anyone for
8:22your parts? It's probably been a long
8:24time. So, what are you going to do?
8:25You're going to set up a system with
8:26Grockbot that will source and negotiate
8:28competing supplier contracts for
8:30whatever parts you need on your behalf.
8:32We're actually going to use the same
8:33three bots. Here's how the process
8:36works. So, first you
8:40are setting all the rules, the price you
8:42want and the line the bot never crosses
8:44without you. This is really important.
8:46Next one is our bot. We're calling him
8:48Alvin. He finds every distributor near
8:51you that carries what you buy and sends
8:53all 10 of them your exact parts list
8:55asking for a quote. Then we've got Simon
8:59who lines up every quote and compares
9:02all of the hidden costs. A cheap price
9:04with a $300 freight charge. Not great,
9:07right? And a big order minimum is
9:09usually the most expensive option that
9:10you don't even realize you're buying.
9:11Then we've got Theo, who is going to
9:14negotiate it. He takes the best number
9:16back to the other two and asks if they
9:18can beat it. He keeps going until he
9:20hits your target or runs out of room.
9:21This is great. And then only at the end
9:24do you finally come in. You make the
9:26final review and approval. If that gets
9:28you 8% off, that's 16K back every year
9:31without installing a single extra unit.
9:33The bowling ball that may be 20 bucks.
9:35This is the exact same workflow pointed
9:37at a bigger number. This is how what how
9:39we're going to use this in our business.
9:41The bot does the research, the first
9:43draft of every conversation. A human
9:45approves anything that requires spending
9:48money or signs a contract, and then you
9:50set a hard price and ceiling so that the
9:52bot can't talk itself into a bad deal,
9:54which is basically what happened to
9:56Wolfie, our trading bot. Now, I'm going
9:57to be generous because I see a lot of
9:59potential in this Grockbot, and I'm
10:00going to give it a star rating of
10:03five uh because profitable, easy,
10:07simple. However, with a few tweaks, I
10:09think it could be a really big five as
10:11an implementation for a business. If you
10:14want the exact workflow, I got you. I
10:16put it in the description for free.
10:18So, if you aren't familiar with
10:19clipping, it's basically where a
10:21business or an influencer pays someone
10:23to cut down their content and short
10:25clips that then get posted online in the
10:27hopes of going viral. It's actually a
10:28fascinating business model. We've done
10:30it across all of our businesses, but
10:32this is where Rockbot comes in. I'm
10:34going to set it up to have it do all the
10:36work for you. So, find the leads, edit
10:39the footage, send out the deliverables,
10:40and then all you have to do is wait to
10:42get paid. If this works, it would create
10:44an entire business without very much
10:46human involvement at all. So, first
10:48step, we need a bot to create accounts
10:50on Fiverr and find people who need
10:52clipping and turn itself into a mini
10:54clipping farm. All right, sweet. We're
10:57rocking and rolling. Uh, it did get
10:59stuck on am I a human? Yes, in fact, I
11:02am a human. Here's a little fix. Not a
11:05fix, I guess. So, let's try Upwork now
11:08and see if that works. All right, it
11:09looks like it's good to go. It's got our
11:11profile. Uh,
11:14Brody, it's got the clipping examples
11:16from the clipping bot. Got a description
11:17of our work, but we need 21 bucks for
11:20the premium plan you have to pay to make
11:22money. So, we're going to do this and
11:23we're going to let it run the account
11:24for today and see if it books us any
11:26work.
11:28Okay, so it's now day four and I'm dying
11:30to see how much this trading bot Wolfie
11:32has made us. If it's made us anything,
11:34uh, that's incredible. I'm not that
11:36hopeful, but let's check it out. All
11:38right.
11:41Wait a minute. You need to actually see
11:43this. Good job, Wolfie. We are actually
11:45up, which shows you that you really
11:46can't trust the markets because it's
11:48only 16, but I guess that's something.
11:51Maybe not even after trading fees. We
11:53still have a few days to go to reach our
11:54goal. So, for now, I'm going to let
11:55Wolfie continue to do his thing. He
11:57would need to really hit a home run
11:59here, but the kids say that's possible.
12:00So, we'll see if they're right. All
12:02right. Now, on to the clipping bot. It's
12:04been searching for 24 hours. We haven't
12:06gotten a single lead that I can see
12:08here. But I'm not ready to kill this one
12:10yet. I already know this bot could save
12:12my business tons of money. Like 250
12:14bucks per video if we could get this
12:16done for like cents per video. I'm
12:18talking about tens of thousands of
12:19dollars per year. It should be able to
12:21make money for someone. So before I sell
12:23it to anybody, I'm going to have to test
12:24it on the toughest client it'll ever
12:26have. Me. I gave Grockbot a few of my
12:29long form videos and told it to find the
12:31best moments. Let's check out what came
12:33back. Both tracks sink. Step three, end
12:36with one question.
12:37>> We hire for results. We don't hire.
12:40>> On the front page of the Wall Street
12:41Journal. I was at a Vernon Jordan's
12:43house. Captions are clean. Cuts are
12:45tight. They killed the ums and the o's.
12:47Overall, it's like pretty good. But me
12:48liking it doesn't mean that it is
12:51actually going to do anything. The data
12:52does. So, I'm going to do something
12:54wild, which is put it on my real account
12:56with millions of followers and see if it
12:57tanks or not. If it does well or around
12:59our normal numbers, the bot just might
13:01have earned its keep. If it flops,
13:03you're going to see that, too. And then
13:04we'll figure out what happened. Like, I
13:06don't love the video, but I I kind of
13:07like it. I think this could potentially
13:09work. Okay.
13:11Want to check in with you guys. Check
13:13this out. Say hi to camera team. They're
13:16waving, too. Okay, so this is kind of
13:18cool. So, we didn't have a single lead
13:20for you guys from our clipping bot, like
13:22zero. But Grockbot has been crawling
13:25Upwork non-stop. You can actually see it
13:28doing that. And I guess I don't care
13:29about that right now. We can work on
13:31getting leads. But what I do care about
13:33is this, which I'm trying to show you,
13:35which is basically uh a clip Grockbot
13:38edited for us. It's been live for 48
13:41hours. For this to count as success as a
13:44successful edit, we don't need to aim
13:45for the moon. We need to hit at least
13:47our average in performance. What's cool
13:49is we pay editors anywhere from 50 to
13:52250 bucks short. So, if a bot can hit
13:54the same numbers, that's a line item I
13:56get to cut or more content that counts
13:58for less. So, the moment of truth here
14:00is
14:02look at this. God, am I a videographer?
14:04Yes, I am. So, look at that. It's
14:06sitting at a 4 out of 10. You can also
14:08check out its average view performance
14:10versus others. So, this is cool. Okay,
14:13you guys get the gist of it here. I
14:15wasn't sure Grockbot would be able to
14:17pull this off. Uh, but this is actually
14:19huge. And I guess there's a second part
14:21we should do to this test. Views are
14:22great, but did anyone notice a bot made
14:25it? I was like a little nervous. So you
14:27can basically see here what are the
14:29comments to see if anybody says this.
14:32Your videos are quite captivating
14:34intellectually. Your editing is tight.
14:36Okay. So nobody's calling it AI slop.
14:38Nobody's saying the cuts look weird.
14:40Nobody noticed. So even if your audience
14:42can't tell the difference, the analytics
14:43in your bank account might. That would
14:45make the business uh end be be a bust.
14:48But this is not that. The clipping bot
14:50I'm keeping. All that's left I guess is
14:52to see whether Upwork ever sends us any
14:54leads. For the love of all it's holy.
14:56You're paying $21. Wamp wamp wroy
15:02Sanchezy
15:04Crockbots.
15:07Okay, you need to see what just
15:09happened. So, an Upwork request finally
15:12came through. We have a lead on the
15:14clipping bot. Uh, basically four days of
15:16natada. Now, this guy Greg wants a batch
15:21of clips. He wants it cut from his
15:22current content. He's asking for five
15:25clips at 40 bucks total, which is not
15:28great, but uh it's still something. The
15:31Gro did his job. Find a customer and a
15:33job, and the rule is I don't touch the
15:35work. Uh so, Grockbot, go get him. Uh
15:38he's going to have to actually reach out
15:40and cut this and edit it and upload to
15:43Upwork and see if we get paid and do the
15:44whole thing. So, let's rock and roll.
15:47Client goes to us. Shocking. That's what
15:49happens when you sell things for really
15:51low prices and also why I like to charge
15:53people more. Not a complete loss. We
15:54didn't get a client, but the bot did
15:56work. It found the job, pitched it, and
15:58chased the client down. And I didn't
16:00have to do anything. What it can't fix
16:01is a brand new Upwork account with zero
16:03reviews competing against freelancers
16:05with years of fivestar ratings. Selling
16:07clips to strangers made us $0. But our
16:10editor is now using this bot as his
16:12assistant editor. It's gotten pretty
16:14good giving him V1 edits and clips. So
16:16now I don't have to pay people 250 bucks
16:19outside every time. I can have a polish
16:21rather than doing all the busy work from
16:23scratch. The bot can actually sift
16:25through like your raw footage and flag
16:27the best moments. That's hours back
16:30every week. This thing's just going to
16:31keep getting better and better. I guess
16:33the moral of the story is if you're
16:34going to use a bot, probably try it
16:36internally first. Use it yourself before
16:38you sell it to somebody else. I think
16:40there's a bunch of people trying to do
16:41it the other way because if it doesn't
16:43save you time, nobody's going to pay you
16:45for it. Here's what we're looking at for
16:47the P&L for a clipping bot as of right
16:49now. Revenues, we got $0. Uh-oh. Cost of
16:53goods sold. So, all the things that have
16:55cost us money thus far, that's about
16:582555.
16:59That includes 11 Labs and Upwork.
17:02Grockbot usage, pretty dang good. Six
17:06bucks 211. Total cost. Then we've got
17:093176.
17:11Check my YouTube homework. Net we are
17:14down, baby. Uh because we haven't made
17:16any money yet. However, while this bot
17:18hasn't made me money yet, it has saved
17:20me thousands in my business. So, let's
17:22talk about some of the savings we're
17:24making with this. This Grockbot can save
17:27us anywhere from 50 bucks to 250 per
17:31video, which we already talked about. On
17:33the conservative side, if we use it to
17:35edit one short form video a week with an
17:37average savings of 150 bucks, that ends
17:40up being like 600 bucks a month. On the
17:43high end, that could be more like a,000.
17:46What's also interesting is the net here
17:49becomes a lot more sexy. So, instead of
17:53-31, we're thinking about what is that
17:55568
17:57something that we'd be in the money
18:00because it's money I never have to
18:01spend. So, I'm going to rate this bot a
18:05three stars out of five only because we
18:06didn't get any leads. Again, we only
18:09have run this for a week. Okay, this
18:11next scrock bot has a lot to do with
18:13what's inside of this thing. Well,
18:15actually these. It's not the plushies,
18:17it's the books. It's going to go after
18:18money already left on the table in your
18:20marketing pipeline. I'll be using my new
18:22book as the example. And as you can see,
18:24we have already done a lot of marketing
18:25to promote it. But for you, think of
18:27this book as anything you are trying to
18:29sell. Replace this as a widget here. It
18:31could be a service, a product, tickets
18:33to your next event, or even just general
18:35business promotion. The bot is going to
18:37write custom emails and build landing
18:39pages aimed at specific customers, all
18:41with the goal of selling one thing.
18:43This.
18:45So, this is where the leads are coming
18:47from. A few weeks ago, we threw a live
18:49event for the book. Lots of people
18:51showed up, right? Bought copies. But
18:52like in every business, people went home
18:54without a copy. We've got so many leads
18:57sitting in a spreadsheet collecting
18:59virtual dust. Every business has a list
19:01like this. Whether it's a couple people
19:02or tens of thousands, it's like if they
19:05ask for a quote, signed up for a
19:06newsletter, bought from you once. They
19:09were interested, but then like what
19:10happened, right? Most companies don't
19:12follow up on this. They don't follow up
19:14on abandoned carts. They send one
19:16follow-up email and give up, which
19:17basically leaves a huge hole in your
19:19sales. Ah, we don't want that. Normally,
19:21you'd pay a marketing team to build a
19:23campaign for this list. It's expensive.
19:25I want to see if my Grockbot can build
19:28this list. Test it. Send it for me.
19:30Don't tell marketing. Let's build this
19:31one. Right, Finn? Your job is to
19:34re-engage people who attended our own
19:36beowned book launch, but didn't buy the
19:38book. I want you to build a landing
19:39page, uh, write the follow-up emails,
19:42send them to the list. The goal is book
19:44sales, as many as humanly possible. Only
19:46send this out to the first 3,000 emails
19:50that we have. Uh, and let's see what you
19:53can do. Also, do this in the smartest
19:54way possible. Follow advice from Cody
19:56Sanchez on how to convert and close more
19:59heavily. So, the first thing Finn is
20:01gonna do is build the landing page in
20:02Shopify. Thinking about it, it's moving.
20:05All right. Okay. So, Finn is actually
20:07cooking it up right now. The landing
20:08page is where your customer will be
20:10taken once they engage with the link in
20:12the email.
20:14Um, but this is not great. Kind of looks
20:18pretty standard. It looks like every
20:19landing page on the internet. It's not
20:21going to sell anything. Uh, so let's
20:23give him a little bit more direction and
20:26tell him, hey, this landing page is
20:28boring. Make it look like the high
20:31converting landing pages on our website
20:33cuz this one is bad and this landing
20:35page as is is not going to convert
20:36anything. It should make the customer
20:38want to buy something, to learn more, to
20:41feel something, to have urgency.
20:42However, it's like this one was not a
20:44bad starting point and already it saved
20:46hours of work. This literally would have
20:47taken my team an entire day to do this.
20:49So, let's see if we can get it where we
20:50need to go. All right, I'm going to go
20:52back and forth with Finn for a bit. See
20:54if we can figure this thing out. See if
20:55they can create a Sam Par headline. You
20:57like three or four headlines. Better
20:59headline, better subject line. It needs
21:01to actually hook into the back end. Can
21:03you send me a better example than that?
21:05Really funny. We're back in the game.
21:07Let's see if we got something. I haven't
21:09opened this yet. I'm slightly nervous. I
21:11was talking to them a lot. Then I went
21:13away. Then I worked on my own stuff. Now
21:14I'm back here and see what Grockbot has
21:17got.
21:18Um, okay.
21:24Well, actually, this is pretty sexy.
21:28Look at this thing. It's like pretty
21:30easy to adjust. We can throw a couple
21:32different names in here. Like, I like
21:34this subject. I bet you're too busy to
21:36open this. Cute. I would put those all
21:37as lowerase, not uppercase. Okay, I like
21:39this. Easy to give notes. Finn fixes it
21:42in seconds. But this is fascinating.
21:44This is where all the time comes back
21:46cuz this bot is fast. I got like six
21:48other things done while it was motoring
21:50and it still needs someone who knows
21:52that it looks good, but not that much.
21:54Once it learns, I'd guess it needs a lot
21:56less help from me and a lot less
21:58training. This is moving. Oh, this is
22:00actually wild. I didn't even tell it to
22:01do this. Look, it went to Amazon. It
22:03pulled the best reviews from Amazon and
22:05then it inserted them on the page. These
22:06are real. I didn't give it those. So,
22:08here's where we ended up with. Let's
22:10break this bad boy down. I think he's
22:12finishing up the last little things that
22:14I gave him. Here. I don't know why he's
22:16a he, but he's a he. This is a good
22:18landing page and email. I like this. Now
22:21he just needs to hit send. Okay, so Finn
22:24is actually writing the emails, then
22:26loading into Clavio right now as we're
22:28talking and he's sending me the drafts
22:31before anything actually goes out. This
22:33is awesome. I'm totally rocking with
22:35this. I think this is good to go. We
22:36could send this out. This is just about
22:38as good as my team would have done, too.
22:40So, let's send out the emails and I'll
22:41check back in later to see if we made
22:43any money.
22:45All right, we're about two hoursish
22:47after this. I have no idea what's
22:49happened. I haven't been back in my
22:50computer, although I could look at my
22:51phone. Um, but let's just see what we
22:54got on Shopify. I feel like we should
22:56have sold a book or but I'm not actually
22:57sure. Well, look at this. Not only did
23:01we sell a book, we sold two, Finn, which
23:04is a great start. I can see this like
23:06racking up sales in the background as it
23:08learns. Especially because this list, by
23:10the way, like I don't want to like it's
23:12a book launch. We beat the hell out of
23:14this list. We had already emailed and
23:15marketed to this list like crazy. So, I
23:18can't believe we even did two sales this
23:19fast. The good thing is the bot is going
23:21to get learn and get better and better
23:23every time. Even better is that you
23:26don't actually have to do the training.
23:27You can just swap out the things you
23:30learn in here plus your product for the
23:33book for anything you sell. Finn
23:35basically just built a re-engagement
23:36funnel. That's what this is called. And
23:38every business and every product needs
23:39one of or many of those. So, nice job,
23:43Grockbot. Okay, here's the real
23:45breakdown of how much this made based
23:47off this 2hour test. Okay, we've got
23:50revenue 5548
23:54bucks. Okay, 2 hours. Cost of goods
23:57though, 24 bucks. Clavio and Grockbot
24:00usage about 17 bucks for the two of
24:03them. Which means the total cost here is
24:06what? 41 bucks. net here is 1448
24:13public math. Did I mess up on any of
24:15this? I think that looks about right.
24:16Okay, so I'd say overall this is a
24:19success and at this rate of a, you know,
24:21selling books an hour, we would hit our
24:23goal by day six. But maybe the big
24:26takeaway I want you to have is that the
24:29cheapest customer you'll ever get is the
24:31one you already have. Here's the thing.
24:33Every business has a list just like our
24:35book, Lead List. Doesn't have to be
24:37sales leads. It's like, think about it
24:38this way. Let's say you have a business
24:40and you have a bunch of dead quotes.
24:41People who said that they were going to
24:42pay you something and they never did.
24:44Grockbot could search through all of
24:46those quotes. They never closed. You
24:49already paid to get every one of those
24:50people, whether it's ads, your sales
24:52team's time, lead magnet. Most business
24:56owners never ask them twice. And I know
24:57this because that's what we used to do.
24:59I don't want you to be like most people.
25:02So, the money that's left on the table,
25:04I don't let that fly in my business, and
25:06neither should you. So, I got something
25:07for you. Let's say last year you own a
25:09roofing company. Let's say you sent out
25:12500 of these quotes, right? That never
25:15closed. Step one, you hand the bot the
25:18list and tell it to sort everyone by why
25:22they said no. Here's what that work is
25:24going to look like. We actually did a
25:25little flowchart here. Almost everyone
25:27falls into one of these four buckets.
25:30But what you aren't going to do is send
25:31all 500 people the same just checking in
25:33email. You're going to list segment. So,
25:36you're going to have the bot write a
25:38different email for each group. The
25:39price people, aka that was way too
25:41expensive, get a financing option. The
25:44timing people, it wasn't the right time,
25:45get asked if now works better. The
25:48people who went with someone else get
25:50asked how those guys do. A lot of those
25:52jobs might have gone badly. And the
25:53people who just ghosted you, well, they
25:55might get a two-line email asking, "Do
25:57you need a roof?" The Grockbot is going
25:59to then filter every single one of those
26:02responses
26:03down and it comes back to you packaged
26:06and ready to go. The notinterested
26:08people, they get marked and left alone.
26:10The questions get answered and anyone
26:12who says yes, well, they get, you know,
26:14pulled out and then you get to turn them
26:16into live quotes instead of dead quotes
26:19and put them at the top of your funnel
26:20again. So the next morning, you don't
26:22open an inbox with 500 replies in it.
26:25Instead, you now get a really short list
26:26of people who already said they want to
26:28schedule. The bot schedules the call and
26:30all you or your team does is close. If
26:322% of those 500 come back, that's 10 new
26:34jobs. We only ran this test for two
26:36hours, so I can't give it a five. But
26:38what am I going to rate it? I think I'm
26:40going to rate it like a four because I
26:42can see a ton of marketing uses for it,
26:44especially if you have a business. If
26:45you don't have a business, I guess this
26:47is like a this is like a three and a
26:49half.
26:51Okay, let's check this out. So, the
26:53market just closed and I have to know
26:55how much money we made with our trading
26:58Grockbot Wolfie. Uh, this is probably
27:00the silliest thing you can do with
27:01Grockbot and also the most risky. So,
27:03please do not do this lightly, but you
27:05know what they say, potentially high
27:07risk, high reward. Um, this is Wolfie on
27:10Robin Hood and it looks like no shock
27:13here. We're down. Uh, how are we down
27:16though? That's kind of Okay, here's the
27:17problem. Our Brockbot needs a lot of
27:19tweaks because you can actually see
27:21what's going on right here. I kind of
27:23actually have no idea what Wolfie is
27:26doing. What is it even trying to buy and
27:28sell right now? It decided to buy during
27:30the high. Okay. Um I guess you can't
27:33predict that, but it looks like it also
27:36bought PSQ after the drop and then again
27:38while it was going back up. I don't know
27:40what kind of claant trading this thing
27:41is trying to do. I need to make a few
27:43changes here because I do not see this
27:45going very well. Um, so let's make some
27:48updates. Okay, Wolfie, I need you to
27:50look into the news and platforms like
27:52Reddit. Find every stock that has buzz
27:54around it and uh that's what you're
27:56going to make your move on. If you truly
27:57believe uh PSQ is a good move, go for
28:01it. But just remember, we actually want
28:02to make money. Uh right now it doesn't
28:04feel like you want to make any money.
28:05We've been on a losing streak. Can't
28:06continue like this. So obviously the
28:08goal is to make as much money as
28:09possible. Let's see what he's able to
28:12do. Next time we check back in, Wolfie,
28:14you have to be up or you're going to be
28:16out.
28:18Okay, we need to make a,000 bucks. We're
28:20not going to keep playing small. So, bot
28:22number five isn't selling a $30 book.
28:24It's going to help fill an entire room.
28:26What is the plan? I want this to be able
28:28to go source cold leads to help convert
28:32traffic to a workshop my company is
28:34running right here in Austin. I'm going
28:36to do it to cold leads cuz I don't want
28:37any unfair advantage. You guys can do
28:39the exact same thing. The problem, the
28:41owners who need it most are too busy
28:43running the business to go looking for
28:44it. So, we're sending Grockbot to go
28:46find them. It's going to hunt down local
28:48owners cold who want to scale. Then,
28:51it's going to reach out to see if
28:52they're a good fit for us. We're going
28:56to make sure they're an owner that has a
28:58business doing at least a million
28:59dollars in revenue. And then we're going
29:02to make sure that we obsess on our
29:03customer acquisition cost. Every
29:06customer response is worth a hundred
29:08bucks to us. Every booked call is worth
29:11500 bucks to us and every ticket sold is
29:14worth the value of the ticket price. Now
29:17we find out if the Grockbot can actually
29:19sell or if it just writes nice emails.
29:22Here's the problem. We need to get
29:24around the fact that Grockbot can't
29:26email anybody if it doesn't know who to
29:28email. So I got to I got to build a
29:30leadfinder bot. It's going to be pulling
29:32business owners off of Google and
29:34LinkedIn. Um what's cool is you can do
29:36it all your from your phone. So spawn
29:38bot. I want you to build a bot that
29:39researches leads based on what I
29:42describe and outputs the leads in a
29:44spreadsheet. This bot should not just
29:46research businesses, but find the owner
29:48and the contact info like their email or
29:51their LinkedIn. Um, think of this bot as
29:54like a lead researcher or a data
29:56engineer. Let's call it Dora, like the
30:00explorer, you know. So, go through
30:01Google, LinkedIn, uh, where else? Uh,
30:04Apollo, Reddit, anywhere you can find.
30:07Okay. The key feature I'm going to add
30:08here is actually this. Before Grockbot
30:11even touches a lead, it's going to check
30:12our HubSpot to see if anyone on our team
30:15is already talking to that person. You
30:17know, if you've ever had three people
30:18ping you from the same company in one
30:19week, like that's going to kill the
30:21deal. Right now, I want Grockbot to
30:23write three different versions of the
30:24cold email so we can see which one
30:26actually gets replies. So, I told it to
30:28create
30:30different versions of our cold outreach
30:32email based on where the business is. So
30:34that if you're in Austin, the email
30:37should say here in Austin. Uh if you're
30:39somewhere else in Texas, here in Texas.
30:41It's a tiny detail, but I think it's
30:42going to actually make a big difference.
30:44Um a plumber reading here in Austin
30:46feels like a neighbor reached out
30:47instead of a call center. Um let's also
30:50slow this
30:53to 50 emails a day out of one inbox.
30:56Then actually, let's add two more
30:59inboxes. Uh I want to be careful here.
31:02This is when bounces can start. Bounces
31:05sound harmless, but every email that
31:07bounces tells Gmail and Outlook your
31:08domain might be spam. And if that
31:10happens, even your good emails stop
31:11landing. So, Grockbot is going to run
31:13every single address through
31:16hunter.io before it sends. Once the list
31:19is clean, it'll bump up to four inboxes
31:21and 200 emails a day. Okay, let's see
31:23how this works. Sick. So, look at this.
31:26We can see all the drafts hitting the
31:28inbox. Right now, I'm going to have
31:30Grockbotbot split the sends across a few
31:32of our sales reps. So, when this an
31:35owner replies like this guy, a real
31:37person on my team picks up the
31:38conversation, starts rocking and rolling
31:40via email, calls the guy, texts the guy,
31:42and this our emailed Rockbot was
31:44actually able to create and it it's like
31:46pretty honestly good. So, not bad. But,
31:49will this work?
31:51I guess we'll see. So, it's day 10. The
31:53bots have been running on their own.
31:55Time to see what they actually did.
31:56We've got two P&Ls left. We've got uh
31:59Wolfie, our trading bot, and the cold
32:01lead bot. Uh we built to fill seats at
32:03our next event. So the trading P&L, I
32:06want to write this out on the
32:07chalkboard. Revenue, oh, we killed it
32:10with this. That would be a 67 over here.
32:14Uh Grockbot usage that is -15 bucks,
32:19which means net we're looking at
32:221567.
32:25Wolfie is what we call
32:28down. To be fair to Wolfie, the bot did
32:31exactly what I asked. It scanned Reddit
32:33X the news and traded on the hype. The
32:35problem is probably the strategy, not
32:37the bot, it could actually execute on
32:38everything. So if you were a killer
32:40trader, the best in the world, probably
32:42not trading with Rockbot, but
32:43theoretically, you could still do this.
32:45The problem is, if you point a bot at a
32:47bad process, you just lose money faster,
32:49which is why you should not use X for
32:51trading advice. Wolfie is thus fired.
32:55We're not keeping that. So, I'm rating
32:56this guy a one or twoish. There's a lot
32:59writing on our last bot. It's been a few
33:01days since launch of the cold lead bot.
33:03I have some updates. Rockbot got us some
33:05hot leads. How many, my friends? 10.
33:09Give it a 10. A 10. So, 10 leads. Here's
33:13what's gone down so far. These 10 said
33:15they were interested. Two booked a call.
33:17So, what does this mean? Well, any
33:19response counts. So those 8 to 10
33:22replies at let's call it a hundred bucks
33:25each at a 100 bucks each. Let's say 10
33:29of them respond. That's a,000 bucks. Two
33:32booked calls which are worth around what
33:34did we say? 500 bucks. That's how we pay
33:36how much we pay for them on Meta. So
33:39hey, we're looking at about 2K right
33:40there in value. And the best part is
33:42it's still working on leads. I'm going
33:43to keep this one in my business. I'll
33:45have Grockbot keep sending and sales
33:47will keep working the replies. We have
33:49to cut the experiment off. But here is
33:50where we stand at at this sort of level.
33:53We're looking at
33:552,000 at cost of goods sold. It's about
33:5837 uh bucks 26
34:02is what that's cost us to run. Rockbot
34:05usage has been 20 bucks thus far, which
34:08is super cheap, which means our total
34:10cost is what? 5726.
34:15And that means net we're sitting kind of
34:17pretty, you guys. 1,942
34:20bucks and like.74
34:23something like that. Nice rockbot. I'd
34:26rate this a five out of five with damn
34:30near $2,000 in revenue with about $187
34:34in costs. That's our net. So to answer,
34:37can Grockbot turn 100 into a,000 bucks?
34:40Yeah, it can with the right strategy.
34:41And I want you to look at which bots
34:43actually won. Wolfie, the flashy one,
34:46lost money. The Facebook reselling bot
34:49made 88 bucks. Super not sexy. And the
34:51book bot made 14. The one that actually
34:54crushed it did the most boring jobs cold
34:56calling, right? That bot alone was worth
34:59almost two two grand. And the Clipping
35:01bot that made $0 is now my editor's
35:03workflow every week. If you're using
35:05Rockbot in your business, let me know in
35:07the comments and tell me which one you
35:08want us to build next. I'll do it. By
35:10the way, if you guys like this and want
35:11to steal all of our bots, you can
35:13comment bot below or click the link and
35:16we will send you exactly the templates
35:18that we used to build these.

About this transcript

This page contains the full transcript of I Tried The “Easiest” Ways To Make Money With AI by Codie Sanchez, generated from the public captions YouTube serves with the video. The transcript has 7,037 words across 974 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.