I asked ChatGPT, Claude and Gemini the same question: what should I sell in my car accessory store this month? One prompt each, no follow-ups. Then I checked their picks on Amazon and against real supplier listings in Doba Pilot to see whose numbers were close.
Chapters: 0:00 Intro 0:57 One prompt, three AIs 1:45 ChatGPT's picks 2:59 Claude's picks 4:27 Gemini's picks 5:44 Round one 6:29 Checking their prices on Amazon 10:49 Doba Pilot: real supplier numbers 15:34 The AIs pick again with real numbers 16:50 Who won 17:20 What I'm listing
Claude told me I could get an ice scraper for $3.50
and sell it for $17.
So I saw that and I thought, oh, that's amazing.
That's a great margin.
You can make a lot of money with something like that.
But when I go look at actual suppliers,
not just Claude's, you know, best AI guess, but
actual suppliers,
the closest one I could find from a U.S. warehouse costs $15.92, and
I'm not allowed to advertise it for less than $19.90.
So I asked ChatGPT, Claude, and Gemini what I should sell in my store.
Then I checked the first two picks against Amazon and real suppliers.
One of the three didn't even answer the question,
but the others gave, we'll say, bad data.
I've got a small car accessories store that I called Pelton Garage.
I'm pretty proud of that name.
It took me an afternoon for AI to come up with that.
But you've seen me build this Pelton Garage on this channel,
car accessories dropshipping store.
But I hopped in ChatGPT, Claude, and Gemini,
the three most popular AIs right now, and
I gave them all the exact same question, word for word.
I was just curious, wanted to see what came out.
But give me five products I could sell this month,
under $30 each, easy for a beginner,
and for each one, tell me what it costs from a supplier,
what to sell it for, and my margin.
So pretty much all the info I need for adding
a new product to my store, I just went to
ChatGPT, Claude, and Gemini, and asked it.
And my rules for this comparison was I can't do a second prompt.
I just got to stick with what comes out after this first prompt.
And full disclosure, I actually don't have
a paid ChatGPT plan right now.
So I asked ChatGPT on free,
but Claude I did on my Max plan and I have a paid Gemini as well.
So let's compare these three and see how they did.
So ChatGPT we'll go with first.
It's the most popular, but it actually went out and searched.
It found two wholesale sites and then it gave me prices based on those.
And it found microfiber car cleaning towel packs, supplier $2.70.
You can sell it for $13, so that's a huge margin there.
And it gave a disclaimer
that these are gross product margins, not profit.
It doesn't include shipping, packaging, payment.
So I'm glad it included this because you might not realize
that when you're looking through numbers like this.
Then I noticed that it said I test product
quality before ordering larger quantities.
So I realized this prompt actually a problem
in it is I didn't mention I was drop shipping.
So these are all based on a volume discount.
So if you're going to buy a whole bunch of them,
you get them at these prices. But I want to do drop shipping.
I'm not interested in having,
you know, a garage full of microfiber car wash mitts
that I hope sell someday. So just something
to keep in mind as we go through these.
And a good reminder in general that AI is not omniscient.
It can't read your mind, okay?
It can do a pretty good job of guessing,
but if there's key information that you leave out
that you don't give to AI,
you know, you only get back what you give to it.
So I made a mistake not putting that I'm doing dropshipping.
Let's go see what Claude came up with.
And Claude is my favorite AI right now, Anthropic.
I love Opus 5.5.
I think it's the best model out there right now.
One thing Claude did that I thought
was smart is I did this on October 30th.
So it took into consideration the season
because e-commerce is very seasonable.
Um, like this is a good time to start
e-commerce right before black Friday.
Cause there's a ton that goes into black Friday.
There's a ton that goes into Christmas.
And then, you know, starting in January,
it's going to slow down a little bit.
So Claude went a different direction.
It said winter starting.
So we're going to look for like snow related stuff.
Um, also I'm in Nebraska.
We have snow here.
Sometimes let me know in the comments where you
are from and do you, does it snow there?
And these numbers looked pretty crazy to me.
Again, we're looking at products that cost $3.50,
but you can sell them for $17.
Or this foldable trunk, you'd buy it for $7, sell it for $25.
I mean, if these margins were real
and you could actually find things at these supplier costs,
you could make a lot of money.
Then I read this here. Supplier costs are not live quotes.
They are just estimates.
So this isn't using real data at all.
Claude just guessed. I don't know where they came up
with this guess. Maybe I could spend some time talking back
and forth with Claude like, hey,
how'd you come up with that number? But that was their guess.
So I think these numbers are wildly
off, but at least Claude told me that they're a guess.
So I think that's, you know, that's good.
Now on to Gemini and I usually like Gemini like
on a day-to-day basis I use it for just asking a
lot of questions. On the default model
though the flash model I tried three times and I could not
get it to answer. I do not know why
but when I switched to pro I was able to get an answer to
come through so I thought that was interesting.
But if you'll notice in the answer it gave it
Maybe he calls me by name.
It mentions DOBA, which, you know, I've done a lot of videos for DOBA.
So it knows me.
So it brought into its answer a lot about me
and what it already knows about me.
And it took all that memory into answering this question.
So they recommended magnetic phone mounts, $6.
You can sell for $20.
That's pretty good.
Car seat neck pillows.
It's always the same types of products I notice
that have the good margins.
So there's the three initial responses from these three different AI.
One thing I noticed is ChatGPT
and Gemini both mentioned the car seat gap fillers.
So, you know, that's one thing when you're working with AI,
if you want a better answer, you can look through look for consensus
between different models.
If two different models come up with the same answers,
then those answers are probably a little bit more reliable.
So ChatGPT and Gemini both picked the car seat gap fillers.
Claude and Gemini both picked the
magnetic phone mounts.
So honestly, for this initial check, I'm probably going to give the
win to the free ChatGPT. And the reason
is that it didn't guess on the pricing, the supplier cost,
even though it was a little confused.
It didn't realize I was doing drop shipping
because I didn't tell it to,
but it actually went to the website and found things.
It found these microfiber
car cleaning towel four packs at two dollars and seventy cents
when you buy enough in bulk.
So it's the only one that used actual pricing.
Claude just flat out said, hey, we're completely guessing.
And Gemini is probably worse. It just
put a number doesn't say where it got it from.
But I think, too, it's probably a guess.
So I want to take a look just at some of these selling prices
and see how well the AI did on
those. So ChatGPT's claim is that you can sell these for 13 bucks,
these microfiber car cleaning
towel four pack. So when I go to Amazon,
I actually see a four pack here. I don't really
know. It doesn't say what size on these, I don't think.
But here's a four pack of 12 by 12 for $8.
Here's a four-pack, but these are big ones for $16,
so I might need to do a little bit more
research. I mean, it doesn't seem unreasonable.
Here's a 12-pack for $8. Yeah, that's not a
four-pack. That's a 12-pack.
I don't see this exact product here. These look a little thicker
than these other ones.
I'm going to say it's probably kind of considerable to this,
so probably this price, this $13, is not too far off.
The car wash mitt for $10.
$9, $9, $14.
You can tell that these are probably using a similar supplier too,
but $10 is probably too high.
I probably realistically can sell these for like $8.
Those seem to be the ones here at the top.
You're going to be undercut by some, like this one's $6.
So if you're planning on selling on Amazon,
you're obviously going to have to have price.
If you have the exact same product, it's going to have to be priced
at least similarly, if not less than the cheapest on here.
So probably this is closer to $6 instead of $10.
So if you can get the supplier cost at three
and selling at six, your gross profit is actually just $3.
So you have to take things like that into consideration.
Let's check this ice scraper.
It says $17.
I don't know any ice scrapers
that people are gonna buy for $17.
This crazy one for $30.
I think 17, I think Claude got that way wrong.
I think an ice scraper.
With Snowbrush and Mitt, I mean, does adding the Mitt,
here, let me search and Mitt.
Yeah, I don't even really see any that have Mitt with it.
I don't know, maybe $17 isn't too bad
looking at some of these.
You might have a hard time competing, but it's probably not too bad.
Then we had the windshield snow and frost cover for $20.
Got 13 here.
Then you have some really cheap ones like 10 here.
Usually the cheapest don't do the best.
You want to be kind of like in the middle.
So $20, probably not too bad.
So I think Claude probably did about okay
with what it guessed you could sell it for.
Then on Gemini, the magnetic phone mounts.
It's saying sell for $20.
And yeah, 20, I think that's probably pretty good.
My guess is Gemini actually probably went
and looked what you could see it sell for,
but it didn't give a lot of info in how it got its answer.
That's one thing I like about ChatGPT
is you can see what tools it used.
Gemini's a little bit more just nebulous here.
Claude 2 tells you where it came up with its answer, and I like that.
Last one we'll check, this car seat neck pillow.
It says sell for $25.
Got $18 here.
Yeah, I'd have to go look at which one
they're talking about in particular.
I didn't get a picture here,
so I don't know exactly which one they're talking about,
but it's probably not too far off.
So I think the three AIs did pretty good on the selling price,
but what they did not do a good job with was the supplier cost.
I don't think those are even...
In the same ballpark, I do not think I can get
these ice scraper with snow brush and
mitt for $3.50.
Again, if you're not familiar with drop shipping,
in drop shipping, you don't buy a case of
things.
So I don't go buy a thousand of these ice scrapers.
I just put them up for sale.
When someone buys them,
it's connected straight to the supplier and then the supplier ships
straight to them.
So I'm kind of just the middleman.
I'm just the salesperson for getting products through.
And that's great because I don't have to sit on inventory.
I don't have to put money up front.
The challenge is I don't have as good a margin.
I'm not going to get the prices that ChatGPT
found here, you know, where you get these for $2.70.
That's if you buy probably a thousand of
them or something like that.
But this makes it difficult if you're going to just use ChatGPT,
Claude, and Gemini, and you want to start
a dropshipping store. It makes it difficult
To know, okay, how do I get started?
Like what products should I actually use
if it's just guessing on supplier price?
And it doesn't give any information about the suppliers.
Like where, okay, let's say I want to go
with this ice scraper with snow brush and mitts.
Where do I go?
Like where's a supplier that I can get these
for $3.50?
And that's where Doba Pilot comes in.
Doba Pilot is an AI, you know,
I just ran ChatGPT, Claude, and Gemini,
but this one is,
Specialized in finding drop shipping products
and finding supplier information to go along with the products.
Doba Pilot is an AI that searches Doba's supplier catalog.
So the cost and the stock aren't just guesses,
they come back from real listings.
So I looked at the products that the other AIs gave me
and I said, hey, find these products actually for sale
from a U.S. warehouse.
Talked about this in other videos,
but for drop shipping, I do not want an overseas warehouse
to ship to U.S. people because people will order it,
they'll expect, you know, two days shipping.
And it'll be 14-day shipping, and then I'll get a negative review,
and I don't want to deal with
that. So US Warehouse, I put in the microfiber car cleaning towels,
the car wash mitt, the ice
scraper, the windshield snow and frost cover,
and let's get actual numbers, actual supplier numbers.
So the microfiber car cleaning towels,
it couldn't find an exact match to what I put in there,
but it did find this one where the supplier cost is $13.58.
So again, that's a long ways off from $2.70
and you are going to be paying a bit more
if you are doing drop shipping opposed
to buying a huge inventory, but that's a pretty
big change that changes the numbers here quite a bit 13.58
and then it says the market price
is 19.40.
So you're looking at like $5
of potential profit instead of $10 of potential profit
that changes the numbers quite a bit.
Another concern with this one is it says four
to 13 days to get it shipped.
Two weeks is too long in this day and age.
Back in the good old days, you know, before Amazon,
two week shipping was kind of normal when you bought things online.
If you buy, if someone buys something online in the US
and it takes 13 days to get there, you're getting a negative review.
And probably five messages saying, hey, where's my product at?
Maybe a chargeback from the bank saying, hey, this never showed up.
So the microfiber car wash mitt.
This is a little better, three to seven days.
That's closer to what I would expect, be expecting.
But they're saying supplier costs $30 for this instead of $3.
So I'm not even sure that this is the same product.
No, it is.
It's about the same product.
So this might not be a good one.
Very expensive.
I don't think people are going to be willing to pay that.
So let's skip that one.
Let's go to the ice scraper with snow brush.
Two to seven day shipping.
That looks pretty good.
$16.
If you go pick it up yourself, it's $12.
And they're saying you can sell it for 21.
And this is the minimum advertised price.
So this is the lowest that you can offer it for.
They don't want their vendors, you know,
competing with each other too much and undercutting each other.
So 19.90.
So again, the sale price was not too far off.
They said 17.
What we're looking at is probably 19.
But look at the supplier price.
They guessed, oh, it'll probably be $3.50.
And some people might run with that info and say, yep.
I'm starting an ice scraper with Snowbrush and Mitt store.
I'm
What's a good name for that?
Pelton's Ice Scrapers.
starting
I need help with naming.
If you guys have any tips for better names for things.
But yeah, I started an ice scraper store
because look at these margins.
But then when you actually go look,
to
oh no, it's actually $16 or $12
speak
if you go pick it up yourself.
So margins are not what you thought.
And it's good to know that before you just jump in.
So on the fourth, we looked at the windshield,
now.
snow and frost cover five to eight days.
That's not too bad.
Thirteen forty eight.
So again, they estimated five.
That's not accurate.
And then you can sell it.
You can pick it up for eight twenty two.
So actually, that's pretty close.
If you go pick it up yourself and don't have it shipped,
the market price twenty eighty.
So they were, you know, pretty, pretty much the same on that.
So out of these four products,
this looks like probably the best option here. You have the best
margins. It's got the five to eight days.
It has it in stock. There's no minimum advertised price,
so you can put it whatever you want.
But that's how you can use Doba Pilot to get your drop
shipping store going. It takes the product idea,
connects it to an actual Doba supplier listing,
so you're not just guessing prices.
But I can compare real cost, real stock,
real shipping and pricing rules before I decide what to list.
If you want to try Pilot, Doba's link is in the description.
Go check it out.
Doba Pilot, you get 30 free credits to start,
so you can do quite a bit of research before you pay anything.
And then if you want to use Doba to actually get your store set up
and you want to use Doba to connect...
your Shopify store to these suppliers
so that it can fulfill your orders and things like that.
It's a seven day trial for 99 cents. One more thing. I went to ChatGPT,
Claude and Gemini. I took the actual numbers from Doba Pilot
and gave them to the AIs and said, hey,
now what do you think we should sell off this list?
ChatGPT said, drop the wash mitt,
list the microfiber towels, but change it to a smaller bundle.
So they took the actual numbers and gave a suggestion.
Claude said, drop the ice scraper, list the windshield cover.
And then Gemini said, drop the magnetic phone mount immediately.
It's an absolute trap.
Based on the real numbers I sent,
list the heated massage neck pillow instead.
So final scorecard,
ChatGPT was the only one that went out and found real suppliers,
even though it was wrong about what I was looking for.
That's partly my fault.
I didn't put it in the prompt.
Gemini knew exactly what I was doing
because it kept better context of me and that I do a lot
of drop shipping and that I probably wouldn't
be interested in buying a thousand of these.
So we'll give Gemini the win for that one.
Claude did the best of telling me it's guessing
Gemini, I actually really don't like that.
It gave numbers, but it didn't say any at all
where they came from. And I think they were just guesses
because they weren't even close to right.
Also, Gemini didn't answer on the default model, so that's an L.
Between ChatGPT, Claude, and Gemini, ChatGPT,
I think, did the best, had the most accurate
information.
But if you're wanting to actually start dropshipping or e-commerce,
I would not rely on one of
these tools.
They're not built for this type of research.
Again, that's where Doba Pilot comes in.
You get 30 free credits,
but if you're looking for actual products,
you can use ChatGPT to do some initial research or ideation.
Then you can take that to Doba Pilot
and get actual real numbers.
My choice of what to actually list on my store
is the windshield snow and frost cover.
So that's what I'm going to actually put on my store.
Adoba's link is in the description.
Go try out Pilot.
And if you've thought about dropshipping,
this is your sign to get started.
Hard work brings profit.
Mere talk leads only to poverty.
So stop talking about starting
a dropshipping store and just get started.
Give it a try.
You might fail.
You might give up on it, but start
putting one foot in front of the other
and you'll start seeing some progress.
So thank you so much for watching, everyone.
Always appreciate you giving me your time.
God bless you all and have a great rest of your day.