September 15, 2025 • 39 mins
AI is revolutionizing e-commerce by enabling brands to scale product catalogs and optimize Amazon listings with unprecedented efficiency. Daniela Bolzmann shares how Mindful Goods uses artificial intelligence to transform data analysis, content creation, and split testing for seven and eight-figure brands.
• AI allows agencies to support enterprise brands with thousands of SKUs by creating template masters deployable across entire catalogs
• Split testing tools like Manager Experiments (free within Amazon), PickFu, IntelliV, and Product Pinion provide critical consumer insights
• Title testing can increase Amazon sales by up to 990% according to Mindful Goods' data
• The three pillars of Amazon success: SEO visibility, main image optimization, and product page conversion
• Data Dive offers powerful keyword analysis capabilities for Amazon listings (use code mindfulVIP for a discount)
• AI-powered synthetic avatar testing allows brands to test messaging against specific personas before spending ad dollars
Guest Contact Information:
More from EWR and Matt:
Leave us a review wherever you listen: Spotify, Apple Podcasts, or Amazon Podcast
Free SEO Consultation: www.ewrdigital.com/discovery-call
With over 5 million downloads, The Best SEO Podcast has been the go-to show for digital marketers, business owners, and entrepreneurs wanting real-world strategies to grow online.
Now, host Matthew Bertram — creator of LLM Visibility™ and the LLM Visibility Stack™, and Lead Strategist at EWR Digital — takes the conversation beyond traditional SEO into the AI era of discoverability.
Each week, Matthew dives into the tactics, frameworks, and insights that matter most in a world where search engines, large language models, and answer engines are reshaping how people find, trust, and choose businesses. From SEO and AI-driven marketing to executive-level growth strategy, you’ll hear expert interviews, deep-dive discussions, and actionable strategies to help you stay ahead of the curve.
Find more episodes here:
Follow us on:
Facebook: @bestseopodcast
Instagram: @thebestseopodcast
Tiktok: @bestseopodcast
LinkedIn: @bestseopodcast
Connect With Matthew Bertram:
Website: www.matthewbertram.com
Instagram: @matt_bertram_live
LinkedIn: @mattbertramlive
Powered by: ewrdigital.com
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:03):
This is the Unknown Secrets of Internet Marketing,
your insider guide to thestrategies top marketers use to
crush the competition.
Ready to unlock your businessfull potential, let's get
started.
Speaker 2 (00:17):
Howdy.
Welcome back to anotherfun-filled episode of the
Unknown Secrets of InternetMarketing.
I am your host, matt Bertram.
We are trying to move intoInternet Marketing marketing
secrets with all our handles onYouTube.
So I'm going to try to startsaying that, but it's going to
be difficult.
But thank you everyone forsticking around.
I want to introduce somebodyvery special that has been a
(00:40):
speaker at Amazon Accelerate.
That's about right around thecorner, that's coming up, and I
know we talk a lot about B2B.
We talk some small.
We've talked a lot about SEOand LLM visibility, but I think
that selling stuff online islike where it's at.
That's where it all started.
That's where Google's goingright.
They're trying to go head tohead with Amazon now and own the
(01:04):
whole funnel.
So I thought it'd be great tobring on Daniela Bozeman with
and I hopefully withmindfulgoodsco.
Yeah, thanks for having me.
Yeah, thanks for being here,daniela.
So we were talking a little bitabout, kind of well, what
everybody wants to know right isAI and how.
(01:26):
AI is transforming our workflowsand how we're doing things, and
you brought up something that Ihadn't heard before but makes
total sense is you're able touse AI to scale catalogs, like
almost indefinitely.
I've also seen, actually, insome of the things with Google
(01:48):
training, where you can withpictures, you can start putting
different backgrounds ondifferent shoes, you can have
the person standing in differentplaces, you can customize
mountains behind it, likedepending on where you're
advertising.
So from an ad creative costlike AI is transforming what's
(02:09):
possible.
Speaker 3 (02:10):
Right.
I mean there's a bunch of toolsthat are coming out on Amazon
side that are focusing on how tomake the ads creative process a
lot easier.
Our focus at Mindful Goods ishelping brands really think
through the conversion piece ontheir product page and on their
storefront.
So in the past we were neverreally able to support those
larger enterprise brands thatwould come to us and they had,
(02:31):
let's say, 250 or 2,500different or 25,000 different
SKUs in their catalog.
For us, that was like we willbe working on that till the end
of time, so it would just be ahard no.
But so many of them approachedus because we, we, we do pull a
ton of data into our process.
We do a lot of split testing inour process.
We create really high quality,almost direct to consumer style
(02:56):
creatives, right, and so everybrand wants that look when
they're, you know, expandingfrom their website to Amazon,
like they want to feel like thetop brand and they want to put
their best foot forward.
We were just never able to doit as an agency, and so that's
one of the most exciting piecesfor us is thinking through okay,
now we can essentially justcreate a set of masters across
(03:18):
the catalog, depending on howmany parent SKUs we're thinking
about, and then we can come upwith a strategy to execute on
that across thousands in hoursor minutes or days, instead of
weeks or months that we weretalking about before.
Speaker 2 (03:34):
Yeah, and you typically work with like seven,
eight figure brands, right, andso now this opens up the top end
, right, because there was justlike the, the amount of like
grunt work that had to get doneto to fill out all those product
pages.
Um, was was a lot of manualeffort.
That that AI and and automationcan help replicate, and now,
with also different tools andimage generation, you can start
(03:58):
to to to scale and personalize,like.
That's what I really like aboutwhat AI is doing is it's
allowing you to personalizestuff.
And if you're thinking about,you know, having a like a photo
shoot used to have to cost a tonof money, right, and and now
with AI, I mean it's almostindistinguishable beyond the
fingers.
I mean they're working on thefingers.
Speaker 3 (04:21):
I have a funny example of that one.
Actually, I recently did awe're working on some video
content with an AI tool and itgenerated a hand holding a
finger, you know, and it wasjust the weirdest scene.
I was like what is happeninghere?
But yeah, that brings up a goodpoint.
We have AI basically in everypart of our process now and I
(04:41):
think in the beginning of 2025,I think a lot of brands were
kind of like whoa, we need totake a beat and just see what
happens with AI, because maybewe don't need to do all of these
things that we had planned for2025.
Maybe we'll save so much moneyif we just don't invest in this
right now.
Let's just wait.
And now we're hitting Q3 andbrands are like okay, wait a
minute, we need to get this done, you know, as we're approaching
(05:02):
Q4.
And you know, if their pagesaren't ready, they aren't ready
for Q4.
So what we're trying to educatebrands on right now is that,
yes, ai is one tool in the toolbelt.
It's allowing us to do all ofthese things better.
It's it's allowing us toanalyze data better, faster and
more precise for your brand.
It's allowing us to, in somecases, replace photo shoots
(05:24):
entirely or just fill in thegaps.
That's fine too, when you thinkabout it from that perspective.
That's something we've alwayshad.
We've had renders, we've hadstock imagery.
It's not like brands alwayscame to us with a full catalog
fresh off a photo shoot.
We had to fill in the gaps andsometimes we had to make
something out of nothing.
So that's still happening.
It's just easier to do with AI.
(05:45):
It's easier to create better,more realistic imagery and more
custom imagery.
So that's definitely one of theways.
And then, like I said, on thebackend, scaling catalogs is
just something that's massivefor us.
Speaker 2 (05:57):
Yeah, I think it's a new layer that everyone's adding
and it's building on what wasdone before.
And you were talking about splittesting before, which you know
early, early, like right when itcame out right and it's like,
oh, this is the smartest thingout there and like whatever it
says is like that's what we'regoing to do and that's even what
(06:17):
I'm hearing on, like thepsychology side of open AI, of
like kids are like putting,asking it questions and doing
whatever it says, like and and,and it's becoming like an issue.
But I can tell you the brainthink of like just giving it
over to a chat, gbt or quad orwhatever you're using Um, it
wasn't always the best.
(06:38):
I thought it was the best.
I like deferred that.
Oh, it is the best.
But you really still need tohave a, a, a, a very senior
operator behind and that knowswhat good looks like, cause I've
tested um meta descriptions, umcalls to action and and uh, it
doesn't always give you the bestthing.
(06:59):
And if you go, take a site andjust click a button, ai, you
know, change all the matterswhich there's tools that you can
do that now and it just autogenerates it.
Like I've seen some bad thingshappen and I've seen some things
that we've had to.
People have come to us thathave AI'd up everything and
we've had to untangle.
Speaker 3 (07:21):
It's basically everybody, especially us.
If we're if we're helpingbusinesses, we need to become AI
engineers, like every businesshas to become an AI engineer if
you're going to help otherbusinesses.
And that's the way I look at it.
It's like it's it's all aboutthe inputs that you're putting
in.
You know, if you're putting ingreat inputs, you're going to
get some great outputs, but ifyou don't even know what those
(07:41):
inputs should be, how are yougoing to come up with a great
output Right?
So, for us, one of our favoritethings of doing right now is
pulling in, and this issomething that was such a manual
thing in the past.
It was like it was like scrapingreviews and then, and then like
basically having our dataanalysts like manually figure
out what, what are the patternsin these reviews.
Now that's done in minutes withAI, right, like we're able to
(08:04):
really feed the AI some reallystrong data across Reddit,
across Amazon reviews, acrossyour competitors reviews, and
then build super strong avatarsand basically have your own
trained AI that can help you,help act as like your own
marketing consultant with youragency to help you cross check
things you know.
(08:25):
So it just makes you smarter.
Speaker 2 (08:27):
I think that that you hit on a key point pulling in
all the data right, because sosiloed.
Point in all the data, gettingthat analysis and then putting
those outputs.
And you know I've hadconversations with with clients,
prospects, whatever about howpowerful this is and you know
what I get a lot.
We know who our clients are.
We know and clients, prospects,whatever about how powerful
this is, and you know what I geta lot.
We know who our clients are.
Speaker 3 (08:47):
We know.
Speaker 2 (08:48):
And.
I'm like and I was like, let's,let's do some analysis.
I bet we can find some thingsand some uh associations that
that, like, no one would everdream of.
Right Like you, you, uh,mckenzie and all the big guys
are are would ever dream of.
Right Like you.
You, uh, mckenzie, and all thethe big guys are are worried.
I think there's an existentialcrisis, because this can now be
(09:09):
done, um, at scale, at afraction of the cost, and you
don't have to pay for all thesedata scientists and all this
labeling and you can just throwall the data in you know, you
can.
You can do prompt engineering.
What you're talking about, theengineering, prompt engineering,
figure out exactly.
You know what are the rightquestions in the workflow to ask
.
You can ideate, you can thenbuild it into workflows and like
(09:31):
what it's spitting out is ispretty incredible and to your
point.
Yeah, the the input's got to beright to get the right outputs
and, and I think early on that'slike kind of where the
trajectory on the curve is.
And that's what that's like kindof where the trajectory on the
curve is, and that's what I did.
I was like, okay, I'm usingthis a lot, like I need to get
serious about it.
I started taking a ton ofdifferent prompt engineering
(09:53):
classes, which gave me a lot ofinsights, and then that becomes
the foundational layer.
And then, okay, like, how do weunify all the data with the MPC
servers?
And so I think there's anevolution here and it's so
powerful.
And I think that the thingthat's lagging the most is
education, and maybe with theclients, and like some clients
(10:17):
that knew everything that therewas to know about digital
marketing, everything shiftedquite a bit and and.
I and I find myself having to um, slow down and kind of like
re-explain stuff and and alsohave to show things and say, hey
, look at this, like look, lookat this output, Cause I I have
(10:37):
one big like publicly tradedclient.
Their stance, kind of goingback to what you were saying,
was no, no AI, and and they wereusing filters on content.
So we were creating content andthey were using filters and and
they were like, oh, this was a90% human written, is 10% AI
pushed it back and I was like,okay, like it was trained on
(11:00):
humans, I know a human wrotethis like like this, this is
slowing down the deliverables,like you were saying Right, and
we're you know?
Speaker 3 (11:09):
and uh, and at the end of the day it's probably
pointier copy that's targetingthe exact avatar, because it was
basically pulled in in a morescientific method than a human
would be doing anyways.
Speaker 2 (11:21):
So those are, these are the kind of debates, right,
that like it's hard to have withclients to say, okay, I, I
understand your viewpoint, butlike, let me show you, like,
give me a little time to showyou a little bit different
perspective, right.
And uh and so so we're in a newworld.
We're definitely in a new world.
Speaker 3 (11:41):
I think, on the Amazon side.
The beautiful thing about theconversion piece, though, is
that, for us, what I always letclients know like my opinion is
my opinion.
You know like I'm just one datapoint.
Let's not lean on my opinion oryour opinion.
Let's actually just let thedata decide.
And now we have managerexperiments, a split testing
tool inside of Amazon.
We have these third party toolslike PickFu, like IntelliV,
(12:03):
like ProductPinion.
They're using real human datato you know, run these split
tests.
Or there's now, like AI-builtavatar split testing, right when
we can pull in completelyAI-generated, you know avatars
to split test in minutes.
Speaker 2 (12:19):
Okay, so, oh, so you're is there.
See, you're so advanced in inin this area.
Like are these avatars?
Can you run fake tests, likewith not real data, and then get
get like an output right?
Speaker 3 (12:32):
Yeah, yeah, yeah, it almost matches our human split
testing.
Speaker 2 (12:36):
What would it, what it would be, what it would be
right.
So go through.
You just listed off a bunch ofdifferent tools.
Could you kind of give us theone-on-one breakdown of some of
these different tools on thesplit testing side of what they
do?
Speaker 3 (12:49):
So manage your experiments is, I would still
say, fairly newer within thelast couple of years, but it's
something that's continuallygetting better and better.
It's a free tool and it'sinside of your seller central
account on Amazon.
Most brands don't use it.
I recommend you should bealways running a test on there,
no matter what.
There's no reason to not be.
It's a 10 minute activity oncea month.
Let it run for the full monthand see if you can get
(13:11):
incremental gains throughout theyear, like you should,
absolutely.
Whoever's managing your Amazonshould be doing that no matter
what, and at the end of everymonth they should be letting you
know what happened with thosetests period.
That's the first thing.
Second thing before we evenupload anything to Amazon, we
use third party tools.
The three top ones out thereright now are PickFu, intelliv
(13:34):
and Product Pinion, and they'reall a little bit different and
they all have their strengths,and so it's really just finding
which one works best for you andwhich one you want to use.
What for?
We use one for actually threedifferent things.
We use each one for a differentthing.
In our business, for instance,we use PicFu for main image
split testing, so our processwith main images is very
(13:56):
specific.
We have another AI tool thatwe've basically trained on all
of our top performing mainimages, like, let's say, 20
different image styles that arelike front loaded with all of
this beautiful aesthetic to andhigh intent keywords to get
people to click on our imagesversus the competitors in the in
the search page results.
And so we, basically we are atthe point where we can drop a
(14:19):
product image into our tool.
It generates 20 images in twominutes and then from there we
have a bunch of differentideation that we can then decide
.
Let's pick three or four andlet's polish these up, let's get
them straight into PickFu,let's test it and then let's
push it straight to manage yourexperiments, to validate on
Amazon, like that's the process,right.
So that's one.
(14:39):
Product opinion is really great.
What I love it for is actuallyvideo polls.
So, again, if we don't want torely on you as a data point or
me as a data point, drop a linkto your listing into product
opinion and ask 50 shoppers orfive to go and look at your, go
and look at your product pageand tell you what could be
(14:59):
better.
And I promise you, even if youjust did it with five.
You would be shocked at theamount of little things that
people say when they areverbally like letting loose on
your, on your.
You know they they are notholding back, you know.
And so that's a really greatway to get out of your own head
and get out of your silo,thinking that your product is
(15:22):
the best, your product pages isworking like.
Go and just let people tell youwhat they think and you'll just
be shocked Like, wow, I did noteven think of that, I did not
even think that people could bethinking about this, about my
packaging.
You know like you'll be shocked.
So highly recommend that as abest practice.
Um, just to kind of takeyourself out of a silo.
And then IntelliV is actuallyreally great at what's called um
(15:42):
product image stack polls andthis is basically the carousel
that lives in the top um of yourproduct page.
When you first go into anamazon listing, you see all of
these images that you can scrollthrough.
Imagine if you could take thoseimages that you have right now
and maybe two more sets that youdesigned, and you could test
them all across each other tosee which ones get up votes,
which ones get down votes, whichorder they should be in, and
(16:05):
basically it'll tell you all ofthat so that you can improve the
images and decide which orderyou're going to upload them in,
so that you can improve yourconversion on Amazon, and so
that's what we recommend usingfor that piece.
Speaker 2 (16:18):
And then, and you said, the avatars for, like,
synthetic data.
Um, it'll, it'll, it'll startto.
I didn't know that that tool isavailable.
Speaker 3 (16:24):
That's cool yeah, so that's a tool, um, it's I'm not
sharing it yet, um, but we'reworking with a team that is
working on that tool and, uh, wewill be sharing it at some
point because it's reallyfascinating.
We're actually pulling in,let's say, reddit reviews,
amazon reviews, competitorsreviews, um, as much data as we
(16:44):
can scrape about our brand as aclient, about their competitors,
and imagine, like, going outand scraping all of this Intel
everywhere, bringing it in andbuilding these really strong
avatars and then building, likeAI, audiences that are based
around those avatars.
So like.
I have an avatar that is justlike a mom who is also a pet
owner, or you know what I mean.
(17:05):
You know you end up knowing whoyour top three avatars are and
you could split test into thosethree as you're developing your
content.
You might have a question asyou're figuring out your
messaging, and you could saywhich one of these messagings is
going to better target my, myICP, you know, or my avatar, or
what else do you recommend toyou know as far as messaging,
(17:25):
and just let it give you ideasback.
So that's going to be a really,really amazing tool, but the
idea there is that it's going todo much more than that.
Speaker 2 (17:35):
Yeah, danielle, I have a question about.
So.
Typically, when I frame it up,like for clients, it's like okay
, what is the ICP Right, andlet's, let's really drill down
on that.
And I think you answered thatreally great.
The next piece that I typicallylayer on is the customer
journey, right, and what like.
There's all kinds of like darksocial and like all right.
(18:00):
There's all these other thingslike what is the funnel look
like or what is the customerjourney look like, cause, like,
I just finished this LinkedInstrategy training and there was
some great data that it sharedwith me in that and I do a lot
of B2B stuff that I didn't knowand 10 pieces of content is, on
average, what someone wants tosee before they convert.
(18:22):
Someone wants to see beforethey convert, and so if that's
what B2B is on, like, let's sayLinkedIn to a website to
account-based selling.
What does that look like on thee-commerce side?
What was the customer journey?
What are the pieces of content?
How do you map that or what areyou looking at?
I'm just curious how thattranslates.
Speaker 3 (18:43):
I think for the brands that we work with,
they're the way that they'refocusing on their traffic
strategies.
Is is kind of all over theplace at the moment, right,
because of AI and because theway that Google search is
changing and Amazon's changingand everything's in flux, right,
and TikTok shop and all of thatlike that's been a huge right,
(19:04):
and so we see brands that aretrying everything and you know
everything at the moment, and wesee brands that are just like
completely doubling down onTikTok, like seeing tremendous
growth.
For me, when I'm trying to workwith brands and trying to
explain to them the funnel onAmazon, I try to keep it just
really simple, because there'salways 150 things that people
(19:25):
are throwing at you that youneed to be doing and really it
just comes down to three thingson Amazon, and even today, this
has not changed Like this hasnot.
I know everybody's talkingabout Rufus, which is Amazon's
AI search that's going to helpyou shop, and it is helping you
shop and that's changing howsearch results are showing up,
which is great.
And there's things that are inmovement and changing and we
(19:47):
need to keep a pulse on, forsure, but the foundation of what
we do today has not changedfrom what we did five years ago
and we're still seeing the sameresults in just staying focused
on the conversion piece, whichis a standard, traditional
funnel, where we're thinkingabout SEO for awareness, like
how do we get found in thesearch results, making sure that
our back end of our listing,the front end of our listing.
(20:09):
And now, most importantly, thebiggest change is our visuals.
Making sure that our SEO is inalignment with those visuals so
that the AI can read it.
That's the biggest changethat's happened there.
The second piece is making surethat your main image is
optimized so that people areclicking on yours instead of
your competitors.
So you're getting found throughthe SEO, you're getting clicks
through the main image and, oncethey're inside your listing,
(20:30):
making sure that every singleimage within there is dialed in
in like a show and tell mannerso that you are pushing for
conversion.
As they're scrolling throughevery single image, right, and
they may not look at all theproduct images, they may not
look at your A plus, but makingsure that every single image is
really thought through and splittested to make sure that you're
converting.
(20:50):
Well, like that's it.
That's it One, two, three.
You know.
Speaker 2 (20:54):
Well, I would love to unpack that a little bit more
Each section, maybe share justsome insights or learnings or
best practices or pitfalls thatwe've seen historically.
If we just think about eachsection.
I would love to have you godeep and just spend a few
moments sharing what you thinkwould be valuable for the
(21:18):
audience in each of thosecategories, because I think that
a lot of people are justoverwhelmed, right, and AI is
now asking so much more, and Ilike how you said look, if
you're trying to build a store,there's three things.
Right, these are the threeareas you need to focus on.
Okay, what are the things ineach area that I need to focus
(21:39):
on?
I'm sure people are gettingtheir pads out and they're
taking notes, you know.
Speaker 3 (21:44):
Yeah.
So for the SEO piece, takingnotes, you know, yeah, so for
the SEO piece, I will tell youand this is backed up by our
data the lowest hanging fruit istitle testing.
So obviously you need to have afull, rounded out SEO copy on
your bullets, on yourdescription.
Your description is not goingto show up on the front end, by
the way, once you upload yourA-plus content, but it still
(22:05):
needs to be there on the backend.
You need to be thinking throughwhat that SEO strategy is.
We use a tool called Data Dive,which I am a huge fan of.
I love it.
If you use any of these tools,by the way, my code if you want
a discount is mindfulVIP, so goahead and give any of these a
shot with that code.
But Data Dive I love because weused to do a lot of things
(22:29):
manually in our agency.
One of them was what we wouldcall like what they call battle
of the titles.
For us it was our titlecalculator.
It was a spreadsheet where wewould go in and do all the
research to pull keywords from,let's say, the search query
reports in your business reportsin Amazon.
We would pull your masterkeyword list.
We'd pull any keywords thatyou're ranking for right now,
that you're indexed for, and anythat your competitors are
(22:49):
indexed for, and then we wouldkind of cross-reference those
against what's converting wellon your advertising PPC.
So we kind of build this masterspreadsheet of keywords, right,
and then we would try tounderstand what is the reach of
all of these keywords and how dowe align, how we're putting our
copy into our listing in such away that it's increasing the
(23:12):
reach, the potential reach forthis customer.
One of the best ways you can dothat is in your title, and so
your title can actually improveyour sales.
I want to say it's like up to550%.
We have a, if you go to mindfulgoodsco, we have a metrics tab
that shows you every quarterwhat we're releasing for our
split tests, and so I will tellyou right now for the SEO copy
(23:36):
oh, I was way off you canincrease your sales, your orders
, by 990% by doing title testing.
So highly recommend doing this.
This is the lowest hanging fruitthat you could possibly do in
terms of activities.
It doesn't require a graphicdesigner.
It requires, yes, you sittingdown and understanding the
(23:59):
keywords that you're using.
But we just worked with aneight figure brand and they have
been selling for more than 10years on Amazon, and they have
been selling for more than 10years on Amazon and they saw
over 600% increase in sales.
And this was one of theactivities that they were
shocked by.
We record videos and we showthem what we did, so they're
along for the ride as we'remaking these adjustments, and
(24:20):
they were just like wow, we hadno idea how far short we were
falling on our SEO, on our SEOprocess, you know.
And so with data dive, itbrings you through a series of
steps where you're reallydialing that in, and one of them
is what they call battle of thetitle.
So lowest hanging fruit.
Highly recommend doing that.
(24:41):
The second piece, if we want todive into that, is the main
images, which I explainedearlier.
Our process has changed quite abit in the last couple months
and now we have an AI toolthat's trained on our best
performing images.
That is really about thinkingthrough the all of the crazy
ideas that you have on how yourproduct could look in that first
image on white and how you cancapture eyeballs quicker, better
(25:05):
, faster, than anybody elsearound you on that page.
So if you go, look for yourproduct and you see 50 products
on the search page, or at leastthe top 20, go and look at what
they're doing.
And then go look in othercategories and see what they're
doing, and this is the point atwhich you want to ideate there
is no bad ideas.
You want a million ideas andthen you're going to kind of
just create them all with AI andjust kind of like look at them
(25:26):
and be like, okay, which ones ofthese do we realistically think
are good options for our mainimage?
Grab a few of the top ones,polish them up, throw them into
a split test and with that youcan improve your sales by 550%
with just that one piece right.
So again, low hanging fruit onthese very simple tests.
And that final piece, the Aplus content.
(25:48):
Once you're inside the listing,what we recommend, there is a
landing page style format.
So now Amazon has what's calledpremium A plus content.
Any brand who's brand registeredselling on seller central can
get this unlocked on theiraccount.
It's something that vendorcentral brands are paying.
We have clients that are payingalmost a million dollars for
(26:09):
access to this and it'scompletely free for brands that
are on Seller Central and thatare brand registered and anyone
can get access to this.
Okay, so the only requirements,just so that everybody knows,
because I feel like I'mscreaming from the rooftops
about this and like 50% ofbrands that come to us still
have not unlocked this.
But it is free to you and youshould unlock it and get it
(26:31):
while you can.
It is the only thing you haveto do to qualify to get it is a
brand story.
You have to upload a brandstory, which, by the way, you
don't have to have anythingfancy.
Put a placeholder image, put alifestyle image and submit it.
So the Amazon approves it, andthen it's going to ask you which
ASINs you want to put this onor which SKUs you want to apply
this.
To Apply it to every single SKUin your catalog.
(26:52):
Submit it, leave it there.
The only other thing you haveto do is you have to have
submitted I think it's 10submissions of A plus content.
This part's changing.
It might be five now.
It used to be like 20.
And so, really, what you wantto do here, it doesn't mean you
need to have 20 SKUs.
It doesn't mean you need tohave 10 SKUs.
It doesn't even mean you needto have five SKUs.
You could have one product andall you have to do is submit A
(27:16):
plus content, which again,doesn't need to be what we do.
It doesn't need to be a landingpage style A plus content.
Just put up a banner, alifestyle image and submit it on
one of your ASINs and then letAmazon approve it.
Once they approve it, you'regoing to duplicate and submit.
Do that 10 times.
And then within a week, this newfeature in your account is
(27:37):
going to be unlocked for you,called premium A plus content.
So that lives below the foldwhen scrollers are shopping.
It's the thumb stopper rightabove the review.
So it's all of those reallypretty big banners that act like
a landing page.
That can be a huge conversionpiece, and so for us, we think
through that exactly as youwould think through your landing
page on your website, andthat's the piece that can get
(28:00):
you up to 1,008% increase insales.
Wow.
Speaker 2 (28:03):
That's a huge upgrade for everyone that's listening.
Could you, Daniela, describe ordefine what is A-plus content
from Amazon's perspective andhow would you look at that?
Let's dig into that a littlebit more.
Speaker 3 (28:18):
Yeah, so A-plus content.
So there's two pieces insideyour product page.
Once they click to go insideyour listing, the product image
carousel is what you see at thetop.
So you have your hero image andyour other product images and a
video.
Then you scroll down becauseyou want to go see the reviews
and there's these other bannersthat live right below the fold
(28:39):
and sometimes and this is whatbrands used to do in the past
brands would just upload justrandom stock images there and a
bunch of text because you couldindex really well in the past
with that section.
Amazon has come out in in in thelast few years saying like
that's not the intent.
The intent of that area is toreally showcase your brand.
(29:02):
So it could be cross-selling,could be educating them on how
you've been you know a legacybusiness around for 45 years.
It could.
There's so many positioningthings you can do there.
But the main point is thatthey're giving you real estate
on your product page to reallyconvert and to to bring the
shopper through something,because there's so many
(29:23):
distractions on that productpage.
Like, if you go look at aproduct page today, there's
advertisements all over the page, so you have to treat it
differently than you would treatyour Shopify site.
You know like you really haveto get the attention, and that
is the spot to do it.
It's the one page, it's the onesection that Amazon is giving
you to put full scale video.
You can have carousels, you canhave hotspots, and so it really
(29:45):
does function very similar to alanding page that sits within
your product page in the middleof all that distracting
advertising.
Speaker 2 (29:52):
So you want to make the most of it that we haven't
really covered, that you wantedto make sure to share about
Amazon specifically.
(30:14):
I think that you've given peoplea lot to think about a lot of
different tools.
There's really no reason whyyou shouldn't be doing split
testing now with large languagemodels.
It's so easy to do.
There's tools coming out allthe time.
This is the magic sauce thatyou've wanted for a long time
and you should start using it.
And when now you're talkingabout the synthetic data on the,
the, the target personas,that's going to be incredible,
(30:36):
because before you even spend adollar, it was kind of that's
kind of like trading stocks with, with the play accounts, right,
like it's like this is so sogood.
You don't have to spend themoney to get the data.
Speaker 3 (30:49):
And there's a tool out there right now that had
that about a year ago and theywere paying.
They were there, theypositioned themselves as an
enterprise tool and I knowbrands right now they're still
using it and that tool.
I mean we're talking about likehundreds of thousands of
dollars and you're going to beable to get that same level soon
.
Like I mean, you can build iton your own right now as your
(31:10):
own custom GPT.
You could do something reallysimilar, but there's going to be
softwares that are going tooffer this that are just going
to make it so simple for you.
Speaker 2 (31:17):
Yeah, they're going to.
They're putting wrappers aroundeverything that.
That's actually what a lot ofthese tools were like um uh was,
was uh just a low grade AI orLLM wrapped around, and that was
the thing to do.
I think that everything's kindof out of the box now, Even
traffic's like left the website.
You were talking about Shopifyand I just wanted to kind of
(31:37):
hear your perspective, because Italked to the brands now and
then that have an e-commercecomponent and like I actually
have some equity in a few, and Iwas like we need to be on
Amazon.
No, they take too big a cut,Like.
I've heard different storieslike and I'm like, okay, you
could mark it up and thenthey'll knock it back down, but
(31:59):
Amazon's like the biggest searchengine for this.
So I've had like a lot of thesedebates so with other
properties that are out there,how do you view them?
Speaker 3 (32:09):
How do you view the ecosystem?
That's such a good question.
Okay, let me give you my takeon it.
So I think you have to do both,and here's why.
So my personal use case I livein Peru.
I'm in Peru right now, but Iwas in New York a few days ago
with the team at Amazon.
And when I go to United Statesat Amazon and when I when I go
(32:32):
to United States, I do my haulright.
So my behavior is my behavioris I already?
had my shopping cart, yeah, preloaded, right.
So I'm about to board my flightand I'm like everything that's
prime, boom, get to the hotel.
So it's waiting for me, right.
When I get there.
If you are not prime and you'reon Amazon, you're not making
the cut.
For instance, I needed shoesbecause I'm going to a wedding
in Italy and I was like, okay, Ineed these specific shoes, but
(32:57):
I need them to either be at myhotel or at my mom's house in
California so that they make itto Italy.
What am I going to do, right?
So I'm trying to think through.
So I go to the Sam Edelmanwebsite.
Am I going to do Right?
So I'm trying to think through.
So I go to the Sam Edelmanwebsite.
They don't have any sort ofguarantee that they're going to
be at my hotel by a certain date, right?
(33:19):
So that's a lost sale.
I wanted a specific toothpaste.
I gave this example the otherday on another podcast.
I wanted a specific toothpasteand this might be TMI, but I
really wanted a.
I use a tea tree oil toothpaste.
It's by very specific brand.
I know they sell on Amazon.
They didn't have prime to set,set up to deliver to New York in
time before I left and I wasonly there for two days.
So they lost the sale rightJust by not having their product
(33:41):
in stock in Amazon's FBAwarehouse.
Am I going to go to theirwebsite to buy my toothpaste?
No, I'm not, because I don'tknow for sure that their website
will get it to me on time,right, and so it's like that is
the Amazon shopper.
I am the epitome of the Amazon.
This all about convenience,right, like I absolutely wait
for the last minute.
(34:01):
I put all my orders in.
They get to the hotel and ifyou don't have prime and you're
not on Amazon and you're notmaking the cut, you lost the
sale Right.
So that is a very specific buyerand there's a lot of us right
Now.
If I lived in California, stillI may have a little bit more
time and I might actually takemore time to go to people's
website.
But if I'm a busy mom and I gotto, I got to take my kids to
(34:23):
school and I got to walk the dogand I got to do all this other
stuff at home I don't have timeto be sitting around on websites
and figuring out like when I'mgoing to get the thing that I
want.
Like I just need to know, likethat's what I need.
It answers all my questionsreally quickly.
I know I'm going to get itcause it's prime and done, done,
just like clear that mentalload, you know.
Speaker 2 (34:41):
Yeah, well, I think that the ecosystem has has
changed the game, because youdon't have to set up multiple
passwords, you don't have to login, you're not putting your
information all in one place,it's all in the ecosystem.
I mean Amazon.
It's been a process to move tolike an AI first company, but I
mean it's changed the game andit's the power of the network
(35:03):
and as it grows, and to yourpoint, I mean I sometimes of the
network and as it grows, and toto your point, I mean I
sometimes I live in Houston andso, um, you know, if I go
somewhere else, um, it's notlike it's delivered within a
couple hours, but like I orderstuff and it's here.
It's scary quick, right, it'sscary quick Cause the, the
warehouse around the corner orwhatever, and I mean there's
(35:25):
Amazon deliveries, you know, onour street every, every day, you
know, maybe multiple times aday, and, um it, it has changed
what our expectation is.
A lot of times, if I ordersomething on a different website
, man, I'll forget that I evenordered it and it comes in the
mail.
I'm like what's this?
You know?
Speaker 3 (35:41):
Um, and now the other thing, though, to consider is,
like, from the brand'sperspective, I get it.
Why you, your margins arebetter on your own website,
right Like, I get that.
So there is a way tomerchandise things differently
on different platforms, rightLike?
Maybe you wouldn't sell theexact same bundle or size that
you would on Amazon, and youdon't want Amazon doing price
checks against any of your otherretailers or your own website
(36:03):
because they want to have thebest price too, right?
So there is a way to kind ofpackage things up and
merchandise in such a way thatthe Amazon experience is its own
thing, separate from yourwebsite, and that's something to
be considered as well.
Speaker 2 (36:16):
Like what I found brands doing is they offer the
same price as Amazon and thenthey offer coupons and stuff
like that on their own websiteto bring down the price.
There's a lot of that that haveto, I guess, be played and from
.
Speaker 3 (36:34):
Shopify, you need, like you need to push up that
average car value too, you know,whereas, like on Amazon, if
you're, let's say, like avelocity product, like a CPG
product, you could focus more onlike subscriptions and getting
subscribed and save buyers, youknow, and pushing that up.
So it's like probably differentKPIs depending on the channel.
Speaker 2 (36:52):
Yeah, you just got to model it and figure out what.
What makes the most sense, butto, not to, not to, to not open
up Amazon is.
I mean, how much Danielle isAmazon of the like overall
market, Like how, how big of apercentage is it?
I don't know Is.
Speaker 3 (37:08):
Amazon of the overall market.
How big of a percentage?
Speaker 2 (37:09):
is it I don't know, of online sales, yeah, or is
there any metrics you have ofhow big of a piece of pie that
is that people want to walk awayfrom it?
Speaker 3 (37:16):
Yeah, I don't want to quote it, but I think it's over
50% of online sales, that'sincredible.
I don't want to misquote that.
But there is really strongfigures out there, that I mean
Amazon is the biggest of onlinesales.
Speaker 2 (37:32):
Yeah, well, awesome, well, this has been great.
So if we had to synthesizeeverything that we were talking
about and maybe surface onething of like what is the one
unknown secret of internetmarketing that you want people
to walk away with, what wouldthat be?
Speaker 3 (37:50):
Oh, don't put all your eggs in the advertising
basket.
Really focus on conversion soyou can get more out of each
dollar.
Speaker 2 (37:58):
I love that.
I love that.
All right, danielle, how dopeople get in contact with you?
Follow you?
I know you speak.
Where are you active on socialplatforms so people can find out
more about your stuff?
Speaker 3 (38:08):
I'm a little bit of a big deal on LinkedIn, so you
can find me there, but also atmindfulgoodsco if you want to
chat with us.
We have a fun thing called amini audit, so if you want me or
one of my creative directors tolook at your listing, we record
a video and give you all of ourtips.
You can either hand it off toyour team or hire us to do the
work, but I recommend checkingthat out.
Speaker 2 (38:27):
All right, Well, awesome.
Well, guys, if you thoughtDanielle was a good guest and
you would like more guests likeDanielle, please hit the
subscribe button.
Please follow us, share like us.
We are Internet MarketingSecrets now on YouTube, so we're
going to get that up and goingthis quarter and just appreciate
all your support.
My name is Matt Bertram.
This is the Unknown Secrets ofInternet Marketing Until the
(38:57):
next time.
Bye-bye for now.
This is the Unknown Secrets of Internet Marketing,
your insider guide to thestrategies top marketers use to
crush the competition.
Ready to unlock your businessfull potential, let's get
started.
Speaker 2 (00:17):
Howdy.
Welcome back to anotherfun-filled episode of the
Unknown Secrets of InternetMarketing.
I am your host, matt Bertram.
We are trying to move intoInternet Marketing marketing
secrets with all our handles onYouTube.
So I'm going to try to startsaying that, but it's going to
be difficult.
But thank you everyone forsticking around.
I want to introduce somebodyvery special that has been a
(00:40):
speaker at Amazon Accelerate.
That's about right around thecorner, that's coming up, and I
know we talk a lot about B2B.
We talk some small.
We've talked a lot about SEOand LLM visibility, but I think
that selling stuff online islike where it's at.
That's where it all started.
That's where Google's goingright.
They're trying to go head tohead with Amazon now and own the
(01:04):
whole funnel.
So I thought it'd be great tobring on Daniela Bozeman with
and I hopefully withmindfulgoodsco.
Yeah, thanks for having me.
Yeah, thanks for being here,daniela.
So we were talking a little bitabout, kind of well, what
everybody wants to know right isAI and how.
(01:26):
AI is transforming our workflowsand how we're doing things, and
you brought up something that Ihadn't heard before but makes
total sense is you're able touse AI to scale catalogs, like
almost indefinitely.
I've also seen, actually, insome of the things with Google
(01:48):
training, where you can withpictures, you can start putting
different backgrounds ondifferent shoes, you can have
the person standing in differentplaces, you can customize
mountains behind it, likedepending on where you're
advertising.
So from an ad creative costlike AI is transforming what's
(02:09):
possible.
Speaker 3 (02:10):
Right.
I mean there's a bunch of toolsthat are coming out on Amazon
side that are focusing on how tomake the ads creative process a
lot easier.
Our focus at Mindful Goods ishelping brands really think
through the conversion piece ontheir product page and on their
storefront.
So in the past we were neverreally able to support those
larger enterprise brands thatwould come to us and they had,
(02:31):
let's say, 250 or 2,500different or 25,000 different
SKUs in their catalog.
For us, that was like we willbe working on that till the end
of time, so it would just be ahard no.
But so many of them approachedus because we, we, we do pull a
ton of data into our process.
We do a lot of split testing inour process.
We create really high quality,almost direct to consumer style
(02:56):
creatives, right, and so everybrand wants that look when
they're, you know, expandingfrom their website to Amazon,
like they want to feel like thetop brand and they want to put
their best foot forward.
We were just never able to doit as an agency, and so that's
one of the most exciting piecesfor us is thinking through okay,
now we can essentially justcreate a set of masters across
(03:18):
the catalog, depending on howmany parent SKUs we're thinking
about, and then we can come upwith a strategy to execute on
that across thousands in hoursor minutes or days, instead of
weeks or months that we weretalking about before.
Speaker 2 (03:34):
Yeah, and you typically work with like seven,
eight figure brands, right, andso now this opens up the top end
, right, because there was justlike the, the amount of like
grunt work that had to get doneto to fill out all those product
pages.
Um, was was a lot of manualeffort.
That that AI and and automationcan help replicate, and now,
with also different tools andimage generation, you can start
(03:58):
to to to scale and personalize,like.
That's what I really like aboutwhat AI is doing is it's
allowing you to personalizestuff.
And if you're thinking about,you know, having a like a photo
shoot used to have to cost a tonof money, right, and and now
with AI, I mean it's almostindistinguishable beyond the
fingers.
I mean they're working on thefingers.
Speaker 3 (04:21):
I have a funny example of that one.
Actually, I recently did awe're working on some video
content with an AI tool and itgenerated a hand holding a
finger, you know, and it wasjust the weirdest scene.
I was like what is happeninghere?
But yeah, that brings up a goodpoint.
We have AI basically in everypart of our process now and I
(04:41):
think in the beginning of 2025,I think a lot of brands were
kind of like whoa, we need totake a beat and just see what
happens with AI, because maybewe don't need to do all of these
things that we had planned for2025.
Maybe we'll save so much moneyif we just don't invest in this
right now.
Let's just wait.
And now we're hitting Q3 andbrands are like okay, wait a
minute, we need to get this done, you know, as we're approaching
(05:02):
Q4.
And you know, if their pagesaren't ready, they aren't ready
for Q4.
So what we're trying to educatebrands on right now is that,
yes, ai is one tool in the toolbelt.
It's allowing us to do all ofthese things better.
It's it's allowing us toanalyze data better, faster and
more precise for your brand.
It's allowing us to, in somecases, replace photo shoots
(05:24):
entirely or just fill in thegaps.
That's fine too, when you thinkabout it from that perspective.
That's something we've alwayshad.
We've had renders, we've hadstock imagery.
It's not like brands alwayscame to us with a full catalog
fresh off a photo shoot.
We had to fill in the gaps andsometimes we had to make
something out of nothing.
So that's still happening.
It's just easier to do with AI.
(05:45):
It's easier to create better,more realistic imagery and more
custom imagery.
So that's definitely one of theways.
And then, like I said, on thebackend, scaling catalogs is
just something that's massivefor us.
Speaker 2 (05:57):
Yeah, I think it's a new layer that everyone's adding
and it's building on what wasdone before.
And you were talking about splittesting before, which you know
early, early, like right when itcame out right and it's like,
oh, this is the smartest thingout there and like whatever it
says is like that's what we'regoing to do and that's even what
(06:17):
I'm hearing on, like thepsychology side of open AI, of
like kids are like putting,asking it questions and doing
whatever it says, like and and,and it's becoming like an issue.
But I can tell you the brainthink of like just giving it
over to a chat, gbt or quad orwhatever you're using Um, it
wasn't always the best.
(06:38):
I thought it was the best.
I like deferred that.
Oh, it is the best.
But you really still need tohave a, a, a, a very senior
operator behind and that knowswhat good looks like, cause I've
tested um meta descriptions, umcalls to action and and uh, it
doesn't always give you the bestthing.
(06:59):
And if you go, take a site andjust click a button, ai, you
know, change all the matterswhich there's tools that you can
do that now and it just autogenerates it.
Like I've seen some bad thingshappen and I've seen some things
that we've had to.
People have come to us thathave AI'd up everything and
we've had to untangle.
Speaker 3 (07:21):
It's basically everybody, especially us.
If we're if we're helpingbusinesses, we need to become AI
engineers, like every businesshas to become an AI engineer if
you're going to help otherbusinesses.
And that's the way I look at it.
It's like it's it's all aboutthe inputs that you're putting
in.
You know, if you're putting ingreat inputs, you're going to
get some great outputs, but ifyou don't even know what those
(07:41):
inputs should be, how are yougoing to come up with a great
output Right?
So, for us, one of our favoritethings of doing right now is
pulling in, and this issomething that was such a manual
thing in the past.
It was like it was like scrapingreviews and then, and then like
basically having our dataanalysts like manually figure
out what, what are the patternsin these reviews.
Now that's done in minutes withAI, right, like we're able to
(08:04):
really feed the AI some reallystrong data across Reddit,
across Amazon reviews, acrossyour competitors reviews, and
then build super strong avatarsand basically have your own
trained AI that can help you,help act as like your own
marketing consultant with youragency to help you cross check
things you know.
(08:25):
So it just makes you smarter.
Speaker 2 (08:27):
I think that that you hit on a key point pulling in
all the data right, because sosiloed.
Point in all the data, gettingthat analysis and then putting
those outputs.
And you know I've hadconversations with with clients,
prospects, whatever about howpowerful this is and you know
what I get a lot.
We know who our clients are.
We know and clients, prospects,whatever about how powerful
this is, and you know what I geta lot.
We know who our clients are.
Speaker 3 (08:47):
We know.
Speaker 2 (08:48):
And.
I'm like and I was like, let's,let's do some analysis.
I bet we can find some thingsand some uh associations that
that, like, no one would everdream of.
Right Like you, you, uh,mckenzie and all the big guys
are are would ever dream of.
Right Like you.
You, uh, mckenzie, and all thethe big guys are are worried.
I think there's an existentialcrisis, because this can now be
(09:09):
done, um, at scale, at afraction of the cost, and you
don't have to pay for all thesedata scientists and all this
labeling and you can just throwall the data in you know, you
can.
You can do prompt engineering.
What you're talking about, theengineering, prompt engineering,
figure out exactly.
You know what are the rightquestions in the workflow to ask
.
You can ideate, you can thenbuild it into workflows and like
(09:31):
what it's spitting out is ispretty incredible and to your
point.
Yeah, the the input's got to beright to get the right outputs
and, and I think early on that'slike kind of where the
trajectory on the curve is.
And that's what that's like kindof where the trajectory on the
curve is, and that's what I did.
I was like, okay, I'm usingthis a lot, like I need to get
serious about it.
I started taking a ton ofdifferent prompt engineering
(09:53):
classes, which gave me a lot ofinsights, and then that becomes
the foundational layer.
And then, okay, like, how do weunify all the data with the MPC
servers?
And so I think there's anevolution here and it's so
powerful.
And I think that the thingthat's lagging the most is
education, and maybe with theclients, and like some clients
(10:17):
that knew everything that therewas to know about digital
marketing, everything shiftedquite a bit and and.
I and I find myself having to um, slow down and kind of like
re-explain stuff and and alsohave to show things and say, hey
, look at this, like look, lookat this output, Cause I I have
(10:37):
one big like publicly tradedclient.
Their stance, kind of goingback to what you were saying,
was no, no AI, and and they wereusing filters on content.
So we were creating content andthey were using filters and and
they were like, oh, this was a90% human written, is 10% AI
pushed it back and I was like,okay, like it was trained on
(11:00):
humans, I know a human wrotethis like like this, this is
slowing down the deliverables,like you were saying Right, and
we're you know?
Speaker 3 (11:09):
and uh, and at the end of the day it's probably
pointier copy that's targetingthe exact avatar, because it was
basically pulled in in a morescientific method than a human
would be doing anyways.
Speaker 2 (11:21):
So those are, these are the kind of debates, right,
that like it's hard to have withclients to say, okay, I, I
understand your viewpoint, butlike, let me show you, like,
give me a little time to showyou a little bit different
perspective, right.
And uh and so so we're in a newworld.
We're definitely in a new world.
Speaker 3 (11:41):
I think, on the Amazon side.
The beautiful thing about theconversion piece, though, is
that, for us, what I always letclients know like my opinion is
my opinion.
You know like I'm just one datapoint.
Let's not lean on my opinion oryour opinion.
Let's actually just let thedata decide.
And now we have managerexperiments, a split testing
tool inside of Amazon.
We have these third party toolslike PickFu, like IntelliV,
(12:03):
like ProductPinion.
They're using real human datato you know, run these split
tests.
Or there's now, like AI-builtavatar split testing, right when
we can pull in completelyAI-generated, you know avatars
to split test in minutes.
Speaker 2 (12:19):
Okay, so, oh, so you're is there.
See, you're so advanced in inin this area.
Like are these avatars?
Can you run fake tests, likewith not real data, and then get
get like an output right?
Speaker 3 (12:32):
Yeah, yeah, yeah, it almost matches our human split
testing.
Speaker 2 (12:36):
What would it, what it would be, what it would be
right.
So go through.
You just listed off a bunch ofdifferent tools.
Could you kind of give us theone-on-one breakdown of some of
these different tools on thesplit testing side of what they
do?
Speaker 3 (12:49):
So manage your experiments is, I would still
say, fairly newer within thelast couple of years, but it's
something that's continuallygetting better and better.
It's a free tool and it'sinside of your seller central
account on Amazon.
Most brands don't use it.
I recommend you should bealways running a test on there,
no matter what.
There's no reason to not be.
It's a 10 minute activity oncea month.
Let it run for the full monthand see if you can get
(13:11):
incremental gains throughout theyear, like you should,
absolutely.
Whoever's managing your Amazonshould be doing that no matter
what, and at the end of everymonth they should be letting you
know what happened with thosetests period.
That's the first thing.
Second thing before we evenupload anything to Amazon, we
use third party tools.
The three top ones out thereright now are PickFu, intelliv
(13:34):
and Product Pinion, and they'reall a little bit different and
they all have their strengths,and so it's really just finding
which one works best for you andwhich one you want to use.
What for?
We use one for actually threedifferent things.
We use each one for a differentthing.
In our business, for instance,we use PicFu for main image
split testing, so our processwith main images is very
(13:56):
specific.
We have another AI tool thatwe've basically trained on all
of our top performing mainimages, like, let's say, 20
different image styles that arelike front loaded with all of
this beautiful aesthetic to andhigh intent keywords to get
people to click on our imagesversus the competitors in the in
the search page results.
And so we, basically we are atthe point where we can drop a
(14:19):
product image into our tool.
It generates 20 images in twominutes and then from there we
have a bunch of differentideation that we can then decide
.
Let's pick three or four andlet's polish these up, let's get
them straight into PickFu,let's test it and then let's
push it straight to manage yourexperiments, to validate on
Amazon, like that's the process,right.
So that's one.
(14:39):
Product opinion is really great.
What I love it for is actuallyvideo polls.
So, again, if we don't want torely on you as a data point or
me as a data point, drop a linkto your listing into product
opinion and ask 50 shoppers orfive to go and look at your, go
and look at your product pageand tell you what could be
(14:59):
better.
And I promise you, even if youjust did it with five.
You would be shocked at theamount of little things that
people say when they areverbally like letting loose on
your, on your.
You know they they are notholding back, you know.
And so that's a really greatway to get out of your own head
and get out of your silo,thinking that your product is
(15:22):
the best, your product pages isworking like.
Go and just let people tell youwhat they think and you'll just
be shocked Like, wow, I did noteven think of that, I did not
even think that people could bethinking about this, about my
packaging.
You know like you'll be shocked.
So highly recommend that as abest practice.
Um, just to kind of takeyourself out of a silo.
And then IntelliV is actuallyreally great at what's called um
(15:42):
product image stack polls andthis is basically the carousel
that lives in the top um of yourproduct page.
When you first go into anamazon listing, you see all of
these images that you can scrollthrough.
Imagine if you could take thoseimages that you have right now
and maybe two more sets that youdesigned, and you could test
them all across each other tosee which ones get up votes,
which ones get down votes, whichorder they should be in, and
(16:05):
basically it'll tell you all ofthat so that you can improve the
images and decide which orderyou're going to upload them in,
so that you can improve yourconversion on Amazon, and so
that's what we recommend usingfor that piece.
Speaker 2 (16:18):
And then, and you said, the avatars for, like,
synthetic data.
Um, it'll, it'll, it'll startto.
I didn't know that that tool isavailable.
Speaker 3 (16:24):
That's cool yeah, so that's a tool, um, it's I'm not
sharing it yet, um, but we'reworking with a team that is
working on that tool and, uh, wewill be sharing it at some
point because it's reallyfascinating.
We're actually pulling in,let's say, reddit reviews,
amazon reviews, competitorsreviews, um, as much data as we
(16:44):
can scrape about our brand as aclient, about their competitors,
and imagine, like, going outand scraping all of this Intel
everywhere, bringing it in andbuilding these really strong
avatars and then building, likeAI, audiences that are based
around those avatars.
So like.
I have an avatar that is justlike a mom who is also a pet
owner, or you know what I mean.
(17:05):
You know you end up knowing whoyour top three avatars are and
you could split test into thosethree as you're developing your
content.
You might have a question asyou're figuring out your
messaging, and you could saywhich one of these messagings is
going to better target my, myICP, you know, or my avatar, or
what else do you recommend toyou know as far as messaging,
(17:25):
and just let it give you ideasback.
So that's going to be a really,really amazing tool, but the
idea there is that it's going todo much more than that.
Speaker 2 (17:35):
Yeah, danielle, I have a question about.
So.
Typically, when I frame it up,like for clients, it's like okay
, what is the ICP Right, andlet's, let's really drill down
on that.
And I think you answered thatreally great.
The next piece that I typicallylayer on is the customer
journey, right, and what like.
There's all kinds of like darksocial and like all right.
(18:00):
There's all these other thingslike what is the funnel look
like or what is the customerjourney look like, cause, like,
I just finished this LinkedInstrategy training and there was
some great data that it sharedwith me in that and I do a lot
of B2B stuff that I didn't knowand 10 pieces of content is, on
average, what someone wants tosee before they convert.
(18:22):
Someone wants to see beforethey convert, and so if that's
what B2B is on, like, let's sayLinkedIn to a website to
account-based selling.
What does that look like on thee-commerce side?
What was the customer journey?
What are the pieces of content?
How do you map that or what areyou looking at?
I'm just curious how thattranslates.
Speaker 3 (18:43):
I think for the brands that we work with,
they're the way that they'refocusing on their traffic
strategies.
Is is kind of all over theplace at the moment, right,
because of AI and because theway that Google search is
changing and Amazon's changingand everything's in flux, right,
and TikTok shop and all of thatlike that's been a huge right,
(19:04):
and so we see brands that aretrying everything and you know
everything at the moment, and wesee brands that are just like
completely doubling down onTikTok, like seeing tremendous
growth.
For me, when I'm trying to workwith brands and trying to
explain to them the funnel onAmazon, I try to keep it just
really simple, because there'salways 150 things that people
(19:25):
are throwing at you that youneed to be doing and really it
just comes down to three thingson Amazon, and even today, this
has not changed Like this hasnot.
I know everybody's talkingabout Rufus, which is Amazon's
AI search that's going to helpyou shop, and it is helping you
shop and that's changing howsearch results are showing up,
which is great.
And there's things that are inmovement and changing and we
(19:47):
need to keep a pulse on, forsure, but the foundation of what
we do today has not changedfrom what we did five years ago
and we're still seeing the sameresults in just staying focused
on the conversion piece, whichis a standard, traditional
funnel, where we're thinkingabout SEO for awareness, like
how do we get found in thesearch results, making sure that
our back end of our listing,the front end of our listing.
(20:09):
And now, most importantly, thebiggest change is our visuals.
Making sure that our SEO is inalignment with those visuals so
that the AI can read it.
That's the biggest changethat's happened there.
The second piece is making surethat your main image is
optimized so that people areclicking on yours instead of
your competitors.
So you're getting found throughthe SEO, you're getting clicks
through the main image and, oncethey're inside your listing,
(20:30):
making sure that every singleimage within there is dialed in
in like a show and tell mannerso that you are pushing for
conversion.
As they're scrolling throughevery single image, right, and
they may not look at all theproduct images, they may not
look at your A plus, but makingsure that every single image is
really thought through and splittested to make sure that you're
converting.
(20:50):
Well, like that's it.
That's it One, two, three.
You know.
Speaker 2 (20:54):
Well, I would love to unpack that a little bit more
Each section, maybe share justsome insights or learnings or
best practices or pitfalls thatwe've seen historically.
If we just think about eachsection.
I would love to have you godeep and just spend a few
moments sharing what you thinkwould be valuable for the
(21:18):
audience in each of thosecategories, because I think that
a lot of people are justoverwhelmed, right, and AI is
now asking so much more, and Ilike how you said look, if
you're trying to build a store,there's three things.
Right, these are the threeareas you need to focus on.
Okay, what are the things ineach area that I need to focus
(21:39):
on?
I'm sure people are gettingtheir pads out and they're
taking notes, you know.
Speaker 3 (21:44):
Yeah.
So for the SEO piece, takingnotes, you know, yeah, so for
the SEO piece, I will tell youand this is backed up by our
data the lowest hanging fruit istitle testing.
So obviously you need to have afull, rounded out SEO copy on
your bullets, on yourdescription.
Your description is not goingto show up on the front end, by
the way, once you upload yourA-plus content, but it still
(22:05):
needs to be there on the backend.
You need to be thinking throughwhat that SEO strategy is.
We use a tool called Data Dive,which I am a huge fan of.
I love it.
If you use any of these tools,by the way, my code if you want
a discount is mindfulVIP, so goahead and give any of these a
shot with that code.
But Data Dive I love because weused to do a lot of things
(22:29):
manually in our agency.
One of them was what we wouldcall like what they call battle
of the titles.
For us it was our titlecalculator.
It was a spreadsheet where wewould go in and do all the
research to pull keywords from,let's say, the search query
reports in your business reportsin Amazon.
We would pull your masterkeyword list.
We'd pull any keywords thatyou're ranking for right now,
that you're indexed for, and anythat your competitors are
(22:49):
indexed for, and then we wouldkind of cross-reference those
against what's converting wellon your advertising PPC.
So we kind of build this masterspreadsheet of keywords, right,
and then we would try tounderstand what is the reach of
all of these keywords and how dowe align, how we're putting our
copy into our listing in such away that it's increasing the
(23:12):
reach, the potential reach forthis customer.
One of the best ways you can dothat is in your title, and so
your title can actually improveyour sales.
I want to say it's like up to550%.
We have a, if you go to mindfulgoodsco, we have a metrics tab
that shows you every quarterwhat we're releasing for our
split tests, and so I will tellyou right now for the SEO copy
(23:36):
oh, I was way off you canincrease your sales, your orders
, by 990% by doing title testing.
So highly recommend doing this.
This is the lowest hanging fruitthat you could possibly do in
terms of activities.
It doesn't require a graphicdesigner.
It requires, yes, you sittingdown and understanding the
(23:59):
keywords that you're using.
But we just worked with aneight figure brand and they have
been selling for more than 10years on Amazon, and they have
been selling for more than 10years on Amazon and they saw
over 600% increase in sales.
And this was one of theactivities that they were
shocked by.
We record videos and we showthem what we did, so they're
along for the ride as we'remaking these adjustments, and
(24:20):
they were just like wow, we hadno idea how far short we were
falling on our SEO, on our SEOprocess, you know.
And so with data dive, itbrings you through a series of
steps where you're reallydialing that in, and one of them
is what they call battle of thetitle.
So lowest hanging fruit.
Highly recommend doing that.
(24:41):
The second piece, if we want todive into that, is the main
images, which I explainedearlier.
Our process has changed quite abit in the last couple months
and now we have an AI toolthat's trained on our best
performing images.
That is really about thinkingthrough the all of the crazy
ideas that you have on how yourproduct could look in that first
image on white and how you cancapture eyeballs quicker, better
(25:05):
, faster, than anybody elsearound you on that page.
So if you go, look for yourproduct and you see 50 products
on the search page, or at leastthe top 20, go and look at what
they're doing.
And then go look in othercategories and see what they're
doing, and this is the point atwhich you want to ideate there
is no bad ideas.
You want a million ideas andthen you're going to kind of
just create them all with AI andjust kind of like look at them
(25:26):
and be like, okay, which ones ofthese do we realistically think
are good options for our mainimage?
Grab a few of the top ones,polish them up, throw them into
a split test and with that youcan improve your sales by 550%
with just that one piece right.
So again, low hanging fruit onthese very simple tests.
And that final piece, the Aplus content.
(25:48):
Once you're inside the listing,what we recommend, there is a
landing page style format.
So now Amazon has what's calledpremium A plus content.
Any brand who's brand registeredselling on seller central can
get this unlocked on theiraccount.
It's something that vendorcentral brands are paying.
We have clients that are payingalmost a million dollars for
(26:09):
access to this and it'scompletely free for brands that
are on Seller Central and thatare brand registered and anyone
can get access to this.
Okay, so the only requirements,just so that everybody knows,
because I feel like I'mscreaming from the rooftops
about this and like 50% ofbrands that come to us still
have not unlocked this.
But it is free to you and youshould unlock it and get it
(26:31):
while you can.
It is the only thing you haveto do to qualify to get it is a
brand story.
You have to upload a brandstory, which, by the way, you
don't have to have anythingfancy.
Put a placeholder image, put alifestyle image and submit it.
So the Amazon approves it, andthen it's going to ask you which
ASINs you want to put this onor which SKUs you want to apply
this.
To Apply it to every single SKUin your catalog.
(26:52):
Submit it, leave it there.
The only other thing you haveto do is you have to have
submitted I think it's 10submissions of A plus content.
This part's changing.
It might be five now.
It used to be like 20.
And so, really, what you wantto do here, it doesn't mean you
need to have 20 SKUs.
It doesn't mean you need tohave 10 SKUs.
It doesn't even mean you needto have five SKUs.
You could have one product andall you have to do is submit A
(27:16):
plus content, which again,doesn't need to be what we do.
It doesn't need to be a landingpage style A plus content.
Just put up a banner, alifestyle image and submit it on
one of your ASINs and then letAmazon approve it.
Once they approve it, you'regoing to duplicate and submit.
Do that 10 times.
And then within a week, this newfeature in your account is
(27:37):
going to be unlocked for you,called premium A plus content.
So that lives below the foldwhen scrollers are shopping.
It's the thumb stopper rightabove the review.
So it's all of those reallypretty big banners that act like
a landing page.
That can be a huge conversionpiece, and so for us, we think
through that exactly as youwould think through your landing
page on your website, andthat's the piece that can get
(28:00):
you up to 1,008% increase insales.
Wow.
Speaker 2 (28:03):
That's a huge upgrade for everyone that's listening.
Could you, Daniela, describe ordefine what is A-plus content
from Amazon's perspective andhow would you look at that?
Let's dig into that a littlebit more.
Speaker 3 (28:18):
Yeah, so A-plus content.
So there's two pieces insideyour product page.
Once they click to go insideyour listing, the product image
carousel is what you see at thetop.
So you have your hero image andyour other product images and a
video.
Then you scroll down becauseyou want to go see the reviews
and there's these other bannersthat live right below the fold
(28:39):
and sometimes and this is whatbrands used to do in the past
brands would just upload justrandom stock images there and a
bunch of text because you couldindex really well in the past
with that section.
Amazon has come out in in in thelast few years saying like
that's not the intent.
The intent of that area is toreally showcase your brand.
(29:02):
So it could be cross-selling,could be educating them on how
you've been you know a legacybusiness around for 45 years.
It could.
There's so many positioningthings you can do there.
But the main point is thatthey're giving you real estate
on your product page to reallyconvert and to to bring the
shopper through something,because there's so many
(29:23):
distractions on that productpage.
Like, if you go look at aproduct page today, there's
advertisements all over the page, so you have to treat it
differently than you would treatyour Shopify site.
You know like you really haveto get the attention, and that
is the spot to do it.
It's the one page, it's the onesection that Amazon is giving
you to put full scale video.
You can have carousels, you canhave hotspots, and so it really
(29:45):
does function very similar to alanding page that sits within
your product page in the middleof all that distracting
advertising.
Speaker 2 (29:52):
So you want to make the most of it that we haven't
really covered, that you wantedto make sure to share about
Amazon specifically.
(30:14):
I think that you've given peoplea lot to think about a lot of
different tools.
There's really no reason whyyou shouldn't be doing split
testing now with large languagemodels.
It's so easy to do.
There's tools coming out allthe time.
This is the magic sauce thatyou've wanted for a long time
and you should start using it.
And when now you're talkingabout the synthetic data on the,
the, the target personas,that's going to be incredible,
(30:36):
because before you even spend adollar, it was kind of that's
kind of like trading stocks with, with the play accounts, right,
like it's like this is so sogood.
You don't have to spend themoney to get the data.
Speaker 3 (30:49):
And there's a tool out there right now that had
that about a year ago and theywere paying.
They were there, theypositioned themselves as an
enterprise tool and I knowbrands right now they're still
using it and that tool.
I mean we're talking about likehundreds of thousands of
dollars and you're going to beable to get that same level soon
.
Like I mean, you can build iton your own right now as your
(31:10):
own custom GPT.
You could do something reallysimilar, but there's going to be
softwares that are going tooffer this that are just going
to make it so simple for you.
Speaker 2 (31:17):
Yeah, they're going to.
They're putting wrappers aroundeverything that.
That's actually what a lot ofthese tools were like um uh was,
was uh just a low grade AI orLLM wrapped around, and that was
the thing to do.
I think that everything's kindof out of the box now, Even
traffic's like left the website.
You were talking about Shopifyand I just wanted to kind of
(31:37):
hear your perspective, because Italked to the brands now and
then that have an e-commercecomponent and like I actually
have some equity in a few, and Iwas like we need to be on
Amazon.
No, they take too big a cut,Like.
I've heard different storieslike and I'm like, okay, you
could mark it up and thenthey'll knock it back down, but
(31:59):
Amazon's like the biggest searchengine for this.
So I've had like a lot of thesedebates so with other
properties that are out there,how do you view them?
Speaker 3 (32:09):
How do you view the ecosystem?
That's such a good question.
Okay, let me give you my takeon it.
So I think you have to do both,and here's why.
So my personal use case I livein Peru.
I'm in Peru right now, but Iwas in New York a few days ago
with the team at Amazon.
And when I go to United Statesat Amazon and when I when I go
(32:32):
to United States, I do my haulright.
So my behavior is my behavioris I already?
had my shopping cart, yeah, preloaded, right.
So I'm about to board my flightand I'm like everything that's
prime, boom, get to the hotel.
So it's waiting for me, right.
When I get there.
If you are not prime and you'reon Amazon, you're not making
the cut.
For instance, I needed shoesbecause I'm going to a wedding
in Italy and I was like, okay, Ineed these specific shoes, but
(32:57):
I need them to either be at myhotel or at my mom's house in
California so that they make itto Italy.
What am I going to do, right?
So I'm trying to think through.
So I go to the Sam Edelmanwebsite.
Am I going to do Right?
So I'm trying to think through.
So I go to the Sam Edelmanwebsite.
They don't have any sort ofguarantee that they're going to
be at my hotel by a certain date, right?
(33:19):
So that's a lost sale.
I wanted a specific toothpaste.
I gave this example the otherday on another podcast.
I wanted a specific toothpasteand this might be TMI, but I
really wanted a.
I use a tea tree oil toothpaste.
It's by very specific brand.
I know they sell on Amazon.
They didn't have prime to set,set up to deliver to New York in
time before I left and I wasonly there for two days.
So they lost the sale rightJust by not having their product
(33:41):
in stock in Amazon's FBAwarehouse.
Am I going to go to theirwebsite to buy my toothpaste?
No, I'm not, because I don'tknow for sure that their website
will get it to me on time,right, and so it's like that is
the Amazon shopper.
I am the epitome of the Amazon.
This all about convenience,right, like I absolutely wait
for the last minute.
(34:01):
I put all my orders in.
They get to the hotel and ifyou don't have prime and you're
not on Amazon and you're notmaking the cut, you lost the
sale Right.
So that is a very specific buyerand there's a lot of us right
Now.
If I lived in California, stillI may have a little bit more
time and I might actually takemore time to go to people's
website.
But if I'm a busy mom and I gotto, I got to take my kids to
(34:23):
school and I got to walk the dogand I got to do all this other
stuff at home I don't have timeto be sitting around on websites
and figuring out like when I'mgoing to get the thing that I
want.
Like I just need to know, likethat's what I need.
It answers all my questionsreally quickly.
I know I'm going to get itcause it's prime and done, done,
just like clear that mentalload, you know.
Speaker 2 (34:41):
Yeah, well, I think that the ecosystem has has
changed the game, because youdon't have to set up multiple
passwords, you don't have to login, you're not putting your
information all in one place,it's all in the ecosystem.
I mean Amazon.
It's been a process to move tolike an AI first company, but I
mean it's changed the game andit's the power of the network
(35:03):
and as it grows, and to yourpoint, I mean I sometimes of the
network and as it grows, and toto your point, I mean I
sometimes I live in Houston andso, um, you know, if I go
somewhere else, um, it's notlike it's delivered within a
couple hours, but like I orderstuff and it's here.
It's scary quick, right, it'sscary quick Cause the, the
warehouse around the corner orwhatever, and I mean there's
(35:25):
Amazon deliveries, you know, onour street every, every day, you
know, maybe multiple times aday, and, um it, it has changed
what our expectation is.
A lot of times, if I ordersomething on a different website
, man, I'll forget that I evenordered it and it comes in the
mail.
I'm like what's this?
You know?
Speaker 3 (35:41):
Um, and now the other thing, though, to consider is,
like, from the brand'sperspective, I get it.
Why you, your margins arebetter on your own website,
right Like, I get that.
So there is a way tomerchandise things differently
on different platforms, rightLike?
Maybe you wouldn't sell theexact same bundle or size that
you would on Amazon, and youdon't want Amazon doing price
checks against any of your otherretailers or your own website
(36:03):
because they want to have thebest price too, right?
So there is a way to kind ofpackage things up and
merchandise in such a way thatthe Amazon experience is its own
thing, separate from yourwebsite, and that's something to
be considered as well.
Speaker 2 (36:16):
Like what I found brands doing is they offer the
same price as Amazon and thenthey offer coupons and stuff
like that on their own websiteto bring down the price.
There's a lot of that that haveto, I guess, be played and from
.
Speaker 3 (36:34):
Shopify, you need, like you need to push up that
average car value too, you know,whereas, like on Amazon, if
you're, let's say, like avelocity product, like a CPG
product, you could focus more onlike subscriptions and getting
subscribed and save buyers, youknow, and pushing that up.
So it's like probably differentKPIs depending on the channel.
Speaker 2 (36:52):
Yeah, you just got to model it and figure out what.
What makes the most sense, butto, not to, not to, to not open
up Amazon is.
I mean, how much Danielle isAmazon of the like overall
market, Like how, how big of apercentage is it?
I don't know Is.
Speaker 3 (37:08):
Amazon of the overall market.
How big of a percentage?
Speaker 2 (37:09):
is it I don't know, of online sales, yeah, or is
there any metrics you have ofhow big of a piece of pie that
is that people want to walk awayfrom it?
Speaker 3 (37:16):
Yeah, I don't want to quote it, but I think it's over
50% of online sales, that'sincredible.
I don't want to misquote that.
But there is really strongfigures out there, that I mean
Amazon is the biggest of onlinesales.
Speaker 2 (37:32):
Yeah, well, awesome, well, this has been great.
So if we had to synthesizeeverything that we were talking
about and maybe surface onething of like what is the one
unknown secret of internetmarketing that you want people
to walk away with, what wouldthat be?
Speaker 3 (37:50):
Oh, don't put all your eggs in the advertising
basket.
Really focus on conversion soyou can get more out of each
dollar.
Speaker 2 (37:58):
I love that.
I love that.
All right, danielle, how dopeople get in contact with you?
Follow you?
I know you speak.
Where are you active on socialplatforms so people can find out
more about your stuff?
Speaker 3 (38:08):
I'm a little bit of a big deal on LinkedIn, so you
can find me there, but also atmindfulgoodsco if you want to
chat with us.
We have a fun thing called amini audit, so if you want me or
one of my creative directors tolook at your listing, we record
a video and give you all of ourtips.
You can either hand it off toyour team or hire us to do the
work, but I recommend checkingthat out.
Speaker 2 (38:27):
All right, Well, awesome.
Well, guys, if you thoughtDanielle was a good guest and
you would like more guests likeDanielle, please hit the
subscribe button.
Please follow us, share like us.
We are Internet MarketingSecrets now on YouTube, so we're
going to get that up and goingthis quarter and just appreciate
all your support.
My name is Matt Bertram.
This is the Unknown Secrets ofInternet Marketing Until the
(38:57):
next time.
Bye-bye for now.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
My Favorite Murder with Karen Kilgariff and Georgia Hardstark
My Favorite Murder is a true crime comedy podcast hosted by Karen Kilgariff and Georgia Hardstark. Each week, Karen and Georgia share compelling true crimes and hometown stories from friends and listeners. Since MFM launched in January of 2016, Karen and Georgia have shared their lifelong interest in true crime and have covered stories of infamous serial killers like the Night Stalker, mysterious cold cases, captivating cults, incredible survivor stories and important events from history like the Tulsa race massacre of 1921. My Favorite Murder is part of the Exactly Right podcast network that provides a platform for bold, creative voices to bring to life provocative, entertaining and relatable stories for audiences everywhere. The Exactly Right roster of podcasts covers a variety of topics including historic true crime, comedic interviews and news, science, pop culture and more. Podcasts on the network include Buried Bones with Kate Winkler Dawson and Paul Holes, That's Messed Up: An SVU Podcast, This Podcast Will Kill You, Bananas and more.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com