Best scripts for Google Ads and eCommerce and how to use them | Next Level eCommerce | Channable

3 2 1

>> Welcome to Next Level E-Commerce, the

webinar scene where top e-commerce

experts share their advanced tips and

tricks.

>> Yes, welcome everyone. Good that you're

all here. I see actually quite some

people again that already uh were here

the last session. So, welcome back. And

for the people that are here for the

first time, welcome to Next Level

E-commerce. Uh today we actually really

have a really nice guest, but uh I won't

introduce him yet. Uh let me first start

with myself. Hi, I'm Juul. I'm the lead

for customer success at Channibal. So

mainly looking into all the yeah direct

uh customers and agencies that we have

and basically see on how they are

successful and that's why we're here as

well to basically give you those

practical tips to uh to help you out and

to be successful in the end in

e-commerce. Uh that's why it's called

next level e-commerce and of course like

the person today that we have here uh

had to be in this session as well. Uh

I'm doing that together with Yam which

he will introduce himself in a bit the

lead for our ads uh team and that's why

we're usually a lot focused on on Google

ads. Um so welcome everyone. So as I

said like these webinar series are

really focused on expertled insights and

best practices. Uh not only showing you

the why around things. So why should you

do more with scripts? Uh but also the

really how and I think often that's

missing and that's what we actually want

to provide you so that if you're

stepping away from this session that you

actually have something you can do uh

which will directly lead into uh best

practices also for yourself uh because

that's in the end what we're doing it

here for and I think this is a nice

stage uh to do so. Um but we do have

some house rules because we have Nils

here. He prepped like a really great

session for you. I already have seen it.

Um but it also means you put time into

it. So we would really appreciate you uh

shutting down all other things that can

distract you. Uh also the best for

yourself because the more you look into

it the better uh you will you can use

the content actually to also get those

best practices out there. Um besides

that we do have a questions tab. So if

any question arise and I know for sure

people will start asking it in the chat

already but here again we do have a

questions tab. So if you put your

question there then we can upvote the

questions and with we I mean the whole

audience basically meaning that we can

also ask the most upvoted question to

Neils. Uh so please bear with me uh if

you can ask the questions in the

questions tab and remind everyone that

ask the question in the chat uh to put

it in the questions tab then you will

help us greatly as well uh and yeah make

sure that we can ask the best questions

to Neils as well. Uh and again uh time

is precious I know but we try to always

give you the best content. So even if it

would take a bit longer uh than we

expected then we like to uh yeah extend

it to make sure the best content is out

there because everything is being

recorded uh so and you're also getting

this recording uh which will help you in

the end to also get the full context out

of it. um last session. So this is

actually a nice follow-up on last

session where uh Austin mainly uh dove

into search and why it still matters uh

and why under 20% non-brand search

revenue is basically leaving money on

the table. Uh so if you want to see more

about that you can still watch the

recording for it as well uh because he

greatly also uh shared a bit of the Pmax

plateau problem problem and actually

Neil's also going to share a bit about

Pmax and at limitation it has um and

it's a great session with very concrete

tips on how to basically uh yeah win and

I must say with search there's a lot to

win because people are getting more lazy

which means with a bit of extra focus

you can get more money out of it. Uh and

then we are coming to the introduction

of the speaker of today and uh now it's

time for Yam.

>> Yes. Hi everyone. Yam here product lead

at general. Um you also see me back in

the Q&A afterwards. So please leave your

questions. I will moderate them. Um it's

an honor to uh introduce our guest Nils

Roys. Um he actually doesn't really need

an introduction. He's already voted as

top 10 PPC influencer. Um, so most of

you probably know him already, but I

still do it. He's um I think his career

started 20 years ago in PPC, even before

uh Google Ads was a thing here in the in

the Netherlands. Um, he worked at the

platform site even at ILZA was the

company like the search engine before

Google. I even used it when I was young.

Um, and now he fully moved into the

agency world, owning a fully remote

agency uh in the Netherlands in in

Amsterdam from his home boat, if I say

that correctly. Neils, maybe Oh, he's

even showing it. It's a little small,

but um that's looking

very very nice, except the weather.

>> The house boat.

>> The house boat. Nice. That's really

really nice.

Um

yeah, I think next to that I think

people mostly know you from um yeah

really being an ads or like an a script

guru. You also are going to show that

today. Um you're a big collector of

scripts even uh on your website. There's

like over 400 scripts in your on your

list. Um so that's really a lot and

today we ask you to show a few of them.

Um

and then except the the mostly the ones

that are used for Google uh for

e-commerce customers and which help you

really scale those those accounts.

Then there is a poll right now about if

people are using it.

I see most the people about 70% is uh

use scripts a few times. Um

20% never use them and 15% uh are also

like real real script gurus. So I think

there's a lot to uh to learn today. Um

also good to mention Neil is not really

show explaining how to do scripting. We

already did some sessions about that in

the past.

Um, so you can watch that back. He's

really going into like the use cases

today. So with that, Nils, welcome and u

feel free to take the mic.

All right. Yeah, thanks for the uh

introduction. Let me quickly test if my

arrow keys are working. They are. Okay.

So today I would like to actually show

you how I use script, not how not so

much how I create them. And I'd like to

use a case study to illustrate uh a

standard campaign type setup that I've

automated with the help of script to

increase the results for one of my

clients. So this is a case study. The

client was selling high-end professional

camera materials. So professional

digital cameras, lenses, microphones,

stuff like that. It's for professionals

and hobbyists with much way too much

money. So you can think of like digital

camera lenses that start at prices like

5,000 and upwards, right? uh it was a

relatively small web shop. He was

already selling it in his offline retail

shop, but he also launched a web shop

and there was a freelancer managing his

Google ads account mostly through uh

performance max campaigns. So he had an

inventory of roughly 50 products. Um the

account was doing okayish spending 10k a

month but he wanted to increase

profitability in the short term and next

to that also get more reliable results

because in the last 12 months he had

seen some months that were relatively

good but also some dramatic months where

he was just spending a lot of money and

didn't make any profit. He was actually

losing some money. So he wanted

predictable growth and increase in

short-term profits in the near run. So

that's when we took on the over the

account and we had a look at the uh the

current setup uh and we immediately

decided okay we're going to redo

everything from scratch.

And this is basically the setup that I

currently use in most of my e-commerce

clients to predictably grow results. So

instead of having this chaotic results

that performance max is generating for

us. Uh please let me know in the chat if

this sounds familiar to you. Performance

max goes up and down with its

performance. it's really hard to scale

it without uh well it's relatively hard

to scale it in a predictable manner. Uh

but a different setup that I will show

today allows you to do so and our client

was really looking for predictable

growth and short-term increase in

profits. So the first thing we did of

course was fix conversion tracking

because conversion tracking almost

always is broken. Uh we fixed the

product feed. Then we moved to a

standard shopping plus DSA campaign

structure. We had some datadriven

standard search ads that I will also

show you today and next to that later on

dynamic remarketing demand gen and only

only if everything has matured then we

uh sometimes layer on top some

performance max campaign. So today we're

going to dive in the first four parts of

this um setup and how I use scripts to

automate that for you.

So let's start with the uh conversion

tracking. Before we took over the

account, the client was tracking

revenue, not so much profit, but the

profit margins between varied widely

between a digital camera, a microphone

or a lens. Yet the the person that was

running the campaigns before threw it

all in one performance max campaign.

There was some segmentation based on

products, but definitely not on product

margin. Uh it was really hard to use the

ROAS targets to scale profit because of

that. Uh so we decided to also next to

the revenue tracking implement

conversion uh uh profit based tracking

based on a secondary conversion tag in

the Google ads platform. So since we

only had 50 products it was relatively

easy to use Google Tag Manager to do

this. So inside Google Tag Manager we

created a lookup table. So you can

basically create a variable, custom

variable. You insert a lookup table and

for each product we entered the cost of

goods sold. Right? Next to that we

created a function that would calculate

the total cost of goods sold for all the

items in a single order.

So for each order for each basket we had

the revenue side of things but also the

total cost of goods sold for all the

products that were in the basket. And

then of course you can calculate the

profit based on the total revenue minus

the tax minus the shipping cost minus

the cost of goods sold. And that value

is the profit for that specific order.

And that was communicated back to the

Google Ads platform through Google Tag

Manager as a secondary conversion action

so that at least we could still uh see

revenue, use revenue for optimization

purposes if we wanted to, but also

compare it to the actual profit that was

being generated in the account. So this

is a relatively easy and cheap for

actually free way to set up profitbased

conversion tracking in your account. Um,

I wouldn't recommend this for big

accounts with a lot of products and a

lot of transactions. Most probably you

would be better off with a more reliable

solution, a third party solution, maybe

even something like profit metrics or

some other server side tracking

solution. But for relatively smaller

accounts, this is a a great way to uh

set up profit based tracking for free.

Now another thing we noticed is that to

our surprise there was a relatively high

return rate on the items. So for this

client um he was selling these lenses

right and the lenses they had like small

variations like autofocus or

anti-trimming or there were just minor

differences between a similar product

the same lens but just slight technical

details that were different. And many of

the clients, they would order these two

variations in the same order. So we have

basically could predict that they're

never going to take both, right? They're

going to return the one that they're the

least um happy with. Um and the same

thing is also happening a lot of course

in apparel fashion, right? So if you

have an order basket where someone

orders like four variations of the same

dress, so two different sizes and two

different colors, you can it's

relatively easy to predict that most

probably three of these four items will

get returned, right? So again, you can

use Google Tag Manager to have a look at

the items that are in the basket at the

time of the transaction and then reduce

the total revenue value and profit value

by this by uh uh subtracting the order

value of the duplicate items. Now, one

trick to do so is to use the data layer

on your thank you page and insert the

item group ID for each individual

product that is in the basket. That way

you can relatively easy create a custom

JavaScript function that reduces the

total order value thereby increasing the

accuracy of your uh conversion tracking

because you you will have uh less

conversions that are being tracked with

uh values that in the end will get

returned.

Hope this makes sense. If not uh we can

we can chat later in the Q&A.

Then of course the product feed. Um uh

since this is a channelable uh session,

I I highly recommend using channelable

to increase your uh uh the quality of

your feed. But you can also use scripts

for that. Um I use scripts to send me an

alert if attributes in the feed are

missing. uh and I also use scripts in

combination with uh large language

models and their APIs to increase the

quality of the titles and the

descriptions and to fix something like

missing uh missing attributes like for

instance the product type. The product

type is also something that is gaining a

lot more importance inside Google ads

because Google uses the product type as

a way to transfer conversion data

learnings. the smart bidding that learns

from the conversion data on the

individual item ids two similar items

that have similar product ids product

types. So imagine that you have a

best-selling product with product type A

that goes out of stock and 20% of the

conversions of that specific campaign

were being generated by that specific

item. Now all this conversion data is

stored at the level of indivi this

individual item that went out of stock.

If you do not use a product type

attribute, you are missing out on the

opportunity of smart bidding to transfer

the learnings from that out of stock

product to other similar products with

the same product type. So I highly

recommend you always look at this uh

product feed attribute to read the

benefits of this transfer learning

inside uh Google ads smart bidding.

Okay, so like I mentioned, the account

was running performance max campaign

with sort of okayish results but also

not so much. And remember the client

wanted to increase the short-term

profitability.

Well, with performance max, as we all

know, if you give it a budget, it will

spend it. Unless, of course, you have a

very high target robots, but if you give

it a budget, Google will will try

everything it can to spend all of your

budget. And in the process of doing so,

it is exploring new growth opportunities

for you, which is a good thing, but it's

also wasting a lot of money on

completely irrelevant clicks, right?

Irre irrelevant search terms and also

clicks from the Google Display Network,

which of course is notorious for fraud,

lead fraud, click fraud. Um, yeah. So

performance max might do a good job in

exploring new growth opportunities if

the overall results of the the the

performance max campaign is meeting your

rowas target but it's hiding the below

average performance of all these

non-converting clicks from irrelevant

search terms and from the display

network.

Now our client wanted to increase

short-term profitability. So an easy

solution would be of course to get rid

of these irrelevant clicks

inside the performance max campaigns

that is relatively hard to do. Yes, you

can negate queries. Now Google

fortunately for us uh added the

possibility to add negative keywords but

that is behind the fact right. You can

only negate them once you see the data

flowing in uh display network. Yeah, you

can add ex uh placement exclusions, but

still if performance max is doing a

great job as in the average

return on ad spend is above your target

return on ad spend, it will expand its

search to the display network and it

will start showing your ads there and

most probably waste a lot of money

there. So, we wanted to get rid of these

irrelevant clicks. How do we do that?

Well, we simply migrate back to the

standard shopping campaign structure and

the standard search ads. Um, we're not

unique in this sense. Uh, this is a

graph that is being created by our

friends at Smarter E-commerce. This

shows the uptake of performance max

campaign over time. And as you can see

in the upper right corner, actually the

line is slightly declining these days

the last few months year uh as in more

accounts are moving back from

performance max to standard uh standard

shopping campaigns and there's a good

reason for it and I'll show you in a

minute.

Now a very important benefit of using of

an advantage of using standard shopping

campaigns compared to performance max

campaign is that standard shopping

campaigns has a feature that is called

prioritization.

You can run a standard shopping campaign

with high priority, medium priority and

low priority. If you're not familiar

with this concept of different

priorities and standard shopping

campaigns, I highly recommend you to

watch a video that has been uh shared by

Martin Litkaden where he explains this

concept in a lot more detail than I will

be able to explain to you today. Uh it

is a very very powerful concept and

especially relevant if you want to take

control and increase profits.

So this is the meat of my presentation

in the sense that this is the standard

shopping campaign structure that I like

to use when my clients prefer profit

over revenue growth. I call it the

gatekeeper architecture and it basically

uses these standard shopping campaigns

with the prioritization to sort of like

filter the search queries that generate

clicks with high conversion value as in

they land in the medium priority money

maker campaign and not spend a lot of

money on exploring new growth

opportunities in a high priority

gatekeeper campaign. So let me explain

the details of this setup. So on the

left part you see two block blue boxes.

the gatekeeper standard shopping

campaign with a high priority and a

money maker standard shopping campaign

with medium priority. Now, for the sake

of this presentation, we're going to

pretend there's only one single product

in there, one digital camera.

Both

campaigns have this one product and a

shared budget, but they run different

bidding strategies. The high priority

gatekeeper campaign runs either a manual

CPC with a very low CPC bit. So let's

say 5- cent, 10-centent, 15 cent or a

portfolio target row as bid strategy

with a bit limit. This is currently my

preferred way. The bit limit is also set

to a very low value. The medium priority

campaign that is the money maker. That

is why we want to set all these clicks

with uh a lot of conversion value. that

one is running a target rorowass uh

bidding strategy could be a portfolio

again with a bit limit but mostly I use

simply a target row bit strategy with a

um with a target.

Now the way we can funnel these

relevant clicks with high conversion

high predicted conversion value to the

moneymaker campaign is by using negative

keywords in the high priority campaign.

So we have this product that sits in

both the same in both the high priority

campaign and the medium priority

campaign. Google is trying to match the

user queries to that product and if the

high priority

campaign is able to match that query to

the product, it will take the bit from

the high priority campaign. However, if

there is a negative keyword in the high

priority campaign that is preventing

that user query from being matched to

the gatekeeper campaign, then it gets

funneled into the medium priority

campaign where we can place a higher

bid. So we can use negative keyword

lists to basically decide what user

queries, what search terms we would like

to land in the medium priority campaign

that has a high bit and all the other

search terms, they get the bit that is

attached to the high priority campaign.

So at the start of this setup, I always

like to create a negative keyword list

that I call the qualifiers. So for

instance, the GTIN numbers of the

products that I'm selling or these very

specific longtail um uh user queries

that are sort of like similar to the

title of the product. So very precise

brand plus type uh combinations in the

user query or something like a user like

buy digital camera, buy digital camera

for professionals online. A ridiculous

example with low search volume, but you

get the idea, right? So I add these um

high conversion value search terms to a

list of qualifiers that is being

attached as a negative keyword list to

the high priority campaign. Thus

funneling all the searches to the medium

priority campaign for these uh user

queries.

If you have this setup running, you will

get the search term data and these

search terms can be analyzed, right? So

you can have a look at all the search

terms from your high priority campaign

and if they have generated a certain

number of conversions meaning that they

have high conversion value you can add

them to a negative keyword list. I call

them converting search terms and that

negative keyword list will prevent the

high priority campaign from placing the

bid the next time someone searches for

that user query. It's get funneled to

the moneymaker campaign where we bid

higher. Right? So this setup allows you

to explore new growth opportunities in

the high priority campaign with a low

CPC bid and exploit the opportunities

that you've discovered in the medium

priority campaign with higher CPC bids.

Now the green parts they are automated

through scripts. So I have scripts

monitoring the search terms. If a search

term has had over two conversions, it is

added to the converting uh search terms

negative keyword list. And other search

terms that have had over let's say 100

clicks and zero conversions, they will

be added to a shopping negative keyword

list that is attached to all my shopping

campaigns. So that we do not show any

more products for these uh

non-converting search terms. So that way

I'm also reducing wasted ad spend in the

high priority campaigns.

Next to that, I have scripts running

that monitor the shared budget. So, of

course, you don't want to be limited by

budget if you're happy with the

performance, right? So, if the overall

performance, the combined performance of

these two campaigns meets my ROAS

target, I would like to increase the

budget if we see that we're spending

let's say 80% of the budget. So, the

script automatically increases the value

of the share budget. On the other hand,

if we see that the performance is so so

we're not really meeting our target,

that basically might mean that we are

spending too much money on

non-converting clicks in the high

priority campaign. So I have another

script that monitors the overall

performance and decreases the bits in

the high priority campaign if we are not

meeting the ROAS target. If we are

meeting the rorowas target or if we are

overperforming so the average rorowas is

above let's say 20% above our target

that means we have more room to explore

new growth opportunities. We are

profitable but

we like to grow uh the total number of

orders uh at a similar profit ratio. So

then the script will automatically

increase the bit in the high priority

campaign thus allowing us to explore

more grow more growth opportunities by

discovering more new uh converting

search terms.

So this setup allows us to scale this

account scale the shopping product ads

in a very predictable way and keeping

control of the exploration versus

exploitation balance.

Now, next to the standard shopping

campaigns, we also run DSA campaigns.

So, if you have a shopping feed, you can

use scripts to automatically create

single product ad groups for your DSA

campaigns. Now, the major advantage of

doing that is that for each individual

product, you can use the script to

automatically create highly optimized

descriptions for that specific product

in your single product ad group. Right?

You could of course run a DSA campaign

simply targeting all the pages in your

DSA feed, but then you wouldn't be able

to optimize all these individual

descriptions based on the data that is

available in your product feed. Thanks

to scripts, you can do so. Head tip to

these guys, Fred Vales and Machu Vont.

Uh they they basically uh presented this

idea in a in an earlier session in the

PPC town hall. Uh I loved it. I

implemented it. It worked like a charm.

Okay, so now you have the standard

shopping campaigns running. You have

your DSA campaign running. And of

course, both these campaign types

generate a lot of search term data. And

you want to monitor these search terms

and exclude the ones that are completely

irrelevant for your business. So I have

another script running that basically

reports search terms that have generated

over, let's say, 50 clicks or 100 clicks

and zero conversions. These are reported

in a Google sheet and me or my team

member we can go in the sheet and inside

the sheet we can basically decide what

to do with that individual search term.

So we could add it as a negative at the

level of the campaign the ad group or we

can add it to a negative keyword list

and we can indicate what match type to

use exact or phrase or broad. We can

also ignore them. So basically we can

inside the sheet we can say okay this

this search term we're not we're not

100% sure that we want to exclude them

so ignore this search term the next time

the script runs we don't want a new

suggestion we do not want to see this as

a negative keyword suggestion in the

future anymore and it simply adds it to

a glorist

now in one of my latest versions I'm

also using an AI to help us predict

um negative search terms

even if they have not generated a

significant amount of clicks yet. So

search terms that only have had let's

say 100 impressions and five clicks, if

the AI thinks it is an irrelevant search

term, it is also added to this sheet.

And the trick here is that the AI itself

already knows

a lot of things based on um uh the way

it has been trained by OpenAI or Google

by Gemini or Claw. But if you provide

more information about your specific

client in the context of your prompt,

then it does a much much better job at

actually generating these negative

keyword suggestions. So what we do is we

use our Google sheet to provide feedback

on the suggestions that have been

generated by the AI. So for instance, if

the AI suggests you don't want to show

for Canon Canon digital cameras anymore

because uh you're not selling the brand

uh Canon. Then you might give it

feedback as in you are right, we are not

selling anything of Canon. That way it

will learn okay in the future if a

search term contains the word Canon it

should be added as a negative keyword

suggestion. Uh the other way around

might also be that okay yeah you

suggested to add Canon digital camera as

a negative keyword but from our

experience the people that are looking

for a Canon digital camera might also be

interested in our product. Therefore we

would not like to exclude it. And if you

provide that kind of feedback in a

separate column in the Google sheet, it

is added to the context of the prompt

that we sent to the AI

to suggest a negative key. So basically,

it learns from our feedback. Now, I have

to be honest, this takes a lot of

entries and a lot of work to provide

feedback that it can actually use to

incre to significantly increase uh the

quality of its suggestions. But the

concept is sound. It does work. But you

have to convince yourself or your team

members to provide very good detailed

feedback on these negative keyword

suggestions.

Okay. So next to the DSA campaigns, we

can also use standard search ads in a

different manner. So we have again a

standard shopping campaign and the DSA

campaigns generating all the search term

data.

Then I have another campaign,

a standard search ad campaign

where I add all these converting search

terms as exact match keywords.

And inside that standard text campaign,

I run a initially a single ad group with

keywords

with dynamic keyword insertion in the

RSA ad copy and I insert the internal

site search landing page for that

specific keyword as a keyword level

final URL. So inside Google ads, you can

actually add landing pages to your

individual keywords. So not to the RSA

app but individual keywords and you can

use the internal site search of your web

shop the landing page for that search

you can use that as a keyword level

final URL and then I use a portfolio bit

strategy with again a target boas and a

bit limit.

Again I use scripts to add these search

terms as an exact match keyword. Thanks

to the dynamic keyword insertions, the

ad relevance is always good or above

average. And thanks to the internal site

search of the web shop, you always have

a highly relevant um landing page. And

again, of course, you can use scripts to

uplate the update the bit limits for the

portfolio bit strategy.

This works like a charm. So here's an

example of a list of exact match

keywords that have been generated.

Again, the result uh the landing page is

the internal reserve uh result page for

the for the site search. And as you can

see, ad scores, the quality scores are

just amazing, right? You don't have to

do any work, but you always have a very

relevant ad copy and the landing page

experience is well, if if you have a

good site search, it's really good.

You can also use scripts to improve the

quality of your feed. Like I already

mentioned, um, one easy way to do

increase the number of impressions and

clicks and those conversions is to

simply add your converting search terms

in your titles and your descriptions.

Right? So I have a script monitoring all

the search terms and for all the search

terms that have had over two

conversions, it checks to see if there

is a product that has that specific

search term in its title. If not, then

it suggests a product to add that search

term to.

Right? So for all the converting search

terms, we check if we have at least one

product that has the search term in the

title. If not, then most probably we can

find one product that is very relevant

for that specific search term and we

simply add it to the title and then that

is added as a supplemental feed in the

Google Merchant Center thereby

increasing the quality of your product

titles where you get more impressions

for search terms that actually um show

conversion value.

Now, so more examples of how I use

scripts in this uh in this setup. Uh

normally the Google Merchant Center has

a feed update limit of the frequency of

maximum once per day. If you have like a

very volatile inventory where products

go in and out of stock and you make a

lot of changes to the product data, I

always recommend to increase that uh

frequency to every hour. Uh you can use

script for that. You can also use CSS

partners for that. But uh on my website

you can find a script that will increase

the update frequency to every hour

instead of the maximum inside er center

of once a day. Another very important

script is the out of stock alerts. Like

I already mentioned, let's imagine you

have a campaign where 20% of the

conversions is being generated by a

single product that suddenly goes out of

stock.

What do you think that will will do to

the performance of that campaign? Right?

So 20% of conversions for that campaign

was being generated by one product that

went out of stock. The next few days the

conversion rate will probably plummet.

Right? Because that product that

campaign is not advertising that product

anymore. And if the conversion rate

drops, then smart bidding takes some

time to learn that the conversion rate

that the performance for that campaign

is being uh reduced dramatically. So the

smart bidding will continue to bid at

roughly the same level for the next few

days spend at the same level but your

conversion rates will have gone down and

thus your overall performance will have

gone down. This is this is a big thing

for for

performance max but also for standard

shopping campaigns that run on smart

bidding. So if you have a campaign where

a significant number of conversions was

being generated by conversion by

products that went out of stock, you

immediately want to know and you want to

tell Google that you expect a lower

conversion rate for the next few days.

You can do so by using seasonality

adjustments. But how do you know when

when this happens, right? We are not

monitoring every each and every

individual product every day to see if

it's if it has been generating a large

number of conversions and now it's out

of stock. Again, use scripts for that.

So, this script basically does that. It

tells us if there is a product that has

a significant number of conversions and

it went out of stock.

The same goes for disapproved products,

right? So, Google has these guidelines

about what products you are allowed to

advertise. These guidelines, they are uh

Google is using AI to help them uh u

make sure that all the products are

within the guidelines. But as we all

know, AI makes mistakes, right? And I'm

not sure about you, but I in inside the

merchant center, there's so many

products that have been disapproved for

faulty reasons. U if that happens, you

want to be in the know and fix it or

alert Google that, you know, uh Google

has made a mistake and you want to

appeal and have a new review because if

a product has generated a lot of

conversions and it is being disapproved,

you have the same issue that you would

have if the product went out of stock,

right? So, you want to be in the no

immediately.

And of course, no-brainer, exclude

non-converting products or add them to a

low priority campaign with a very low

CPC bid. Um, you can use scripts for

that. Basically, this scripts monitor

all products. For products that have

generated over, let's say, 200 clicks.

It creates a supplemental feed. Um, uh,

this sheet says custom label 5. You need

to change it to a custom label that you

are not using already. custom label 0ero

to4 and then you can use that in your

campaign structure to filter these

products. Either exclude them completely

from all your campaigns or add them to a

low priority campaign with a very low

bid so that you're not wasting more

money on this uh these products.

Now, I'm going to tease you a little

bit. Uh I just showed you the nice canal

that I'm overlooking right now. Uh I

live in Amsterdam. I love Amsterdam

during the summertime. Not so much in

the winter because now the leaves start

falling from the trees. It's getting

windy, rainy. Uh so I leave to Dutch

Caribbean island called Kurasau close to

Venezuela. This is the view from my

office. Right.

Imagine me sitting there in my flip flip

tops and my speedo

and there is this product. I'm let I'm

going to let you guess this product that

is being sold by one of my clients. It

is a mouse mat heater. Now, of course,

me sitting there with this view, I could

never imagine it's December, this

product being a trending product, as in

this product was gaining a lot of

momentum both in impressions, clicks,

and in conversions. If I hadn't had not

been running a Google ad script, I would

never never noticed this. But thanks to

the script, I did notice that this

product was gaining traction and I could

separate it into a a new campaign so

that we can reap the benefits of the

seasonality trend by giving it its own

budget and its own target, right?

Because the weeks before this product

was simply doing nothing. And if you

separate it in a separate campaign, you

can actually tell Google, okay, let's

let's let's go after the opportunity

here.

Now, all these scripts combined with the

setup allowed us to replace a

performance max campaign structure that

was sort of okayish but not generating

enough profit and was very unstable in

its performance like choppy results

every month. We completely replaced that

with something that was more stable and

showed a steady increase in short-term

profitability. So, the client was was

happy. But then the question becomes how

far can we take this right? How how far

can we increase these profit? Um this is

a a chart that shows you the

relationship between your ROAS target or

your POA target if you're tracking

profit and the actual profit that you be

that you would generate. So on the

horizontal axis you would see your

target. Let's say you you start with a

relatively low target. It could even be

below your break even ROS target. Then

of course you would bid very high gener

get a lot of impressions and clicks also

conversions but these conversions

wouldn't be profitable because you're

below your break even target right

then you start increasing your ROS

target and as a result the CPC bits will

also go down. You get less impressions,

less less clicks, less conversions, but

the profitability per conversion goes up

there therefore increasing your profit,

right?

But if you increase your lowest target

too much, you might start losing a lot

of profitable conversions. So instead of

100 conversions a week, you would only

get 50 conversions a week. And although

the relative profit margin per

conversion is much higher because of

your higher ROS target, the absolute

total profit starts decreasing because

the number of conversions is decreasing.

Now the shape of this curve is very

important and it differs per client per

campaign. So you would have to test it,

right? You don't know the shape of this

curve up front. You don't know where the

maximum or where is the target that

generates the maximum profit. you have

to experiment and test it. So the way I

do it is I use custom columns to do so.

So I create a custom column named gross

profit after ad spend.

And that custom column basically tells

me uh yeah it's basically calculated as

the revenue minus the shipping cost

minus tax minus the product cost minus

the ad spend. Right? So that is the

gross profit after ad spend or some

would call it net result.

And then I run an AB test with my

standard shopping campaigns using a

geosplit. So I have geo region A, geo

region B. And then I use different

targets for my smart bidding algorithm.

And I monitor the performance in my

gross profit after ad spend column. And

as you can see for this specific client,

a higher target rows or BOAS because we

are tracking profit here. So the target

draw us of 150% outperform the target

draws of 130%.

Does uh yeah basically telling me that

you know if we want to increase the

profitability our target should be

closer to 150 than to 130 and most

probably I would create a new experiment

where I would test 150% versus let's say

180%. To see which uh target growers was

generating the most profit. So this way

you can sort of like test and experiment

with the shape of the product curve.

So the client was happy. Before they

were spending like roughly 10k a month,

hardly break even and very hard to grow

in a predictable way. But after eight

weeks already uh we actually cut down on

irrelevant spend. Uh but we were

definitely profitable. Uh and we were

able to grow the account in a very

predictable way. Um which uh which made

the client very happy.

Now, the good news is you can do this

too, as in I've showed you the setup.

I've showed you examples of how I use

scripts. You don't have to be a

developer. You don't need any coding

skills to use these scripts. Some of

them are for free. You can get them

through my website. Others are paid for.

Uh you can send me an email or you can

simply create them yourself using my

custom GPT the script sensei. Um yeah,

and I hope you start experimenting with

that. All right. Thanks you for your

attention and uh now let's open it up

for uh for questions.

>> Great. Thanks a lot Neils. I actually

quoted or I will quote some comments

actually in the chat. Uh I actually

already found it really really

interesting because there are too many

things that uh you also told where my

brain uh almost exploded on things that

I have to look into myself as well. Uh

but I was not the only one uh because

Alli for example said my brain hurts.

This is so interesting. So you're even

giving people headaches which maybe

wasn't your purpose but you did so but

on a positive way.

Another one. This has got to be the most

valuable webinar I've seen so far.

>> Oh thank you.

>> So I would say thank Zeon for it. Uh and

thanks Neil's great webinar and I think

this great you deserve the sun. So maybe

you should go to Kurissau anytime soon.

Nice. Uh and yeah, I completely agree.

There were so many good things in it

from the mouse mat warmer uh sitting in

Kuro to conversion tracking for free to

using AI for negative keywords and to

name many more. I actually have a lot of

questions but uh I feel we should open

it up for the uh audience to ask those

questions but as you know us we always

do that via Yam. So Yam what are the top

questions uh from today's session?

Yes. Uh thank you Nils also learned a

lot. Um

maybe for you Nils. Um I'm going to the

questions that are upfoded the most.

>> Sure.

>> So you can also read along. Uh the first

question is from Marco. Do you recommend

running standard shopping campaigns

along with PMX? If so, why?

>> Yeah. So like I described I personally

like to start with standard shopping

campaigns and only if that setup has

matured if I feel that we've reached the

limit of its growth and we have we are

already running dynamic remarketing on

display and demand genen if we want to

do more upper funnel only then I would

like to test performance max campaign

just to see if it can

uncover some hidden things that we

haven't uh discovered with our standard

setup and I'd like to limit the

performance max campaign to the

bestselling products.

>> Okay. And there is I think a follow-up

question from Marott. What is your idea

about PMAX shopping only campaigns or

feed only campaigns?

>> Right. Yeah. So again uh I prefer the

standard shopping approach and the

reason for that is that compared to

performance max you have much much more

uh control um be also mostly because of

the prioritization within the standard

shopping campaigns. You can literally

filter and prequalify the search terms

uh before you get the click. That is not

possible with performance packs or

feebased. Uh and another big advantage

of standard shopping campaigns is that

they won't show on the display network.

Uh performance max will show on the

display network if Google feels like uh

it's yeah it feels like it.

>> Got it. And do you think it's future

proof?

>> Sorry.

>> The focus on standard shopping. Do you

think it's future proof?

>> Uh future proof. We never know, right?

As in Google is making changes to the

platform every day. Uh but I don't see

standard shopping campaigns disappearing

anytime soon. Uh simply because of you

know Google still making a lot of money

thanks to them. Um, also what I've

noticed is that Google tried to

introduce performance max campaign for

uh, a couple of years ago as the one and

all campaign type to rule it all, but

they are backing off from that statement

and that philosophy as in they realized

that, you know, the campaign type isn't

doing what they thought it would be able

to do. Um, so I think standard shopping

campaigns are going to be here for a

little while.

>> Good to know. then we uh this webinar

will be useful in the future.

Um a question from Abby is the

gatekeeper architecture scalable for

large feeds or would you mainly use it

for small feeds?

>> Yes, it's definitely scalable for large

feed but you would have to most probably

would have different pillars as I call

them for different

product categories or product types or

you you would have to think

strategically about the campaign setup.

it gets more difficult of course little

bit more complex uh but you would also

need to do something similar with BMAX

so um yeah so it's definitely you can

definitely use it with multiple products

in different product categories with

different margins

>> yeah nice then a question from Yans do

we get the scripts Neil is telling about

I added the link already are all the

scripts mentioned in that you're

>> the free ones are mentioned uh at the

landing page of uh I think the link that

you shared, right?

>> Yeah.

>> Yeah.

And for the rest uh feel free to uh to

use either my custom GPT to create it

yourself or find me if you are

interested in a paid version.

>> Nice.

Um another question from Abby. Does the

semi-automatic

add negative keyword script include pmax

search terms or only search?

>> Uh both. Yeah. So the paid version is

both for pmax and for standard campaign

types.

>> Um

then a question from Paul. Uh, regarding

the product type, would you recommend to

aggregate items? So, clothing, shoes or

be as specific as possible? For example,

clothing, shoes, sneakers, uh, and the

same for boots. Uh, for categories where

performance and margins are about the

same.

>> Let me reread a question. I'm not sure

if I understand.

>> Yeah, you mentioned it at the beginning

of your presentation.

Um,

I'm not sure if I really understand the

question, but the idea would be that

each individual product has a product

type and similar products would use the

same product type. So, for instance, if

you have uh let's stick to sneakers. Um,

if you have sneakers that are for trail

runs, the product type could be trail

run sneakers versus sneakers that are uh

not specifically for trail running.

Right, you get the idea. Hopefully.

Okay. Um, I think the question from

Michelle is answered.

A question from Mari.

My questions are flipping around.

Uh I believe the question was about DSAs

like with AIA max. Um do you still need

DSAs or would you move to AI max?

>> DSAs.

>> AI max is still uh

experimental just like performance max

in the early days. Google is constantly

making changes which makes it very hard

for us PPC professionals to optimize for

it. And of course the machine learning

behind AI max is still learning just as

it was you know during the early days of

performance max. So with these type of

new

campaign types or new beta features in

the platform my approach is to always

test it always be testing but only spend

a very small percentage of your budget

on these new campaign types or features

and let the other advertisers pay for

Google to learn and fix things.

Nice. And let's uh do one more question

and I think uh then it's time to round

off uh for the other open questions that

are there. We always try to answer them,

but we have to double check with Neils

if he's okay to also answer the other

ones, but we'll then share it later on

uh with you. So, one more question.

>> Yeah, I'm getting through the

>> if you have any and otherwise

>> um

Um,

I'm just looking at questions we didn't

answer. Um,

the 40 gatekeeper structure. It's

actually a question for myself like you

mentioned you're using uh a few scripts

for it. How do you track all those

scripts? like if you not sure how many

customers you manage but let's say uh 10

20 customers how do you manage all those

scripts across all those uh customers?

>> Yeah, great question. Yeah, we we we

could do a whole webinar about that as

well, but uh I can make it very complex,

explain how I do it or I can make it

relatively simple. And the simple answer

is simply to start running them and then

you will discover for yourself how easy

it is to make it client specific and

keep keep a script setup that works for

your specific situation.

Uh

it might feel overwhelming at first, but

if you start running these scripts in

your client accounts, you will discover

that it's not that hard. And if you

transform transform them into a Google

Ads manager account scripts, the MCC

level scripts, uh you could you could

yeah basically

it's it's not easy, but you could do it

with the help of my custom GPT or ask me

to do it for you. And then you can run

them from your manager account and

easily manage all these scripts from a

single Google sheet where you can

basically indicate this script should be

running in this client account with this

email setting, this campaign threshold

setting, stuff like that. Um so that's

basically a sort of like sheetbased

dashboard to manage all the scripts for

all your individual uh clients accounts.

That way um you keep track of what's

running where and the reports are also

generated in that Google sheet.

>> Nice. Great. And now then again thanks a

lot Neils for all the great insights. I

think everyone learned quite a bit.

Maybe too much to directly implement but

there were a lot of takeaways that I

think they can uh they can take home. I

hope to see you another time. So we'll

stay in touch. Uh and as you saw like

the audience was really really happy

about it and to be honest we didn't get

these uh types of comments uh so far. So

uh I would say you did an amazing job.

Thanks.

>> Thank you. Thank you.

>> Thank you for joining Next Level

E-commerce.

If you want to see more, check out our

YouTube or give us a follow on LinkedIn

to get more e-commerce tips and tricks.

See you at the next