Tackling Shopping Ads and E-commerce in 2021

Episode Description

If 2020 was the year #eCommerce became a necessity, 2021 could be when it becomes the norm.

What PPC strategies should you use for retail and eCommerce accounts? Which shopping trends are going to continue from last year? These questions are answered in our chat with Google and our friend Andrew, who comes armed with plenty of lessons from his agency.

This panel covers:

• Which 2020 online shopping trends Google expects to continue in 2021
• What happened during Standard vs. Smart Shopping campaign tests
• Pros and cons of different account structures for e-commerce advertisers
• Tips and tricks for maintaining more control while using Smart Shopping campaigns
• Feed optimizations to try for better PPC performance

Episode Takeaways

2020 Online Shopping Trends into 2021:

• The shift to digital shopping has intensified due to COVID-19, with significant increases in local search and curbside pickup.
• Consumers are trying new, safer services to receive their purchases, including a 600% increase in click-and-collect searches.
• Trends suggest that even traditionally in-store shoppers, including older demographics, are adapting to online shopping.

Standard vs. Smart Shopping Campaign Tests:

• Smart Shopping campaigns automatically adjust bids and allocate budget across platforms (Google Search, Display, YouTube, Gmail) based on performance.
• These campaigns prioritize products and ad placements dynamically to optimize conversion value, often outperforming Standard Shopping campaigns in tests.

Account Structures for E-Commerce Advertisers:

• Diverse account structures can impact campaign performance, with segmentation based on product profitability or seasonality.
• Structuring campaigns according to margin or inventory levels can help target specific business goals beyond just conversion value.

Control in Smart Shopping Campaigns:

• Advertisers seek strategies to maintain control over automated Smart Shopping campaigns, such as excluding certain products or adjusting targets based on specific insights.
• Enhanced visibility into campaign mechanics, like budget allocation across channels, can help advertisers optimize strategies and justify campaign adjustments to clients.

Feed Optimizations for Better PPC Performance:

• Optimizing product feeds by improving titles, descriptions, and images can enhance ad relevance and performance.
• Regular updates and strategic adjustments to product feeds are essential for keeping up with changing consumer behaviors and preferences.

Episode Transcript

Frederick Vallaeys: Hello and welcome to another episode of PPC Town Hall. My name is Fred Vallaeys. I’m one of the co founders and the CEO at Optmyzr . And today we’re going to have a very special session. We’ll make it the best one yet. And we have some guests back who you’ve seen before. But the topic we’ve got lined up for today is about e commerce and shopping ads and kind of thinking about all the things that happened in 2020, the new trends that we saw.

A lot of e commerce growth a lot of growth during the holiday season And sort of trying to figure out what are the takeaways from this? What can we expect to continue into 2021? What do we think is going to shift? And how do we build our campaigns for the next year around these new trends? So who better to join us for that session than two awesome speakers from from Google.

We got Peter and Emmy who will be joining us. But of course, we also want to hear from another practitioner. So after Google tells us what we can do better, let’s hear from someone who’s done it, who’s tried it and what his experience has been. So we’ve got Andrew Loke from Savvy Revenue is going to share that with us today.

So thanks for joining us on PPC Town Hall, and let’s get rolling.

All right, Emi and Peter, thank you for joining me. So thanks everyone who’s watching us. One thing we like to do, by the way, is if you can say hello in the comments, tell us where you’re joining us from. We’ll also use the comments on the YouTube stream. To take your questions. So this is live and me and Peter, where are you guys calling us from?

Emi Wayner: I am calling from Los Angeles south side of LA. Beautiful day. It’s a little chilly, but it’s a really nice today.

Frederick Vallaeys: Nice. Thank you for coming on again. Yeah. And so Emmy is a. Partner platform partner lead over at Google. And I mean, I work together quite a bit and then we’ve got Peter. Peter, you’re coming from the East coast, right?

Peter Oliveira: Yeah. Tuning in today from Raleigh, North Carolina, getting some quite gray weather here, but it’s okay. We’ll make it through it. We’ll be somewhere again soon, hopefully.

Frederick Vallaeys: Yeah. And I’m still here in the the San Francisco Bay area. And we’ve got A plume of water coming in a tropical, a pineapple express, I guess it’s called.

So for those of you not on the West Coast, that basically means that all the rain from Hawaii makes it over to us and we just get drenched with a foot of water, which we need because it’s been dry here. So yeah, lots of people joining from all over the world. We got Italy, Germany. Okay, we got some people getting their popcorn and notebooks ready.

We got Canada on the line. So thanks everyone for joining our session today So, as far as the flow so quickly level set here But emmy and peter we’re going to talk a bit about the trends that we’ve seen from 2020 moving into 2021 Then we’re going to talk to peter about smart shopping campaigns versus regular shopping campaigns Kind of how you would choose between those That’s when we’ll bring in andrew from savvy revenue Who’s actually played with these campaigns and we’ll do a quick introduction of him at that point But so, emmy with that lots of people watching.

So tell us what google has seen Shopping wise in 2020 and what you think is going to happen in 2021.

Emi Wayner: Sure. If you could put my slide quickly so shopping trends, I want to share something more consumer focused trend. So as we all know, we’ve shifted very quickly since COVID happened and digital is going to be, and is already a critical touch point for, you For many shoppers, right?

And search locally, and there’s three things. So search locally, and shopping shopping also in store availability. So I think when you think about myself too, and when you think about yourself, you look for items and looking for the inventory, if there is going to be available nearby stores. So that’s 60 percent increase.

And then trying new and safer services. As we all know, We want to make sure that the contact contacting the store and going to the store and picking up items is safe for everyone. So 65 percent increase in a curbside searches is I think for me, it was a shocking stats. So I just wanted to share these three main stats.

There are some other stats available for from Think with Google. So I’m going to share the link with everyone. It’s a public. Website. And then the lastly on the bottom if you 600 percent increase for searches for click and collect that’s, that tells you how much people, how many people are looking for services to pick up items from local places.

So that’s that’s a really interesting insights I wanted to share. These

Frederick Vallaeys: are the types of trends that we think might also stick, right? So, and I was just watching another study basically saying that the, the older generation in large droves is also moving to, to. Doing shopping online, but what’s interesting here is that shopping online no longer means like buy online and get it shipped to your house It’s sort of this hybrid model that’s now coming in that’s really making it easy for people to bridge that gap and you know have a different comfort level

Emi Wayner: Yeah And we’re going to see some category trends as well.

But I’ve seen an article saying that people are more comfortable purchasing big furniture. So furniture is another one in a home, you know, home improvements, of course. So I’m going to show you that slide later. So related to this, I wanted to show you the store pickup with shopping. I’m sure some of you have seen this.

So when you go search the item, you’ll see the picture and under the picture you see pickup today. There’s also another new option called pickup later. So that’s released in January. And there’s also another release website that you can look. So that’s insights too.

Frederick Vallaeys: That’s really interesting. So what do retailers do to get this running?

Peter, how do you get this pick up today, pick up tomorrow, pick up later running on your ads?

Peter Oliveira: Yes. These are all extensions that you can add onto your, onto your shopping campaigns. I’m not up to date with the, with the exact specifics on there. So I’m not sure if, if. If you’d have that handy, but we’ll look it up, right?

We

Emi Wayner: can we can include in a playbook, that comes after this webinar.

Frederick Vallaeys: Awesome. Yeah, we we do have the slides So if anyone’s asking we do have the slides that we can share with you later We’ll pop up the link at the bottom where you can put in your email and then we’ll send the slides everything we presented today As well as an additional playbook for shopping online All right.

Sorry to interrupt you. I mean no,

Emi Wayner: no, no. Yeah.

Frederick Vallaeys: People still, I knew there was

Emi Wayner: going to be a question for this one because it’s great. So why, as a marketer, when I use pickup today, it’s, it’s beneficial for all of us, for consumers and retailers and marketers. And then this is a one one stats I wanted to show you was a performance lift.

So the, when you have pick up today extension, then CVR increases 13 percent and add CTR increases 2%. So why not using to help everybody? And on top of that, you can deliver higher performance. So that’s that’s one message I wanted to share. Okay.

Frederick Vallaeys: Yeah. So obviously people they shop online, but they want it right now.

So if you have the capability of doing store pickup, why not do it? Right. And I think the other interesting thing that we certainly saw in 2020 was that local guidelines are shifting all the time. So if you, you know, if you’re not allowed to have people coming into your store and I don’t know what it’s like in the places where people live right now, I was just doing a webinar with some folks in in the UK, they’re pretty much on the strict, Lockdown, us in california.

We just came out of it. So but a lockdown in california. You can still go shopping strange sort of thing, Right, but but it really depends from area to area So the more that you can have these different options and be very nimble As far as putting them in play when it when it makes sense, obviously people still need stuff So they’re going to look to to the ads and this is awesome because like emmy showed the ads now actually show you You know That there is an option to just drive up to the storefront and have somebody put it in your trunk And take it home that way But then i’m also curious emmy so I mean you talked about like a lot of furniture purchases, right?

I think we also see a lot of people buying second monitors for their computer webcams to work from home Obviously, we’re seeing huge spikes in these categories, but can we expect that sort of e commerce to to continue? Or do you basically have your new couch and now you’re good for at least a decade?

Emi Wayner: Yeah, I think overall, all categories went up in terms of purchase online. But particularly what we’ve seen is, you know, computers all the electronics, right? And then clothes and furniture and home furnishing. And I think we all know everyone is working on a house renovating the house. So that’s a high demand.

And then, interestingly, I’ve seen health care and a personal care increased searches. And so those are the key categories that went up, I think, highest. And then when you look at, on the right hand side, I have a chart of search trends. So people are more comfortable purchasing non-brand items. They’re more focused on inventory.

Can I get it today? Can I get it tomorrow? So when you see brand versus generic you see much higher demand for non-brand searches.

Frederick Vallaeys: Right, so sort of the interesting story that I take away from this is obviously there’s been a huge boost in e commerce across All categories that are operating travel sort of being probably the exception here and so you get all these new consumers now looking for stuff online, but they’re brand agnostic They don’t care about a specific brand.

So basically all this new opportunity It is up for grabs to a large degree because people don’t look for that specific brand. So it just makes a huge case for get your shopping ads up because people are willing to be convinced by whatever it is Whether it’s price or easy access or like emmy was saying like is your supply chain working?

Can you actually get us this thing that we need and can you get it to us quickly? And brand matters far far less now. I I think brand overall Is important, right? Like I focus on brand, Google focuses on brand. You probably all do, but this is sort of like an opportunity to start engaging with all the customers.

And once you have that connection, then you can start building the brand with them, start doing email campaigns build that longterm relationship and eventually get cheaper clicks when that consumer does now go and search for your brand specifically in the future.

Emi Wayner: Yeah, and I think a lot of it is a social e commerce.

I think a trend as well. I think I share with you earlier. When you’re watching, let’s say YouTube or when you’re doing something else there are a lot more advertising or useful information coming up. I ended up purchasing so many. So that’s that’s one experience. myself, I can’t remember how many things I purchased from YouTube.

You

Frederick Vallaeys: know, I actually deleted the Instagram app off of my phone because I

Emi Wayner: am

Frederick Vallaeys: buying too many things from Instagram. I was like, I didn’t know I needed that. But yeah, sure.

Emi Wayner: Yeah. And a personal services too. I think we all know that, you know, consulting services to fitness services. And yoga teachers do any anyone who provides useful services?

That’s I think I’ve seen a trend that I think the other opportunity that was interesting that you had pointed out was that there’s been a huge increase in Streaming behavior. So someone like disney plus for example, they launched last year, but within that first year they saw The subscription levels they had expected to get after five years.

Frederick Vallaeys: So it was a huge acceleration, right? Yeah. Where else do people watch a lot of online content? Well, YouTube, obviously. And so YouTube actually has a lot of people watching and the inventory is pretty cheap. And and there’s some new opportunities to put ads on there as well. Right.

Emi Wayner: And then we’re going to talk more success cases later, but yeah, I’ve seen a lot of uplift from YouTube actions too.

Frederick Vallaeys: Good. Okay. Well, so we’re all here. We all believe in e commerce and shopping ads. So Peter, let’s talk a bit about smart campaigns, smart shopping campaigns versus regular shopping campaigns. And maybe if you can see slide 10, Amy, then we can talk about the differences between the two. So Peter, why don’t you explain a bit to folks how Google at least thinks about smart versus regular shopping?

Peter Oliveira: Yeah. So there are a couple of key differences, right? I think a lot of people understand that the smart shopping is going to use automated bidding but then a lot of the times the question that we’ll get is, okay, so what’s the difference between smart shopping and automated And a traditional shopping campaign with smart bidding enabled on it.

The key difference there is that smart shopping is going to dynamically allocate your budget using machine learning across different channels and formats. So, whereas with traditional shopping, you may have a Display campaign, a dynamic remarketing campaign. You may be wanting to surface products on YouTube and Gmail all through separate campaigns and through separate strategies.

Smart shopping is going to do that automatically for you. And the really nice thing is that it’s still keeping your goal in mind. So it’s only going to tap into those additional formats or surfaces when it thinks that, Tapping into those could bring you additional leads or sales,

Frederick Vallaeys: right? And I think that’s an important distinction that we’ll talk about that in a minute But like what is a lead and what is a sale and how do you value?

New customers and how do you value lifetime? I mean, I think fundamentally that that’s really a mind shift I think in in ppc Is we don’t have to worry so much about like what’s the cpc bid we want to set? But like how is it that we want to value when that customer comes to us? Like is that going to be a lifetime value?

Is it going to be high because that customer is going to come back to us? They’re not going to return a lot of things. Is that a customer who tends to visit us by walking into the store or only online, and what do we know about the difference in behavior between these two categories? So these are the questions that we have to start asking ourselves much more.

And how do we feed that data back into Google? And then I think regardless of whether you use smart shopping or regular shopping with smart bidding of some kind, the system is going to benefit from that because it actually knows what it is you’re trying to achieve. So we’ll talk a bit more about that and I’m kind of guessing that Andrew might have some some thoughts on that.

So Andrew, we’re going to add you to the stream here, Andrew Locke. Hello, how are you?

Andrew Lolk: Happy to be here. So far, so good.

Frederick Vallaeys: So far, so good. And I hear it’s pretty cold where you are. Where are you calling us from today?

Andrew Lolk: So we’re calling in from right now I’m sitting in, in Northern Mexico, right across the border from from San Diego.

And nothing down here is. Built for it’s built for, for 50, 50, 60 degree weather. So it’s just, it’s just, it feels cold. I have a space heater here behind me.

Frederick Vallaeys: It’s

Andrew Lolk: so bad. Like the people in Denmark we have a, we have company in Denmark as well. And they just like, they always like, whenever I say what the weather is, they’re always like, yeah, we’re at 30 right now. So it’s the same every time.

Frederick Vallaeys: You know, that’s why you moved, right? So you’re from Denmark.

That’s the reference there. Also the one of the co founders at White Shark Media. So you’ve obviously been in this space quite a long time. Currently tell us about Savvy and what you’re kind of working on now.

Andrew Lolk: So, yeah, we we, we do a primarily a paid search. We. Where it’s, it’s search shopping, YouTube display for for e-commerce.

So, so we’ve had a very, very interesting year with with Covid. We’ve been in, in one of the industries that’s really been booming. So, so that’s been interesting to follow how lockdowns have affected the different markets we operate in. Because we have a team in in Denmark and we have we have offices here in in the states.

Then we have, we operate with e commerce brands across the States Europe, and primarily Scandinavia. So we started seeing these really interesting things. What happened when there’s lockdowns in in individual countries and how it affects it it differently. And it really affects the same every single time.

So it’s been, it’s been really interesting to, to compare across across borders.

Frederick Vallaeys: So yeah, what have you seen across borders or between the different types of campaigns that you can run for shopping?

Andrew Lolk: So is one of the things that there was that, that, like the question we all ask is like, how long will the e commerce boom last?

And we all see different numbers and there are some overall charts that say, okay, we went up to 19 percent and then down to 16 percent and all that, but. Overall, what we’re seeing is that there’s just a boom across the border, and it’s, it’s more or less held in most industries And so that’s

Frederick Vallaeys: a behavior change, right?

So people who may not have been doing e commerce are now used to it. They’ve kind of explored it, so likely they’re going to continue it because they’ve seen the convenience of it.

Andrew Lolk: Yeah. So for instance, in Denmark, we had a lockdown in March of last year that lasted about one to two months. And we saw a huge boom in most e commerce markets at that time.

And we’re all sitting, just waiting for it to go away because COVID was more or less non existing in Denmark from may up until winter time. So we’re all waiting for it. When, when, when is it done? I got to go down again. And, and in most markets and most industries, it didn’t happen. didn’t, which was really interesting.

And we’ve been correlating some things from different countries. It, it just looks like it’s, it’s here to stay. People have now gotten used to buying online.

Frederick Vallaeys: Okay. So how do we make the most of that opportunity? And you guys have tried smart shopping and regular shopping. How do you think about that?

Andrew Lolk: So, so, so I’m known for being very skeptic. So, so, so one of the things that. That always surprises people a little bit is when I say, like, I recommend smart shopping and smart bidding for that matter, for all in house teams and full service agencies. When, when we consult with in house teams, especially smart bidding is the one we recommend.

But, but like smart shopping wins over regular. Manual bidding setups 10 times. And that’s coming from me. I run it. This is all we do So it’s like you’re

Frederick Vallaeys: actually pretty good at ppc, right? So yeah, this

Andrew Lolk: is all we do So it’s like but it’s just for it’s just in house teams and full service agencies They don’t get as deep as the rest of us do.

Not tooting our own horn, but we This is all we do. So we have some more complex setups we can use and I’ve been running for a very long time that worked really well. It’s just the standard setup that you see in Google ad accounts these days with one shopping campaign, maybe a little bit of smart bidding, or maybe a little bit of manual bidding that you update now and again.

It’s just smart shopping beats that. Nine out of 10 times. And

Frederick Vallaeys: I agree with that. I mean, I think for a lot of companies, they, they’re kind of used to keyword advertising and then they figure out, well, Hey, Shopping ads obviously are huge. If you’re a retailer they dominate those search results pages So we need to do it, but you’ve never really focused on it You’ve never really learned it and then you set up that single campaign with a single target and that’s not going to work well, right and so without Exactly without divulging your secret sauce, but what like what’s the key factor?

In terms of manual management and structure and how you think about bidding, that does make it work better if you’re actually super focused on it.

Andrew Lolk: So the, the one thing that, that you need to use in order to get men quote unquote manual working is is the priority levels. So there is this ability within Google shopping where you can have a high, medium and low priority, and then you.

You’re basically telling Google, which which products or which campaigns are prioritized over the other. And you can play around with that in many different ways and forms. The most, the easiest example to understand is having best sellers in your high priority and having. All your regular products in the secondary campaign.

That means you’re telling google if you have two different cables and you have one best seller and another that’s usually regular Well show my best seller first because that’s where we get the most conversions from is the one we get the highest conversion rates from That like, but you can use that for promotions, inventory levels with weather with if you’re proactively using seasonal bidding based on, okay, now we’re getting into clearance time in February.

So you need to focus on your clearance products. You can use that across the board to proactively showcase focus on the products that, that sell the best.

Frederick Vallaeys: Right. And so I guess what you’re saying is that you have a huge inventory. So you just may have different margin on certain products, have certain products ready in the warehouse to ship.

Other things might be drop shipped. So maybe you don’t want to focus that much on those. So so basically how good is the limit? Yeah.

Andrew Lolk: Sky’s really the limit on, on, on, on that part.

Frederick Vallaeys: Now, of course, we have the high, medium, low priority campaigns. So do you typically use just two levels or will you use all three levels?

Yeah.

Andrew Lolk: It depends on the complexity. Sometimes it can be a lot to, to, to try to bucket search terms or products into three different layers. And especially with in house team, that’s just getting used to this, this kind of setup, you can just use two layers. So you have a high priority and a medium priority.

Perfect. That’s, that’s, that’s a good start.

Frederick Vallaeys: Yeah, that makes sense. And here Antonia saying another piece of good advice, I think. So smart shopping, use it on good sellers and push manually. Actually, sorry, the point is confusing me a little bit, but I think what you’re basically saying, so these, these things that you can actually focus on, then focus on it, regular shopping campaign And split the levels, high, medium, low, but then the smaller clients that you may be managing where you don’t necessarily have the time to think about these things.

Smart shopping beats kind of like setting up a campaign and then not worrying about it.

Andrew Lolk: Pretty much. Yeah. And, and this is like one of the the example that Antonio is mentioning is, is really good for when it works. Let’s, let’s say that your best sellers has a lot of data. And we all know smart bidding, smart sharing works best with a lot of data.

So, so putting those bestsellers with a lot of data and your smart sharing campaign, and it can like kind of squeeze the performance out of them. And then you kind of just manually gut feel, run the rest of your, your products in a manual campaign. You can easily do that.

Frederick Vallaeys: Makes sense. Hey I want to add Peter back to the stream here because we’re talking about some some pretty good tactics and I know Peter loves his tactics.

Peter, any comments on what we’ve heard so far from Andrew?

Peter Oliveira: No, I think it’s totally in line with what we’ve seen and what we’ve done before. I mean, obviously, Google talks a lot about smart shopping, and we push it pretty heavily, but that’s because we do see good results. And even for myself, I started at one point just reading the case studies and reading the survey results.

But at one point in time, if I believe it, I’ve worked with a lot of different partners in the past that Didn’t necessarily just buy into the stats. Right. So through a lot of like recurring tests with different partners, different budget levels, different user groups my experience at least has been time and time again, that we do see smart shopping outperform traditional shopping in all of the use cases that I’ve encountered so far, at least.

Frederick Vallaeys: And you touched on some things Andrea was asking. So are there like certain budget levels for which it works better than others?

Peter Oliveira: Yeah. So as far as I’ve experienced, we’ve tested smart shopping at least within our organization at some levels that are, are quite lower even than probably the recommended Google one overall.

And what I’d say is I’d probably try to keep it, I probably want to make sure that the campaign is generating sales, right. To a certain extent, if it’s not generating any sales at all I might hold off on it, on, on testing it for.

Frederick Vallaeys: Sorry to interrupt, but that’s basically because Google makes a point, there should be a roughly a hundred conversions per month.

On a smart shopping campaign, right? So it’s maybe not so much about the budget, but how many Quote, unquote sales. You try.

Peter Oliveira: Yeah, I’d say the sales and conversions matter a bit more. We have seen success at least with testing that we’ve done within our team that I’ve done with, with different partners that I work with, we have seen some pretty strong success with much less than a hundred conversions a month.

But once you start getting into single digits, I might. Want to keep an eye on that and test it a bit carefully.

Frederick Vallaeys: Thanks for bringing in Peter. I’m going to ask Andrew one more question and then we’ll bring you and Emmy back in here. But Andrew, we had a great question from Cole. The smart shopping campaign outperform a more advanced query level filtering set up a priority settings. Nine out of 10, you said it does work for us. Smart shopping

level filtering. So let’s talk about it. And a lot of people may not know exactly what the query level filtering is.

Andrew Lolk: So the query, so, so the query level filtering is, is using the priority levels that I mentioned before. But, but through some tactics, which are hard to explain on a, on a quick call, but basically you’re trying to categorize the different keywords into high, medium, and low.

Priority campaigns based on how likely they are to convert. So for instance, a branded term converts better than a generic term in most markets. So you’ll put that you’ll, you’ll assign those in one campaign and set a higher bid for it. That’s basically what, what Cole is mentioning. And, and the question will it outperform a more advanced query level setup that’s actually.

The best question I’ve gotten today. We had another team meeting last week as well and savvy and our verdict on that is, is, is evolving a little bit right now because we’re getting less search data. So the less search data or search term data we have access to, the less powerful it is to use the query filter because we can’t see what is actually performing well and what is actually performing poorly.

So we can’t find the keywords or search terms. That needs to be moved into the different categories or buckets, so to speak. So we have seen cases where we would have used the curve filter in the past, but where we either go over to another priority split, or we, in some cases with various simple products actually opt for, for the smart shopping, which seems to doing really well.

Frederick Vallaeys: Yeah. And so my take on the query level filtering is that there’s sort of two levels, right? The one you talked about was brand versus generic performs differently. And in that case, you don’t really need to know what the search queries are. You can still say, okay, well, here’s my generic campaign. So I know what the brands are.

So I’m just going to make those negative keywords right from the get go. And so you can still maintain that you can still experiment and listen, does it work better or worse? I think you always have to test it. And it depends on a lot of other factors too. Like, how are you bidding for the different categories?

But

Andrew Lolk: one thing, one thing that is interesting with, with the query filter thing is that it’s both smart, smart, big and smart shopping. Have access to the full library of of search terms. So. Basically, if you want to run a query filter or a query split on your shopping campaigns, theoretically, it should be better to run it with smart bidding because smart bidding has access to all of the search term data that we don’t.

So that’s like, that’s where our thinking on that evolves as, as stabilities and the in, in in Google ads also changes.

Frederick Vallaeys: Peter, can you deny or confirm? That might be the case. I don’t think you can, right?

Peter Oliveira: I’m learning a lot from this, too. We work with a lot of a lot of SMB. So I haven’t gotten to the advanced query filtering yet.

Frederick Vallaeys: And then the other sort of query level filtering that is becoming more challenging for us. And we have capabilities to do all of these things and optimize versus check it out if you haven’t seen it. But basically, we can identify when a generic query. Something like t shirts may match against different product groups, and we can see, okay, well, this product group sells an orange t shirt and this one, a black t shirt and this one, a white t shirt.

And for some reason, Google always wants to show the orange t shirt. And I don’t know why, because it doesn’t convert as well. Like, how many times do you see someone wearing an orange t shirt, right? But now at least we know, because we look at the queries and we see why it doesn’t convert as well. So we make it a negative.

So it’s query sculpting. So we forced the query to go to the ad group where we think it’s just a better match. And that could be because of conversion rate could be because of profitability could just be because we just have more black t shirts in stock than orange ones, right? So we prefer selling that one.

Anyway, all good advice and tons of good questions coming in. So I’m going to bring me into the call as well again but let’s look at some of these questions and how people want to optimize So garrett is asking what about smart shopping? Versus actually having the three channels split out by campaign, right?

Because like peter was saying a smart shopping campaign will show you right on youtube display and search at the same time Andrew makes the point that if you manage it really well, you can still do better. But does that also mean you have to have all three campaign types?

Peter Oliveira: Yeah. So I think the core benefit that we’re, we’re getting from smart shopping versus having the separate ones is that we’re dynamically going to allocate the budget between them based on the results you’re saying, right?

So if you’re having a standard shopping campaign and a dynamic display campaign, you might, you’re probably gonna have to determine how much budget is going into each of those. So if in any case, let’s say you allocated less budget to the dynamic display campaign, but the dynamics display campaign starts giving you more favorable results, starts giving you more conversion value.

At the cost that you’re giving. If you have those in separate in separate campaigns, we won’t be able to then shift more money into that shift, more budget into that so that we can capture those additional sales. Smart shopping will take care of that. Okay, how much budget do I allocate to each of these and simplify it to just hey I’m trying to get as much conversion value as possible within my goal.

Serve my ads on whatever channel are gonna best help me do that.

Frederick Vallaeys: Let’s move on to the next question here So marco is asking about scaling smart shopping campaigns and there seems to be this invisible wall That they keep hitting when they think there’s more people searching, but it’s just that that volume is not coming there.

Peter, I know you had a great visual, which is a bit of a flow chart. I mean, I don’t know if you can show us the last slide, slide 24, which we got right there. So Peter, do you want to walk us through kind of like how you scale here?

Peter Oliveira: Yeah. So I think the core levers that we have for, for scaling the smart shopping campaign in general, right.

It’s going to be the robust, the budget, and also how many products are in our feed. And it’s simple terms like products in your feed are going to help because the more products we have to choose from the, the more inventory we have to serve up from, right, that would be akin to in a search campaign to adding more keywords.

If we add more products in there, we’ll be able to get more reach. The ROAS target and budget, those are sort of served as trade offs. So if I am lowering my ROAS target, I’m essentially telling Google ads. And I’m telling the machine learning that I’m okay with slightly less efficient conversions if I’m able to capture more volume from it.

So we can use that as a lever in order to achieve, achieve more reach in that case. And I sort of think that this graphic a bit works through that, right? And I think if your budget is, is limited that might be a case where we want to increase the ROAS because that could be an indication that, hey, we could be getting more, more revenue at an actual better ROAS if we were to increase the budget.

Frederick Vallaeys: And that slide, by the way, we’ll put it in the deck that we’ll share. So anyone who wants it will pop the URL to request it on the bottom of the screen here. There’s a question about from Harry here. So SmartShop works great, but then all of a sudden it doesn’t. What do you do? How do you fix it when so much of it is kind of like automated?

Peter Oliveira: They start performing, stop performing. Well, I think it’s, it depends on a bit sort of like the flow chart thing. Right. First check, like, did we set the ROAS target correctly? Do we have the right products in there? Is the budget appropriate for, for what we’re serving on? If all of those seem right, then it may be the case that we dig down into.

The feed right? Are our feeds set up really well? Is there some optimization we need to do? And the, the titles, the, the descriptions, there’s a lot of complexity that can happen there and a lot of different levers we can pull. But not sure if, if Andrew, you or Fred have encountered question

Frederick Vallaeys: is, is it is the performance dropping in terms of the conversions are drying up or is it that the impressions and clicks are drying up?

And, and so then could be that the price is the issue, right? So there’s a competitor who came in, who’s underpricing you that would cause your conversions to go away. Well, the search volume might still be there. I

Andrew Lolk: think one of the things like, so we, we see this we, we get a lot of calls on, on, on this particular scenario.

And, and the, the most often thing is like like, like Peter slide, it might have looked a little bit complex for everybody who’s, who’s looked at it, but it’s, it’s really, that’s all it takes. Like one month, for instance, you go from November to December, you’re seeing great results, and then you move into January and, and, you know, And ROAS started sliding a little bit down and, but you still have the same ROAS target as you had in December, which is a high one.

The campaign can’t really generate the same volume as it did before. So it’s still trying to chase that high ROAS that you had before, where the solution is really just adjusting your ROAS levels or even changing to maximize conversion value within a budget that you have available so that, so that you use the, so you use the, the options you have available now.

Yeah. Too many people just look at this smart shopping, smart bidding, et cetera, and look at us as a set and forget. And, and there’s so many things you can go in and tweak when it comes to the bidding and the budget and all that force it to, to relearn some of these things and think of like when it’s seasonality changing and all these things and then just changing the target rows is a really good solution.

So that, that before you start diving into the feed and all the other things.

Frederick Vallaeys: Yeah. I I mean, to me, it’s always curious that when you Talk about people who’ve been doing PPC for a long time. They never thought of that CPC being a static value. Like we all went in like every day, changing every CPC that we had.

And then all of a sudden we got smart bidding and it’s like, Oh yeah, here’s our target and we’re done. Like, can we walk away? That’s fine at some level, because part of what the system is doing is it’s figuring out what is the correct CPC. And that’s part of the variance that we did based on changes in conversion rate.

But what it’s not doing is looking at what you were saying. Andrew seasonality, like how have your margins changed? How has your supply chain changed? Right? Those are the things that still warrant you to to take a look at at the targets, the target row, as in the target CPA. And again, I mean, I love having tools like Optmyzr events, so you can slightly tweak it and you can see if that tweak is working automatically and then reapply and keep going.

Chasing basically after the right targets that work for you.

What question do you guys want to take next? There’s lots of questions. Well, why don’t we take Paul’s question here? So, and this is another optimization tip before we go into Paul’s question, but one way to get smart shopping and smart bidding to work better is to attach audiences to it so that Google has a little bit more information about like, who is it?

Did they come to my product page? Did they go through my shopping cart? They can use that, right? And so then the question is, is it better to do this tagging through Google ads? We’re going to use analytics or does it not matter?

Peter Oliveira: I think both of them work, work pretty well. I think if you have an existing analytics setup, you may want to use a lot of the different a lot of the different goals you have there and import them into Google ads as well.

But both are very compatible. And I think you can go either way.

Andrew Lolk: Yeah, same, same, same thing. It’s it’s, it’s, it’s about the same.

Frederick Vallaeys: Yeah. And Garrett who’s watching is also confirming that it’s both of them work. So we have this question here, so we kind of touched on this before, but Herbert’s asking minimum number of products for a smart shopping campaign.

I guess this talks about granularity, right? So. And I think we have a great slide. So, I mean it was slide number 14, if we can show that one. So slide number 14 is talking about splitting things out, keep going, and I want to write there. So how do you put things out? When does it get too granular?

And then, and Peter, do you want to talk about what this slide is kind of talking about as far as a smart shopping campaign split out?

Peter Oliveira: Yeah. So I think this is also referring to like testing different shopping campaigns, how far, how far you can split it. Listen, I’d say, wouldn’t say there’s like one specific magic number.

I think in a lot of slides, we’ve called out like a hundred conversions per product group. But ultimately you want to make sure that you have enough volume to meet whatever goal you’re, you’re, you’re tackling, right. The smart shopping ultimately that in smart bidding in general now optimize at the entire account level.

So it’s learning from all the different things that you have going on in there. But the risk of segmenting it too much, much is that with each segment that you put that you break out, you’re introducing an additional constraint, right? So if we have one single, if we’re just looking at conversion value overall, and the amount of conversion value we’re able to generate in our account, it is likely that we’ll be able to generate the most conversion value with one campaign with a single target versus three campaigns that are each introducing different targets.

Because with each campaign, we’re introducing an additional constraint. So if we have only one product in a campaign and we have a ROAS target of 300%, that may never show. Right. So I think we just have to balance that out. And testing is, is always a good way to do it because there’s no, Secret saucer or magic recipe there,

Frederick Vallaeys: right?

But I think what you also touched on then is that you really have to understand your goal, right? So if you’re looking to maximize revenue then put everything in a single campaign and just like Google figure it out obviously that’s now also premised on correctly reporting your values back into google but then on the flip side if you have more of a profitability goal Well, guess what?

You probably have different margins on the different products that you sell so it doesn’t make sense in that case To have a single t row ass against that Absolutely

Peter Oliveira: Yeah, and that goes into also, like, why would you want to segment your campaigns to begin with? Right. I think a single smart shopping campaign for your account works really, really well.

If you’re North Star, and if what you’re trying to achieve is get as much conversion value as possible within a row at school. However, in a lot of cases, we understand that that isn’t the only business school that that a retailer may have, right. Retailer may have different profitability goals. Certain products have a higher profit margin than others.

Certain products, some products may be in low stock. Some products may be in high stock, some products may be seasonal, and we really got to get rid of them this month. Right? So all of those things are different signals that aren’t necessarily just the conversion value. And by segmenting our campaign, that’s really how we can inform smart shopping and smart bidding of how they should be treating these different products in these different groups, right?

So we may want to have our higher margin products in a separate campaign that has a lower target robots, right? Because by having a lower target robots, we’re saying bid more aggressively on these. We’re okay if it’s a little bit less efficient, but if a product has a much lower margin, maybe we only want to get sales and we only want to convert those if it’s at a really, really efficient row ass.

So segmenting your campaigns are a good way of doing that, but also we don’t want to go too overboard to a point where we can’t meet the new row ass targets and these additional constraints that we’re introducing into the machine learning.

Frederick Vallaeys: Right. And to get really tactical here for a second. So Google doesn’t have a margin field, I believe, as of yet. So how do we build out campaigns based on different levels of profitability?

Peter Oliveira: There’s like, there’s a bunch of different Andrew, go for it. Can

Andrew Lolk: I start off by saying I don’t like doing it because it’s, it’s so, it’s, it’s so hard to know what people are actually buying after clicking on a specific product.

We we’ve seen them, we’ve run the number so many times and there’s very little correlation between what people end up buying and what they’re actually clicking in the shopping ads. So it’s, it can be very misleading to run at one campaign at 800 percent ROAS because you have a low margin and another campaign at a 200 percent ROAS because of a huge margin. So we, we, we love to use profit metrics for it. Huge plug. It’s a, it’s another Danish company. We use it for, for a lot of our clients and it’s, it shows you the profit derived from each transaction. So you actually, you’re skipping a couple of steps. You’re getting straight to the part where you get the profit from each sale.

It’s, you know, what clicking a product, it actually ends up Yeah. Resulting in when it comes to profit. So you don’t guess what the profit is based on what the margin is on the product.

Frederick Vallaeys: Yeah. We had Frederick poison, I believe from profit metrics on a couple of weeks ago. So if you want to see that episode but yeah, I mean, I think regardless of how you go about it, like there’s different tools that help you plug in profitability and you can bring it into Optmyzr .

I can do split outs based on historical ROAS. But one of the key points we’re hearing here is if you want to get the most success out of something, then oftentimes. If you do have the time to manage it, split it up in some way that makes sense. So Paul here is kind of lamenting. I think that he’s trying to convince a client of his to do what he thinks is right, but they want to do it the other way.

When it comes to experiments, actually, and Peter, maybe you can weigh in, like one of the difficulties I’ve heard is that smart shopping basically sucks up all the traffic that’s possible for it. And so completely ignores a competing regular shopping campaign that you have. So if you’re in a scenario where you actually want to prove to your client that one works better than the other, what sort of a split up might function correctly and give you the data you need?

Peter Oliveira: Yeah, so there is some truth to that. So with if a product group exists in two separate campaigns and one smart shopping campaign and once regular shopping campaign, the smart shopping campaign will have priority, right? It would be similar to like the priority setting that we have a traditional shopping campaigns.

It will take priority over the product group. That’s in the regular shopping campaign. So we really recommend testing. If you are trying to compare by product group. So I actually did this a few weeks, but basically I downloaded a report of all the different product groups that existed in an account.

And then assign certain ones that I’d want to put in smart shopping and certain ones that I’d want to keep in traditional shopping. I, the only sort of caveat that I had while choosing these is that I wanted some product groups that did have. A considerable amount of conversion value to give it a chance to succeed.

But you may also want to have a representative sample, right? Maybe pull in some product groups that have a higher conversion value and some that have a lower as long as you have some there that are that are able to succeed. And then once we have those product groups selected, We can create a new Smart Shopping campaign with only the product groups that we want to test.

And actually, because Smart Shopping is prioritized, we don’t even necessarily have to remove those product groups from the traditional shopping campaign. Smart Shopping campaign will just take priority and you’ll leave the certain product groups just in the base campaign that aren’t included in the Smart Shopping campaign.

So you have something To test against, right?

Frederick Vallaeys: I mean, what’s the role of the marketer in all of this? We’re talking a lot of tactics here. So what’s the, what’s that shift that we have when so much of it can be automated?

Emi Wayner: Yeah. I wanted to ask the audience if that’s okay. I think they’re all saying smart shopping campaign as well as manual shopping.

But I would love to hear, I know there are so many steps to get there, but what are the features or tools that are missing on the market? That you will like to see on optimize or any other platforms today, if you have any lab to see, cause then we can help.

Frederick Vallaeys: So let me ask, right. So, and they’re here to listen.

So yes, please put it in the chat comments on YouTube. What features you want to see? Okay. I’ve got the first one here and Andrew, I’m sure you’re going to ask for a feature too, but what is the split out for smart shopping between the different channels, how much of my budget is going to remarketing?

It looks like Google actually has a beta for this. So some people have noticed under the attribution reports, there’s a couple of new dimensions that have opened up that actually do show the split out for smart shopping campaigns. So my request would be open up this beta to more people and make it an official product.

So that’s a, that’s what I’m asking for. Andrew seems to agree. Anything else, Andrew?

Andrew Lolk: Like the biggest thing I have. So, so, so we, we, we see us, we see ourselves moving, moving into that strategy role as, as we have taken on a lot of time. I think most PPC people like myself have been in the game for 10 plus years.

If, if we need to add more negative keywords, we kind of like, we, we don’t want to do that anymore. Like, so nobody wants more control. But, but visibility, like that thing, the visibility, like the latest things that come out with smart bidding for, for target CPA, at least, or start showcasing, we’re bidding higher and lower based on these parameters.

The more of that, the better, because that like, there’s so much data that that’s being used today. And the better we can use that data to also inform clients, inform ourselves. And hopefully down the line, say like, listen, this parameter with weekends that doesn’t seem to be converting. We want to ignore that because we want to just hammer away on the weekends because we don’t want to leave all of the market for the, the competitors on the weekends, even though we know ROAS is lower on the weekend.

That could be one example where we’re at that control that a lot of people are asking for, which sometimes gets misunderstood as, as the need to control, or it’s just like, we have some insights and we have some client requests and we have our own requests as well, that, that we like that limitation sometimes hurts.

But I know a lot of work is being done and listening to that already. And the visibility seems to be getting there, but I just, I want to mimic all the others in the industry that says that, yeah, that makes a ton of sense. I mean, be the strategist, right? So don’t, don’t pine for going back to the days of having to manage every single little thing manually.

Frederick Vallaeys: The figure out what’s the bigger picture stuff, the data about your business that you can, how do you feed that into the machine so the machine can make better decisions? And then, like you said, if it’s a strategy based decision, like we’re okay making less money on the weekends, we just don’t want to lose these potential customers down the line.

That’s a big one. And I think, and Andrew, maybe we’ll have you back in a couple of weeks here, but that also speaks to attribution, I think, right? So if you look at the last four attribution model, Then yes, you we will shut off your ads on any day when people are just beginning the process of shopping and that’s especially dangerous when you have smart shopping or smart bidding And last click attribution in combination with each other because it just it’s so hyper focused on a narrow slice of what’s happening It ignores everything around that and that basically disappears but that was actually a driver of what eventually brought people to to purchase.

All right, we’re Oh dan is asking us. You Showcase shopping ads. Do some smart shopping campaigns supersede those?

Peter Oliveira: Yeah. So I don’t know if I have stats specifically. I would imagine showcase performs quite well on, on both. I think the core difference here is that on smart shopping showcase ads are, are automatically served, right?

You don’t need to create a specific. Showcase ad group. If if we determined that it would be a good user experience and that it would maximize your goals, your results to serve a showcase shopping ads will automatically group products together and be able to serve that right. So I think one of the benefits that you’ll get there is that right.

It’s going to be. We won’t be setting a specific Row S goal in advance that we want to hit. We’ll just let the machine learning determine when or when not to serve showcase ads based on the ability that it has to maximize our overall campaign goal.

Frederick Vallaeys: Okay. Correct me here if I’m wrong, right? But isn’t that a query level decision that Google says, this is a query that seems not specific enough that maybe we should show a showcase ad, which shows a variety of products as opposed to that one specific thing the user was looking for.

And so then if Google says it’s showcase, well, then. If you have a showcase ad that basically is the one that pulls in and your regular shopping, I wouldn’t have not even qualified.

Peter Oliveira: Yeah. So if you think about like what showcase ads are, right, they group products together and allow the user to sort of like filter through oftentimes by brand.

So they can go through the catalog, see the different products that you have to offer. If I know exactly what I want and I have a very, very high intent query, it might not make as much sense to service a whole catalog of different products. Right. We surface. The product that the user’s looking for and make it super, super easy for them to convert on that.

But if I have a query that’s more potentially mid-funnel, right? I’m still in the discovery phase, I’m still going through that purchase journey, then presenting a catalog with a lot of different product options might be a better user experience. It might drive some more engagement that way. So that is sort of how, how we’re determining when to serve those showcase shopping ads when the user’s sort of in that stage of the purchase journey where they’re still open to learning more about different products.

Frederick Vallaeys: Okay. And we’re getting pretty close to time here, but. This is the last life question. I think we’re going to take before we wrap it up. Oh, no, sorry, showed the wrong question. No, oh, my God, sorry. The questions keep moving around on my screen. So I click where it used to be, and then it’s a completely different question.

My apologies. So the people whose questions I showed, we’re not going to answer them, but I do want to answer this one from Alexander. So what about learning? Right? So smart shopping, smart bidding, it needs some budget to figure out what’s what. How do you avoid that being wasted spend? How do you get the machine as close to where it needs to be as possible?

And Andrew, I’m curious if you have any thoughts on that one. So the, the, the name of the game is trying to help. Smart shopping or smart bidding predict what will convert the best. So if you’re worried about worried about, worried about burning through a budget, start with the 10 percent or 5 percent or 20 percent of the products that That are bestsellers like 20 percent of the products that account for 80 percent of the pro the revenue Start there instead of throwing 10 000 products into a smart shopping or any shopping campaign It’s the good old rule spend a dollar per click per product per day and you’ve spent ten thousand dollars in a day And you’ve learned nothing so Limit the amount of products and then slowly start adding more products to it start with the best sellers and you’ll burn through less budget so to speak, but you will burn through some You That’s just it.

Andrew Lolk: That’s also part of the smart shopping and smart bidding game. That part that you will, you will learn for yourself when you do manual bidding. That’s basically what Google is doing. So you have to take some waste initially for, for the, for the long term gain.

Frederick Vallaeys: Right. But also think about some exclusions potentially, right?

So if you’ve already figured out that you don’t really sell this product in New York City, then Maybe exclude that from the beginning. And then later on when things are working well and maybe you have some budget to see, can we actually make New York city work? Then you bring that location back into it.

Andrew Lolk: And if you’ve learned something from your previous manual campaigns, like for instance, we, we have clients that sell posters or, or like those wallpaper things and certain products, like it’s funny, like the, the world map wallpapers. They seem to. All show up in manual in shopping on the word word any sort of map related query.

And it’s such a waste of everybody’s budget. So that product might not do well in a, in an environment like smart sh where we can exclude something. Some of those learnings just take ’em with you. And some of the products will go in smart shopping, some won’t. So, so every learning that you have that you can somehow mimic in smart shopping, give it an, give it an an advantage instead of trying to start from scratch every time.

Frederick Vallaeys: Great questions keep coming in. We just don’t have time. We know people have to go on to other meetings. But what we will do is we will look at these comments as great suggestions for some blog posts and we’ll see if Google will help us answer those. And maybe we’ll get Andrew to weigh in. By the way, Andrew has great blogs on his website, savvy revenue.

com. So after this, go and check those out as well. Okay. Closing thoughts. Who wants to start anything? I know we haven’t really touched on the YouTube for action as a great opportunity. To expand your shopping volume. I mean, do you want to tell us maybe a little bit more about that?

Emi Wayner: Yeah. Maybe we can set up another another session focusing on the YouTube action and maybe we can share insights and then get feedback from everybody.

Frederick Vallaeys: Yeah. We do lots of PPC townhouse. So always open for more topics. By the way, the next topic we’ll have in two weeks is Running a successful PPC agency in the age of automation. It’s a very pertinent question. Like once you, basically the people you’ve hired to do the account management for you, once they’re no longer working on spreadsheets all day long, what should they be working on and how do you remain relevant when Google automates so much, and then the, the two weeks after that, towards the end of February, we’re going to do one on attribution and I haven’t asked Andrew yet, but if he wants to come back, he’s certainly welcome to talk about that as well.

Andrew Lolk: I can speak all day long,

Frederick Vallaeys: but also in

Andrew Lolk: attribution, I’ll ask a ton of questions of of the other guys, because it’s such a, it’s, it’s such a wild game that, that whole attribution thing.

Emi Wayner: Yeah, that’s exciting.

Frederick Vallaeys: Peter, do you you know, as the, the guy who probably sees most of what’s happening with all of Google’s automation, like what’s your one takeaway?

Like obviously we’ve seen that it can work really well, but what sets you up for having that best chance of success?

Peter Oliveira: Yeah, I think two things to highlight. One is just test, test, test, keep on testing always. I think as. As we start introducing like more and more like different types of automations, I think it seems to be a pretty great opportunity for us as PPC practitioners to test different items and make sure that they work for the different user bases that we’re servicing.

And in terms of automation, like how we can help that and help inform it, help reduce the learning, right? Make sure that we’re importing things, importing different conversion values, communicating segmenting your campaign, setting different ROAS targets, making sure that it knows what it. Doesn’t know through the just the traditional baseline conversion value.

Frederick Vallaeys: Great thoughts. Okay, well, thank you everyone for watching and for the great interaction we saw again in the questions and people even answering each other’s questions. I love seeing that. So the slides, like I said, we do have them available. Actually, we don’t, we don’t quite have them available yet, but if you go to this landing page right there you can sign up and we’ll have them in about a week or so.

But we’ll, you’ll have the slides from this session. You’ll have some additional slides that we didn’t get to talk about today that talk about best practices. But Emmy, Peter, Andrew, thank you for your great insights today. Thanks for joining us. And like I said, we’ll be back with the next PPC town hall on Wednesday, February 10th, and then again, two weeks after that.

So thanks for joining. We’ll see you for the next one.

Andrew Lolk: Thank you. Thanks guys. Great