Podcast transcript
Introduction
Hi, and welcome to Marketplace Masters, the podcast dedicated to uncovering the strategies that drive agency success on Amazon.
Paul Sonneveld
I'm your host, Paul Sonneveld, and today, we're going to be diving into an advanced and often very complex topic, Amazon DSP and Amazon Marketing Cloud, and specifically the tactics that you can take as an Amazon agency or an enterprise seller to really take performance to that next level.
Now, to help me break all of this down, I have a great guest today. I'm thrilled to welcome Jens Jokschat, a very seasoned expert in e-commerce advertising and a pioneer in Amazon marketing. Now, for those of you who don't know Jens, Jens is the Founder and Managing Director of PrimeUp, a full-service e-commerce agency that spans vendors and sellers, really help them scale their revenue on Amazon and other marketplaces.
With a career spanning back to 1999, Jens has deep expertise in performance marketing and Ad technologies. He has managed campaigns across Google, Facebook, Amazon, optimizing results for more than 100 clients during his time at agencies like Pilot and D3 Media. In 2017, he launched PrimeUp, becoming one of the first specialized service providers focused entirely on Amazon Advertising. Jens understands firsthand the power of optimization technologies and best practice strategies, particularly within the Amazon marketing services and systems and sponsored products. Jens, it's an absolute pleasure to have you on the show today.
Jens Jokschat
Thanks for the warm welcome, Paul, and hello, everyone.
Paul Sonneveld
All right, I need to lift my face a little bit too close to the screen, I think, but great. Just a reminder, before we dive in, everyone, We've got a jam-packed session for you. So we're going to try and do our best to stick to the time. We may run over a little bit. Also a reminder that this is a genuine live session. It's interactive. If you do want to ask Jens some very compelling questions, I encourage you to use the comment section on LinkedIn to do that. They flow into our system and we will try and answer as many of your questions as we can.
All right, let's dive into it. Maybe just a scene-setting question first, Jens. We always hear these phrases like endemic, non-endemic, and Amazon DSP in particular has become a very powerful tool for both. But maybe just to set the scene, can you just break down the differences between these two types of advertisers and between their different approaches at a high level to set the scene?
Jens Jokschat
Yeah, absolutely. Good point and a good question. So basically what Amazon says is how they define non-endemic advertisers. That's pretty much all industries that do not basically sell products or potentially sell products on Amazon. And as an endemic client, that's pretty much e-commerce in the broadest sense. So if you have either your own online shop or selling on Amazon or both, then you're considered an endemic client.
And in terms of the marketing funnel that Amazon offers basically to advertise your products, your services, what Amazon's been doing just recently really is moving towards the upper funnel and pretty much broadened their offerings specifically when it comes to activating consumers that have not been searching for your product. I mean, we all know Amazon PPC when people punch in their keywords to buy your products. That's like the lowest effort, the lowest funnel pillar, so to speak. And recently, Amazon has really broadened this up to the TV market by launching Prime Video TV Ads, which we will be talking about today, same as Fire TV Ads and obviously, through the DSP, you can not only run display Ads that we are all familiar with, but also online video Ads on thousands of websites.
And what we're also going to dive in a little bit is the different metrics and KPIs that both non-endemic clients and endemic clients should look to. And these KPIs might be a bit different depending on your objectives for the campaigns that you have. It's not necessarily all about direct sales and driving revenue, specifically for complex products where the user journey, the decision-making process might be a bit longer. You also want to activate and get into consumers' minds a bit earlier when they start their thought process. And yeah, that's something we're going to look at today.
And when we pretty much think about how endemics have been using the Amazon platforms, what we've observed is actually that the DSP pretty much is not really the icing on the cake, but it's the last or the latest stage of adding it to the marketing mix as an endemic client. I mean, what we all probably know and have seen is first thing you do is you start launching your products on Amazon, you make them available, you maybe run some Ads, but in the early stages, that was sometimes really reactive and less focused. And this obviously nowadays doesn't cut it at all anymore.
So you have to actively manage your listings, both content-wise, manage your brand on Amazon and also run all the sponsored Ads, sponsored brands, sponsored display Ads. And what many of these endemic clients are seeing nowadays is that it can be a cutthroat business. Competition is super high, super fierce, sometimes feel like you've maxed out your PPC campaigns. And this is oftentimes when the DSP comes into play.
Paul Sonneveld
That's a great intro. I wanted to ask, for agencies that are looking to introduce DSP to their clients, you know, whether it be endemic or non-endemic, what do you think are some of the biggest misconceptions that they need to overcome?
Jens Jokschat
Yeah, I think sometimes, I mean, both agencies, but also like clients that directly run the advertising, not everyone employs an agency. They always oftentimes have the ambition and the expectation that every click, every euro or every dollar that you spend on your advertising must basically materialize into direct sales immediately.
And I think this has a couple of flaws and a couple of drawbacks, especially also when it comes to measurement, what we're going to talk about later on a bit. I think what's important to understand is that what Amazon and also other platforms obviously bring to the table is a lot of targeting options, a lot of segmentation options that allow you to find your audiences and get into the relevant set of consumers.
And especially when we also think about the non-endemic clients, what they've been using these especially the DSP for more and more is using it as a tool to diversify from their Google and Meta world that most of them, if not all of them, usually primarily use it for their marketing mix. And without diving too deep into the details here, what Amazon, from our experience, brings to the table is not only the same, but also even more precise sometimes and more specific options to segment your target groups, your personas, and basically address your consumers that at some point are likely to buy your product or use your services.
And specifically, obviously, this is related to all the buying behaviour that Amazon has. I mean, with all the shopping information, what particular products you're buying, not only for you, but also maybe for your spouse, for your family, for your kids. This reveals a lot of information about your living situation, about your spending behaviour.
And this is something that, in that particular regard, only Amazon can bring to the table and also combined by the way with the option to do a pretty perfect regional geographical targeting because obviously, Amazon has the shipping address of all their clients so they know exactly by zip code where they live and this is also an option to be used for for running Ads and targeting specifically to locations which for many non-endemic advertisers, specifically when they come from the brick and mortar store industry, for instance, retailers that have branches in certain cities, but maybe not nationwide, they can pinpoint pretty much their Ads to the locations or around the locations and the consumers, and that can be pretty powerful.
Paul Sonneveld
So, in other words, much more focused and defined, that's certainly a big distinction between some of the traditional platforms. So let's talk about DSP and AMC, right? Amazon Marketing Cloud. They're often talked about together. Can you explain how these two platforms actually, how do they interact? How do they work together? And how do you leverage AMC for better DSP campaign optimization?
Jens Jokschat
Yeah, that's a really good question that's been coming up a lot more recently, since Amazon obviously is investing a lot in building out the AMC, the Amazon Marketing Cloud Platform. And maybe to give a rough overview of how this can be utilized, it's helpful to understand again, what the Amazon DSP brings to the table. So it basically allows you to address consumers both on Amazon's owned and operated inventory or O&O inventory, as well as thousands of third-party websites, apps and inventory that other partners provide and have hooked up to the DSP. That's combined with all the buying intent data we just talked about and also that can be utilized across the entire marketing funnel from big screen TV down to actually the more transactional Ads like sponsored product Ads, for instance, that you run on Amazon themselves.
And the AMC pretty much is the part and the pillar that brings the analytics to the table. While the DSP can obviously give you reportings on how your Ads perform directly on what particular website or in what particular remarketing segment, the AMC allows you to dig much, much deeper. And actually, it's a really broad and complex topic that would probably fill an entire slot by its own. But to give you just a couple of examples, what the AMC can do is it allows you to analyze user journeys, so the touch points that particular users have with multiple of your Ads across multiple of your channels.
It allows you to also incorporate non-Amazon data. So if you're a non-endemic client, You are able pretty much to put tracking pixels on your website, to hook up Amazon tracking APIs, conversion APIs, and feedback that information into the Amazon Marketing Cloud, so you get a much broader picture of how your Ads and where they are performing. How and where they are influencing your target group's behaviour. It's also very flexible in terms of attribution models. Gonna dive into a little bit deeper on this later.
And also what's been pretty helpful, and this is still rather new and still in the process of being built out by Amazon, is the opportunity to actually build target groups in the Amazon Marketing Cloud in the AMC and export them directly as segments, as targeting segments and campaigns into the DSP.
To give you just one example is what you could do for instance is you find that people who have clicked or seen certain Ads, maybe your sponsor product Ads for really high ticket items, but they haven't purchased yet, like build a segment for these and export it to the DSP so you can have a specific targeted campaign towards these audiences after you find out, for instance, that they might be converting much better after they've been in touch with both these Ad channels. And that requires a lot of analysis sometimes, but it can be really helpful and powerful to just move away from these single view reportings that only give you information about one particular campaign or one particular keyword.
Paul Sonneveld
Yeah. Look, I think targeting is clearly one of the key strengths of Amazon DSP. And you've touched on it already, but you know, clearly the targeting is, is far superior in terms of precision and effectiveness, but you know, I'd love, maybe you can just make that a little bit more real in terms of some of the other advertising platforms out there, right. Or maybe just some examples in terms of how you can go deeper and more focused on the Amazon platform on DSP.
Jens Jokschat
Yeah, maybe quickly what was touching already on the targeting options actually we see in this slide and most notably from our point of view obviously, it's all buying-related targeting opportunities. And buying behaviour, I already touched upon, but if you really imagine what you can do on the Amazon platform is you can pin it down to people who viewed a particular ASIN or who were interested in a particular brand, maybe a competitor's brand. And you can really build these target groups out and target them specifically and that can be helpful if you want to gain market share for instance or if you want to attack in air quotes a specific competitor these type of things is going to be a node.
And without diving too deep in here but amazon also provides a type of we call it cheat sheet internally where you can actually really dig down deep and find out and look up what, if I'm an endemic advertiser that runs campaigns that link to Amazon, to my Amazon detail pages, what am I allowed to do? What specific targeting options can I use? And as different advertisers, non-endemic or someone who's not selling on Amazon, what I'm not allowed to do, for instance, one particular case that Amazon is not happy about and that therefore is not allowing it is if you run your own online store for instance and you sell products that can be sold on Amazon. You can't run targeted Ads on the Amazon DSP and link then the consumer to your own shop. This is obviously a restriction that you have to keep in mind but that's pretty detailed and pretty clear of what's allowed and what's not allowed and yeah.
Paul Sonneveld
Yeah, I was gonna say, clearly, they don't like you to basically take that traffic and redirect it to your website. Right. So, look, we've got a really interesting case study that we'll get to very shortly around the Black Friday and how this all works together. But before we do that, I just wanted to sort of ask, from an agency point of view, what are some of the most important signals or data points that agencies should pay attention to when they're running DSP campaigns?
Jens Jokschat
Yeah. I mean, it always depends on really your marketing objective. That's like the number one thing that we always want to keep in mind. And I touched on this earlier, that it's not only about driving sales and revenues or your ROAS or your ACOS. It's really depending on what type of marketing objective do you have? And what I think is also very helpful when we talk about signals and what we are actually able to measure is maybe also take a look again at the AMC and the use cases you can use the AMC for. And as I said already, it's pretty broad. It's much too complex and flexible to go through all the use cases. But I think a couple of them are probably worthwhile pointing out.
For instance, one thing that you can analyze, and that can be interesting signals when you're running like upper funnel campaigns, is to analyze how many brand searches, so people who punched in your brand name or your product specific name, how many people have been doing this after they actually saw an Ad, a TV Ad for instance, a prime video Ad or a display Ad or an Ad on fire item. So this is actually some information that can really help you to nail down the impact and understand the impact that upper funnel Ad formats or channels can have, you know.
Another example, I'm going to dive in deeper as part of the case studies, is looking at the return on Ad spend compared to the penetration within some specific user segments that can be interesting. Also helpful is what we call the time-to-conversion analysis. That's an interesting part, especially, for instance, if you sell high-ticket items. For instance, say you're an online travel agent or you sell freezers or fridges for $1,000 upwards.
These are usually products where people don't just click and buy immediately and do some impulse buying. This is something where that requires research, where you think about it, where you compare. And that means usually that the user journey, the decision-making process can be a bit longer for users, even weeks sometimes. And this also determines pretty much, if you're running retargeting campaigns, for how much, after someone showed initially interest in your product, should you be showing them Ads. And this can be analyzed by the AMC and can be really helpful.
And obviously, you can do also campaign mixes, which I mentioned earlier, where you can basically see how certain channels interact. And also, one other helpful tool that is often overlooked from our perspective is to look at different attribution models. What we all know from sponsored Ads, for instance, and from the Amazon Ad consoles is what they call a last touch attribution or last-click attribution, pretty much saying like, okay, the last Ad someone clicked or someone saw, and then that person converted, that actually that keyword, that Ad gets all the credit, it gets all the attribution, and if they clicked or saw other Ads before, it's worth zero, it's worth nothing.
And this hardly ever gives the right picture, and with the AMC you are more flexible in terms of looking at it from different attribution model standpoints, for instance, a linear attribution. You can just analyze your data and your campaigns by assuming like every Ad that a user saw before they made a final purchase decision has some value and you can attribute it towards every Ad partially. That can also be helpful in finding out actually what type of channels and Ads contribute to final sales instead of just focusing on the last touch.
Paul Sonneveld
That is a great overview and some great steers for other agencies who are venturing into this space. Thank you for that. Let's make it real. I mean, this sparked a lot of interest for me. When you and I were having lunch in Hamburg a couple of months ago, we started talking about these, some of the experiments that we were running at the time around Black Friday and it would be great to revisit that and share some of the learnings with our audience today. Maybe let's just start with, can you just walk us through some of the structures and experiments, structure of campaigns and experiments that you run during Black Friday at the end of last year?
Jens Jokschat
Absolutely, happy to do that. Before we run into the concrete Black Friday examples, I also wanted to share some just examples in a more general fashion that we also did for non-endemic clients. The Black Week example that you're talking about, specifically run for the endemic clients, because we wanted to find out pretty much how can also the upper funnel, the prime video Ad, the TV Ads, can they really drive sales for the deals that advertisers are promoting during Black Friday and the Black Week. But even on top of that, it's really interesting what you can find out by using these type of analysis.
And just to give you a quick intro here, maybe leading up to these examples, is we did one case study for a company pretty much manufacturing and selling luggage, so travel accessories on Amazon. And the question we ask is pretty much, okay, this client has been doing obviously sponsor product Ads, PPC Ads ever since. And what happens if we now bring display Ads to the mix? We use the the Amazon DSP to just run display Ads, both for target segments that have not been touched with sponsored Ads before, but also as a type of retargeting people who clicked on sponsored Ads already but haven't purchased yet.
And what we found out actually is two things primarily. One thing is that it works pretty well. To retarget users that have clicked on certain Ads. So you find a decent amount of people that you can address with display Ads on top of it. And the primary outcome was also that if people get in touch with more than just one Ad format and one click, that it can highly improve the conversion probability or conversion rate. And this was really helpful in understanding how it can drive sales on top of your sponsored Ads campaign that's been existing already. So that was, I think, a good example of how these two channels can work together.
And when you look then specifically at the streaming TV options that we've talked about, like Prime Video Ads and Fire TV Ads, that was for a different client, online travel agents, or non-endemic in this case. The objective here actually was finding out, OK, What happens if we combine TV Ads, like activating users on the big screen, lean back situation in the living room, but obviously they can't click on a TV Ad, they can't do any direct response, but we want them at the website of the advertiser, you know. And so what we did is we analyzed pretty much, okay, how many of the people that saw the TV Ads ended up on the website of the advertiser, so probably got out their phone and punched in their website.
But what happens if we run display Ads in conjunction with it? Because display Ads obviously they run on the mobile phones, they run on desktop Ads, they are clickable, so it's much easier for these users. If they see, okay, I've seen this company, I've seen this Ad, I maybe like their products, let's check them out. And what we found out that. So if you combine those two channels with each other, you not only have a much higher frequency of Ads that you can serve to people, because obviously on a TV, we're able to serve between two or three Ads on average per person, which is not super high.
But if you add display Ads to the mix, it was more than 10x of that. So that's actually good for driving more impact and Ad exposure. But even more important was that we found out that the probability, the likelihood of people ending up on the website increased 14 fold. And that was also a much stronger outcome than what we had hoped for. And that was really, really interesting to see.
And so one of the low-hanging fruits that we nowadays do with clients is pretty much whenever we run TV campaigns for them on Prime Video or Fire TV, that we recommend basically, hey, the same users that already saw you on TV, why not serve them, by the way, much cheaper in terms of CPM, some display Ads, so they have an easy time of getting in touch.
And last case here, that is the audience analysis. I mentioned one analysis we did for a beauty client. So selling beauty products here is what you can also find out is pretty much if you target a certain segment, in this case, for instance, people that were new to the category beauty, they were buying beauty products for the first time on Amazon. That's like a lifestyle targeting you can target. what we found out as part of this campaign is that the penetration of users we actually reached was rather low. So we only reached six percent of users as part of this display campaign here in a campaign but at the same time this segment performed with the highest ROAS.
So it was really efficient to show Ads to these users and if you compare to other segments and it's pretty easy to use it for optimization purposes and you can just basically allocate more budget, for instance, to these segments that perform really efficiently, and maybe reduce budgets in other segments, as long as you know, okay, there's still a large portion of these segments that I haven't even reached yet, you know, and this is also information that we can derive from this.
And by the way, if people really want to dig in deep into the audience segmentation, Amazon offers some paid features as well that allow brands pretty much to, if they're willing to spend a couple of extra bucks, to analyze market shares, analyze user segments that have not been in touch with any of their Ads to understand more pretty much about the potential, if they will broaden their investments. So that's also an interesting option, especially for marketers with a bit deeper pockets that really want to optimise their marketing mix now.
Paul Sonneveld
Fantastic. Thanks, Jen. Some really great, I love each of those three case studies with clear insights and actions and what does this mean? It does make me pause and ask, was there anything that didn't work? Things that surprised you, go, gee, I thought that was going to be a winning strategy and it was a complete flop. Sometimes there are valuable learnings in that too. Anything that meets those criteria?
Jens Jokschat
Yeah, absolutely. Because I mean, obviously, what you can come up with this hypothesis, you try to ask yourself certain questions and then see whether they are right or wrong. And giving you more examples of specific examples about our Black Week cases that I want to shed some light on is it's helpful to understand again, because that's something we haven't spoken about yet in detail.
These prime video Ads, the TV Ads that Amazon now allows you to run against their audiences. What does this really incorporate? And when we talk about prime video Ads, what we usually use it for is really for the big screen. I mean, you could also run it on apps that people have installed on their phones to watch movies. But the biggest impact, obviously, is if you have it on your big screen in your living room. And Amazon pretty much allows you, since last year, to run these full screen Ads from 15 seconds to even 60-second spots on people's TVs. It has large reach.
CPMs, obviously, are much higher than what you would expect from display Ads. But it's non-skippable. It's also non-clickable. That's also important to know. So there's no direct response opportunity, obviously, as you know from TV. And pretty much the same with Fire TV. Fire TV being a bit different in terms of that it's just shown on the starting page. The ones from the audience that maybe utilize the Fire TV stick at home or have a TV set know how this works, that there's a spot called what they call feature rotator in the upper part of the screen, where usually some movie trailers, for instance, are being played. But more and more, it's also being utilized and offered for TV Ads.
So these Ads you can run pretty much on the big screen. And we wanted to find out, OK, if we combine these into the marketing mix with Amazon, can these Ads actually drive sales through the Black Week? And we have three clients of ours that we could convince to run this test with us. One is from the supplements industry. Another one was a beauty client. The third one was pretty much selling cookware. And we wanted to find out okay during this black week all these clients they had special promotions obvious special deals special rebates that we wanted to communicate in their tv spots is this suitable and efficient to basically drive sales on their amazon deals and product pages and to what extent does it even make sense or is it not really helpful.
So what we did is We combined these Prime Video Ads and Fire TV Ads with display Ads. Like I said earlier on, the display Ads are great for pretty much offering a response channel to users after they've seen the TV Ads. And obviously, all these clients, by the way, were also running their regular sponsored Ads on the Amazon website. Obviously, that was part of the picture. So it's a total of four channels that were utilized. And a couple of outcomes here, I think, that were pretty insightful was the first question we asked ourselves was, can these prime video Ads, if people got in touch with the TV Ads, does it drive brand searches? So do people get active somehow?
And what we saw here in one case, for instance, was that, yes, indeed, after audiences saw prime video Ads on the big screen, their probability to punch in the brand name into their Amazon search bar increased by factor 2.4. So 4.2x higher brand searches after TV contact, that was pretty interesting. At the same time, we also discovered these people who were obviously triggered to research more on this brand were a bit less likely to convert.
So the interest was not immediately translated into more sales, yes, to a certain extent. But obviously, there were also people interested in just finding out more, but not immediately buying. And then we looked at what happens if we added Fire TV to the mix. So the people that we reached on both of these channels, so obviously more contacts and more impact. And the factor, actually, of brand searches increased even a bit more to 5x. And in this case, interestingly, also the purchase probability increased. That was an early indicator here that obviously we're reaching the right audiences and the ones that have been activated through the TV Ads would be willing to buy as well. So that was also helpful.
And in terms of what works, what didn't work, it's also important to get an understanding ideally of how many Ads do you want to show you and that was also a question we asked ourselves? So if we run these different formats across the DSP TV display across different channels and You keep dishing out these Ads and this was obviously in the black week rather short period of time, the total campaign ran I think across 10 days or more and so what we then did is finding out, okay, for how many Ads are we actually able to increase sales probability? What's the ideal sweet spot, pretty much, of serving Ads? And at some point, obviously, you will be spending more money, you even send out more Ads and reach consumers more, but their likelihood to purchase doesn't increase anymore.
On a very high level what we found out across the board here was that apparently in these cases at least in these three cases the sweet spot seemed to be between 12 or 16 impressions somewhere that can be analyzed much deeper and will be done but as a rough indicator here, that can be helpful because obviously this is a metric that can be controlled by setting frequency caps or even optimizing frequency on the DSP this is not just a nice to know information you can really use this to increase your efficiency of your campaign setup. So that was pretty interesting.
And then also, that's touching on both the question what worked and what didn't work, we took a look at these different kind of user journeys. So when we hit users with Ads across certain channels here, how did that influence the likelihood of purchases? And on a very high level, again, what we could see is, if we hit users on only just one channel, specifically, they only saw a Prime Video Ad or only a Fire TV Ad, that didn't activate them really. So it didn't make the cut at the lowest purchase rate of all user journeys, so to speak. And even interestingly, if only you hit them with a PPC Ad, that was also not one of the top performing options here.
So in simple terms, the more channels, the more multiple touch points you get with users, the more likely it is that they finally end up purchasing. And if you look at the real top two options here, what we found interesting to see is that definitely PPC obviously was part of the picture. That's not a question that you would want to run your sponsored Ads and pick up your users once they're on Amazon and they find your brand there. But also, both Fire TV and Prime Video seem to contribute to the likelihood. And what we just assume, obviously, that would need some further clarification is that if you're seen on TV, that probably gives your brand credibility. That makes you being a TV brand. And yeah, helped in total definitely drive the probability of people purchasing here.
Paul Sonneveld
That's fantastic, Jens. There's so much detail there in terms of case studies and insights and metrics. Very compelling. So, you know, I appreciate that. We are at the 34-minute mark, so we're a little bit over time. So, But before we wrap up, I do want to sort of just ask just a couple of questions really from an agency point of view, right? Agencies that are trying to build capability in this space. I had a series of questions here. I won't get to all of them.
But just let me just ask this one, for example. You know, measurement and attribution. When it comes to this measurement and attribution, you know what? What are some of the best practices that agencies should really think hard about to truly understand the impact of their DSP campaigns? You've been in this game for a long time. What advice do you have on this front?
Jens Jokschat
Yeah, I think what we all learned for many, many years in online marketing in general, that thankfully all the direct response, if someone clicks somewhere or someone takes an action on a website, it can be measured, you know, whether it be cookie-based or with other measurement systems, but the direct response usually can be measured.
And as we now start talking about upper funnel metrics, where people might not be really inclined to buy immediately, you're getting more and more into the field of what's been done in the past by market research. I mean, market research has been around for decades, where people are just being asked through panels, for instance, hey, did you see an Ad? Are you aware of certain brands? To figure out how impactful actually marketing campaigns were. And the interesting part here is that Amazon also brings this to the table because, I mean, we said earlier on a TV Ad that can't be clicked, Prime Video, Fire TV, yes, but Prime Video, no. You can't even take out a direct response and measure how impactful this Ad directly is.
So what Amazon does is that's something that agencies and marketers can use. It's just beyond the typical e-commerce KPIs that you can select pretty much from the DSP. What do you want to track? And you can differentiate between tons of e-commerce KPIs that you can track on your own website or from the Amazon detail pages. You can nowadays also utilize what Amazon called brand lift studies. This is also not super new for the people who've been dealing with Google or Meta for some years. Because the same concept is available on those platforms as well.
But what we find really interesting is that Amazon has set up this, what they call the shopper panel. It's pretty much a panel of users that have signed up pretty much to be asked about certain brand research surveys and so if you run a campaign from a certain amount of budget what you can do is you can set up in self-service on the DSP this type of brand lift study where you can ask pretty much about very traditional marketing metrics like brand awareness, Ad recall, preferability, purchase intent, that's like these typical steps along the funnel.
You can ask people, hey, do you know a particular brand? Or do you remember seeing an Ad from this brand? And you can also compare yourself against competitors. So which of the following brands do you prefer, your own being one of them? And also ask, are you inclined to buy this product? And then people can just answer what they think. And what is really interesting is, and just give you maybe one or two quick examples here, is if you look at these metrics, they are really easy to understand from these brand lift studies.
And one example we did here for a client where you could see, OK, along this funnel of awareness, Ad recall, preference, intent, what Amazon does is they have a control group of people who have been not exposed to certain campaigns and compare them to people who have been exposed to these campaigns and can directly see, okay, how much of an uplift was there, how much of an impact. Do more people actually know my brand now? Do more of them remember seeing my Ad? Are they more likely to buy my brand here? And that's really helpful, I think, from an agency and advertiser perspective to have these types of measurement options beyond direct response metrics that you see in any consoles.
And this can also help you not only by answering, am I more known or popular right now with my brand, it can also help you in optimizing your campaigns because you can even drill this down to see how does awareness uplift look like in terms of different Ad exposures, for instance. And in this example, for instance, you would see that if people had 10 or more Ads seen, then their uplift, their record is much, much higher than if they've been only exposed to one Ad, for instance. So yeah, that is pretty helpful and insightful, especially when it comes to these upper funnel marketing formats.
Paul Sonneveld
Thank you, Jens. Really interesting. I wasn't even aware that Amazon offered those shopper panels. So thank you for drawing attention to that. We are at the 39-minute mark, so we're going to have to close it out. But I do want to do our audience a little bit of justice. It is a live session. I did ask for questions. We have a couple of questions in the room here or in the comments. We won't be able to get to all of them, but I just want to ask you two of them. And I just want to say thank you to our audience for hanging in there. It looks like a lot of people are staying back. I really appreciate that.
So let's start with a question from Mohamed. I'll read it out to you so you can sort of digest it. What if our click-through rate is good, like 4% to 5%, but unfortunately, our conversion rate, not very good, like 15% in DSP Ads? What matrix needs to be checked, please? Or maybe more broadly, what advice or questions would you ask that maybe can help shed some light here?
Jens Jokschat
Yeah, that's a good point. So the first thing that I would probably look at is where does the good CTR come from? Because some cases we've seen that marketers sometimes tend to optimize for high click-through rates because it's assumed that this shows real interest. That's not always the case. In terms of inventory, for instance, it's pretty easy to get high CTRs if you start focusing on mobile inventory, for instance. So mobile Ads have much higher click-through rates, generally spoken, than desktop Ads.
So one of the reasons here might be that the clicks that you're actually seeing at the high CTR is not coming all of the time from users that really show an interest in purchasing the product. So inventory analysis and optimization And also looking at what type of devices are you serving your Ads on might even help optimize here. But if this doesn't find the truth, then obviously conversion rates could be also analyzed by maybe optimizing product detail pages or your website, wherever the conversion is happening.
Paul Sonneveld
Great question, Mohammad. Thank you for that. Last question here. It's actually from, I don't know who's asked this, but it's just come in. And I think it's a little bit of the elephant in the room type question, right? So how much sales increase did you experience after utilizing DSP and AMC? From previous experience, this person saying the DSP stats, they look good, but they did not truly increase the overall top-line sales in the long term. I think it's a very, pertinent question. Your thoughts, Jens.
Jens Jokschat
Yeah, absolutely valid question, because I would completely agree that you can easily dish out tons of DSP Ads, specifically display Ads, and probably find users that at some point would have purchased anyway, but it doesn't drive any incremental sales. So the answer in this type, in this case, as usual, it depends. I think what you would want to really set up to find this out is have some setup put in place where you can really find out under equal circumstances. What are the usual baseline sales that you see with your traditional setup of marketing campaigns? And if you add certain segments, if you add certain campaigns or add formats to the mix, how does this influence incremental sales? And we've seen the results from 0 to 15 or 20 percent.
One general tip probably or general experience is that obviously, if you want to run a DSP Ad specifically also on TV, if you have any promotional, any specific offers that can really activate consumers, it's much more impactful in driving immediate sales. And if you would just rather do some sort of brand advertising where you might focus on your products, USPs or benefits, promotional, price promotion specifically always have a strong impact and can also have a strong impact in driving incremental sales. So that's all we see.
Paul Sonneveld
Great. That was a great answer to a great final question. Jens, thank you so much. Unfortunately, we have to wrap it up today, but I really appreciate you joining me on the show today. I appreciate you bringing lots of insights case studies, metrics in relation to DSP, AMC, and how it all works together. Really, really appreciate your contribution here.
I hope for our audience that this has stirred some excitement or interest or maybe has given you some inspiration to try something new or pursue some different strategies here. Jens, for those that are interested in maybe learning more about your service and what you do at PrimeUp, what is the best way to get in touch?
Jens Jokschat
Yeah, just shoot me an email or get in touch through our website, especially around the DSP topic. You can directly address our team here through this email on top of the screen. And yeah, happy to provide further information, input, discussions on how you or your brands or your agencies can utilize DSP and the different formats it provides. And yeah, just shoot me an email and get in touch. I'm happy to hear from you.
Paul Sonneveld
Fantastic. Jens, thank you so much. Take care.
Jens Jokschat
Thanks for having me.
Paul Sonneveld
All right, everyone, that is a wrap for today's live episode of Marketplace Masters. Thank you so much for tuning in today. Now, of course, if you are looking for more insights, please ensure to check out our video on-demand library at merchantspring.io for a wealth of video content. And of course, if you are looking for analytics, whether it be agency-level analytics for Amazon or advertising, feel free to reach out to me. I'd love to show you what Merchantspring is all about.
And last, before I forget, last but not least, I'm always on the lookout for new topics, new content, and things and topics that you would like me to tackle next. So if there is an interesting idea that you'd like me to explore and perhaps find the right speaker or expert for, please send me a message and I will do the research and recruitment for you. All right. Until next time, take care. See you then.
