Our transcripts are generated by AI. Please excuse any typos and if you have any specific questions please email email@example.com.
Speaker 1 (00:00):
Welcome to unpacking the Digital shelf where we explore brand manufacturing in the digital age. Hey everyone. Peter Crosby here from the Digital Shelf Institute with market share and profitability becoming ever more important in this economic era, using data to get more revenue juice out of every marketing squeeze is on the top of everyone's list. Today's guest thinks that a good amount of that juice can come from a savvy use of Amazon Marketing Cloud, Ross Walker, retailer marketplace's, paid media team lead at Acadia. Join the podcast to walk Lauren AK Gilbert and me through the data that will drive the business metrics you want from your Amazon marketing campaigns. Trigger warning, incremental ROAS may be discussed. So Ross, thank you so much for being on the podcast. We are thrilled to dive in on the topic of Amazon Marketing Cloud with you today.
Speaker 2 (01:02):
Thanks so much for having me.
Speaker 1 (01:04):
Yeah, I think so much of this present moment that we find ourselves in is about trying to squeeze performance out of every marketing dollar spent, every campaign, every product page, et cetera. And Amazon Marketing Cloud really does feel like an effort on Amazon's part to provide data that can help their brands do it. And therefore Amazon makes more money, we make more money, they make more money, and the consumers are hopefully happier. They'd always done that, but they really have upleveled the playing field of that with Amazon Marketing Cloud. So we are delighted to have your brain here. So tell us more about AMC as we can shorten it for the rest of the podcast and why it's important to brand manufacturers who are trying to make money on Amazon.
Speaker 2 (01:55):
Yeah, it would be my pleasure. So AMC is, it's kind of all over right now and you see in a lot of thought leadership it getting mentioned, and you'll probably have heard an Amazon rep mention it. Here's what we could do. And maybe you have a vendor calling you saying, we can provide AMC services. And it is all kind of nebulous and it also sounds super complex, but I think at base what AMC is, it's an open database. So it's a place where you can ask Amazon questions about the data of your account and complex questions, questions that we haven't been able to ask before about how your advertising efforts for sponsored products interact with your advertising efforts for DSP and programmatic about how your streaming TV investment later impacts your sponsored product investment. So there's a lot of really cool data that's now available to us that never was before. And I think the main reason why is because Amazon has so much data, their problem is actually, well, how do we package this and get it to people? And so AMC is the first really big new foray into that to offering you the ability to ask complex questions and not wait for them to develop knowledge products for you to consume.
So I think that in a nutshell, that's what it is, and it can be both very complex and require someone to know how to write SQL code and not just any SQL code but the SQL code that Amazon uses. So that's a very unique skillset that raises a bit of a barrier, but that's not all that it has available. There's also a set of templated questions or queries that any brand can use to get insights into how their advertising is performing in a deeper level. So that I think I would say is what AMC is in a nutshell. And it's super valuable because a lot of it is net new data that helps us make tactical decisions that can improve our performance today.
Speaker 1 (04:02):
Yeah, we had a really smart Amazon expert on our podcast, Martin Hoyle, who has his own firm called Consult. He's super, super smart. And he was saying one of the trends that he's seeing is Amazon, is that they're trying to have fewer and fewer humans doing things. And essentially his wrap up was kind of like your growth on Amazon is essentially your total responsibility. Amazon's not going to have a lot of people or human resources or even their joint business planning is getting a little bit more distant or spread thin. And so they're building these things to automate as much of it as possible and put the onus on it. That was the first time I'd sort of heard, that doesn't surprise me, but does that make sense? Yeah, that's
Speaker 2 (04:49):
A great way, that's a great way to think about it. They've outsourced the analysis and even the data pulling portion of the business, so that what would've required an army of people on their side to deliver insights. They've kind of just said like, okay, here have at. There you go.
Speaker 1 (05:07):
Good luck with that. Figure
Speaker 2 (05:08):
It out. Which is great, honestly, I much prefer that,
Speaker 3 (05:13):
And I'm sure brands also prefer it because they're getting access to the data, but they have to have the ability to splice and dice and figure out what it's saying. But I can imagine it paints a much broader picture than they've ever been able to look at where I feel like previously you get one dataset and you'd have to have an analytics team internally to take that dataset to match it to your sales data to match it to your other dataset to make that full picture. But it sounds like you can do that with Amazon Marketing Cloud. Is that the case and how would a brand think about doing that?
Speaker 2 (05:45):
And that's a great question. So it's like if a lot of things in the past seemed causal, but a lot of activities in different areas seemed causal, but there was no way to link them together. What AMC does is it just breaks down the barrier and it allows you to say like, all right, when I invest in this area of my business, if I spend more on certain DSP campaigns, this is how my sponsored product campaigns are affected. I can see my customer journey from first seeing an awareness campaign to them finally making a purchase when they search for my branded term, far down the line. And with that data, it's always been a challenge how much data there is, and most brands sort of like are a wash. And now I think they feel more than ever now that they're a wash in data, they don't know what to do with it all.
And so what we've tried to do as much as possible is to open up the storytelling that comes with AMC so you can make really good tactical decisions. And we've zeroed in on some queries that we think are much more impactful for the short and medium term that matter to e-commerce managers and a whole other set of queries that are really valuable for managers of brand and marketing budget and where their KPIs are very different. So there's really something for everyone. There's something for the executive that wants to understand which of their products are driving the most new to brand customers, so we know where to invest in new product development in the future. There's a set of queries that are really great for telling you which of your campaigns are the ones that are helping to drive final conversion and what the path to conversion looks like for your e-commerce manager so they know where to invest their budget in different tactics. And then there's something in there for the super data nerds to let you know what frequency and on what device is the optimal media mix. So there's really something for everyone.
Speaker 3 (07:46):
And could you use it to target a specific audience? Have you seen a brand say, okay, I'm looking for this kind of shopper in this area, or who shops at this time of day? Is that how you're seeing a lot of brands using AMC from a targeted audience perspective?
Speaker 2 (08:03):
So AMC doesn't add any kind of newness to that kind of targeting.
You can do with Amazon DSP you can do. And now with Amazon sponsored products, there's some new day parting features to be able to target the time of day a little bit more effectively. But what AMC does do when it does give you an insight into is more of the holistic questions about what time of day people might be converting best or what time of day your ROAS is best. Those are really great insights to be able to then layer into other tools that actually activate on your campaigns. From an audience perspective, what's new for DSP is a set of queries that we always knew should exist but never did in the past. So a set of audiences that always knew should be available but never were. So for example, if you're in the e-commerce world, you'd be very familiar with a campaign that hits your cart.
So you can target people who have added to cart but not purchased, and there was nowhere in the entire Amazon ecosystem for you to be able to do that until AMC came around. And so now AMC says like, dear Amazon, give me a list of all the people who've added to card in the last 90 days and not made a purchase and turn that into an actionable audience that I can leverage on DSP. And we found, for example, that in the lead up to a big event like a Prime day, those kinds of campaigns perform really well. They remind people who've already raised their hand to show interest that the product is now on promotion, they're increasing their likelihood to purchase. But then incredibly after the event, tons and tons of people will have added the cart, added their product to cart, and then done more research and done more consideration and decided, you know what? This is the product I want to buy. I just discovered it on Prime Day, but I wasn't ready to purchase it. But a month later, yeah, now I am. We see that that DSP campaign that's utilizing that unique AMC audience is performing incredibly well. That was a really cool insight. We wouldn't have had that without AMC. That's completely net new.
Speaker 1 (10:12):
I'm picturing a whole bunch of prime campaigns that are just think about it, come back to it in a few days and pay us more money. Right?
Speaker 2 (10:21):
I mean, honestly, the focus of a prime day is always like, now's the time to buy, but it's, we've always said that it's one of the best times to increase your reach to touch new customers. And now we have a bunch of really interesting tools through AMC to retarget some of those unique behaviors. So another really unique audience that you can build out of AMC is to say, for every person that was exposed to campaign X, I want a list of those customers that didn't purchase. So if a bunch of your customers are exposed to one of your campaigns, they click but they don't purchase or they view, but they don't purchase, I can now set up a follow-up campaign for all of those people. So this is a set of people who've been exposed to message one, let's serve them message two. And that is a really unique opportunity in the DSP platform that's been really, really powerful for our streaming tv, more like upper funnel campaign. So for someone who's watching something on their connected TV and they get served a streaming ad, I can now retarget that whole group of people who saw my video ad with a more e-commerce focused ad. And those have been really impactful in increasing the ROAS of our overall marketing.
Speaker 1 (11:36):
And do you see that these kinds of impacts, I'm thinking of as we were talking about the opening, the desire to drive more top line growth through existing without unnatural acts. Do you feel like this presents the opportunity to actually forecast what the uplift of AMC could be to my business if I'm doing it? Do you have a sense of, I use the word promise, I'm doing the air quotes, but
Speaker 2 (12:07):
Speaker 1 (12:07):
You can promise to a client like you give us six months, we'll give you X, or what are you seeing there?
Speaker 2 (12:13):
I don't really see the opportunity to put a number value on utilizing AMC just yet because it's super dependent on each brand and it's super dependent on what kind of tactics they are running. But I will say this, for all of those brands for which we're doing full funnel marketing right now, or even smaller brands that aren't doing too much full funnel marketing that are keeping things pretty close to just Amazon maybe dipping their toes in DSP, that what AMC has done is it's unlocked all sorts of insights that has helped them make marketing decisions and catalog decisions that has grown their top line revenue and help with the bottom line overall. So for example, with AMC, we're identifying which of the products in the assortment is the best at bringing new to brand customers. We're putting more dollars behind those products and we're seeing the channel grow as a result because those are the products that people are willing to pick up at a higher rate. And so for a lot of brands that have been stable or plateau for plateauing for a long time, they don't know where to place their betts. AMC is really helpful in helping you place smarter Betts.
Speaker 1 (13:24):
So one of these things that, I mean, you've been talking a lot about tactical campaigns that are finding opportunities in the moment and going, but is there a way in which AMC can help you predict or think about future purchases to lead to opportunities that the consumers may not even necessarily have thought of themselves yet?
Speaker 2 (13:50):
There's a couple of really cool queries. One of them is called the, I mean, I hate to say it out loud, it sounds so corny. I don't know who's in charge of naming these queries at Amazon, but it's called the Ace and Overlap report for upselling. And I'm just in love with this report. It does exactly what it says it's going to do. It tells you after someone purchases product number one, what is the likelihood that they're going to purchase product number two and product number three and product number four. So it takes your whole, say you have 20 different products that you're selling, it will tell you after someone buys your hero product, 10% of them are very likely to buy this other product in the assortment, and 2% of them are likely to buy a third product in your assortment. And if I know that data, then I can calculate really clearly what the future total growth of the account could be if we sell more of the first product. So super, super exciting for us to be able to give clients insights into how we can merchandise their product better, how we can create more virtual bundles that package together, products that people are likely to, and also how we can cross market upsell their current assortment in a way. So I can say, yeah, people are really likely to buy these two products together based on the AMC data. Let's market them together.
Speaker 1 (15:06):
So talk to me about, this makes me think a lot. There's obviously, when I think about the org that's required to make use of this,
And it seems like there must be kind of a maturity curve of your customers that have not even heard of AMC yet or something, and then all the way to people that are organizing around it in a very sophisticated way. How do you think about, and in your case, their clients, I imagine, how do you think about that range of clients and what do you find that those that are best successful have in-house that they need? And then what do they do with you? What is that to make it the biggest use of this? How should you organize yourself around it?
Speaker 2 (15:52):
I'm a bit biased on this question. I know you are. Thanks
Speaker 1 (15:54):
For asking. People have their grain of salt out, Ross, but I think it's important because you spend every day in this, so I'm giving you a chance.
Speaker 2 (16:02):
I have yet to work with a brand that has outsourced AMC insights to a third party tool that delivers them a dashboard or a database or recommendations. I've yet to meet a brand that's utilizing that in a way that successfully drives actual changes in their account. So a lot of the times companies will pay for a third party tool, they'll get access to a ton of data and then no recommendations, no idea what to do with it. And they're in the same place there were before, but just with more data. Or they partner with Amazon and Amazon says, we'll do the AMC for you. We'll help build you a dashboard. We'll give you the insights so they get another report. No idea what to do with it, no recommendation that comes with it. So what we've tried, what I think we've thread the needle really well on, and what I encourage any brand to do is to work very closely with your agency to look at the queries that are available and then to just set touch points and say like every month or every quarter, let's look at this data, let's see how it's changed, and let's be very intentional about developing a recommendation set that comes from the AMC insights that I've seen.
So I think it's important to have someone on the team, on your agency side who is very knowledgeable about what the data is that's available and what are the new queries that are available because it's always developing. So you need to have that. You need to have an agency partner or someone in-house that can do that. You don't have to have someone who's really SQL savvy to be entry level. You can get away with, I think a lot of AMC insights without that, when you move to the medium tier of AMC utilization, then you might start to be ingesting someone a little bit more in-house to have someone who's analyzing the data more frequently. And the highest tier of AMC use would be connecting via the API building a lot of dashboards for your team. But then you've got to have someone to interpret that data. It's not enough just to collect it. You've got to have someone who's dedicated, who can boil it down to tell a story to say, okay, CEO E-Commerce Manager, here's how our budget allocation needs to change. Here's how our marketing should change. Here's how our catalog needs to change. So you've got to have that person with that capacity somewhere sitting in the mix. Otherwise it's just more data and it's not functionally useful.
Speaker 3 (18:30):
And what I'm also hearing from you is AMC is not this golden thing that's going to solve all problems. You use it in conjunction with the other data sets that Amazon provides to really paint that whole picture. Is that true based on what I'm hearing from you?
Speaker 2 (18:46):
Yeah, a hundred percent. And even on my team, I've got team members who will, okay, I'm really excited about AMC and I've got the report and I provide it to the client, and then, okay, nothing happened. And I'm like, all right, well hold on a second. Let's look at this. You've got to have the data and you've got to have some analysis, and you have to have a recommendation that's attached to it. And that's the hardest part. You have to be somewhat creative about it in order to make it work for you.
Speaker 3 (19:11):
And I might be incorrect in this, so let me know if I'm totally off base, but I believe you can also upload your own data to AMC, is that correct? And they totally, they get rid of any of the proprietary information and they make it easy for you to de-identify, is the word I'm trying to find, right? Correct. And then you can use that data in conjunction with what's already in AMC.
Speaker 2 (19:32):
That's right. So part of what makes AMC even more unique is it's not just a, it is functionally meant to act as what's called a clean room, which is to say that anonymous data can pass in and out and no one will know exactly who the users are, which is a really important promise of Amazon that they don't expose your user data to your advertiser or to anyone really. It's only they know who your users are. So it does that with its own databases, its own data about DSP and PPC and Amazon attribution. It pulls all of that and anonymously will tell you this number of users do this action but not who those users are. And it can do the same thing with your own CRM data as well. So you can upload or even pipe in via API your own CRM data to understand, so here's all the customers that we have currently that can all be uploaded and anonymized.
So Amazon doesn't know who your customers are currently, but they can take the metadata around it, where they live, what their zip code is, what their phone number is, and match it up with their Amazon customer database and say, okay, your website customer is also an Amazon customer and their path to purchase looks like this. So they started on Amazon and they ended on your website, or they started on your website and they ended on Amazon. And you can create lookalike audiences from that and say like, all right, the core customer set that I have on my website, I want to create a lookalike audience and target them with Amazon DSP. And those are tactics that are very familiar for anyone who's in the Facebook marketing world or any just kind of e-commerce marketing. Those are tactics that they'd be familiar with. And Amazon with the Marketing Cloud has added that capability to Amazon sellers.
Speaker 3 (21:20):
And do you find that a lot of brands do that and they're uploading their own data or are they more No. No. Okay. No,
Speaker 2 (21:29):
Very few brands are utilizing this capacity, mostly because they're cagey about sharing their data with Amazon. But also I think the use case hasn't been proven out too strongly yet that there's a lot of incremental value from going down this path. But for me, I think it is the future of a holistic marketing strategy not to segment Amazon and e-commerce into a silo and just think, no, this is where we can do a lot of marketing efforts here. Let's have maximum data available.
Speaker 3 (21:58):
So in terms of AMC and what is available, what else is there from a customer standpoint in terms of new brand data or new to brand data? Is there anything around that that brands can use to identify who's new to their products or their brand?
Speaker 2 (22:13):
I think this was one of the things that I was most excited about when AMC first got released. And it was the ability to understand how some of my current marketing efforts on Amazon are performing from a neuter brand perspective. So it's always been the case that we only knew this neuter brand metric, which is to say someone who's purchased from our brand and who hasn't purchased from our brand in the last 12 months is counted as new to brand by Amazon. So we only had this metric for a couple of ad types sponsored brands and sponsored display. And that typically makes up less than 20% of most brands investment in the Amazon ecosystem. So a lot of brands are investing 80% of their budget in sponsored product ads, which is to say search ads. And they never knew how good those ads were at driving new to brand customer acquisition because for no reason at all, you don't have that KPI associated with that ad type, but AMC does surface it.
So that was one of the things that I've been most, it's one of the first queries that I run for any brand is to say, let's look at your actual data. Let's see which campaigns are the best at driving new to brand customers, and let's include that in our evaluation and how impactful and effective they are. So I can take a brand that's been spending very little money on a certain campaign type or targeting using keywords that don't perform super well from an ACOs or ROAS perspective. And I can show them that most of their new to brand customers are being driven by this campaign, and we can scale it up and see great impact to the revenue of our account.
Speaker 1 (23:49):
Ross, when you think about that balance in this next couple of years, uncertain economy and all the blah blah things we know about the macroeconomic environment, I do you feel like what is the balance between new to brand, which I think is often a lot, like you said, where people put their attention versus getting more out of the people we already know. Do you feel a shift happening there? Is there sort of new logo, new logo versus existing upsell kind of thing?
Speaker 2 (24:27):
I think that as Amazon sellers and vendors have become more savvy over the years, I think that neuter brand was a metric that a lot of brands cared about because it helped to show incremental new customers, which is a super valuable KPI for anyone looking at growth. But more important now than ever, I think has been the value in incremental growth for any brand and neuter brand is a great KPI to show that. So I think it's only gotten more important for brands to understand that metric. And in many ways it's a better metric than ROAS to go off of when you're looking at the effectiveness of any one channel. And I've seen a lot of retail media networks bring that KPI to their own platforms because it's such a powerful story. However, as brands have become a little bit more savvy, I think it has actually lost a little bit of its interest.
I think brands are much more interested in overall market share now than they've ever been before. And so new to brand feels it's valuable, but it's started to become a little bit more of a vanity metric in the way that ROAS has become very much a vanity metric. You can kind of achieve ROAS with any mix of advertising that you want. And now I think customers are understanding like, yeah, neuter brand is actually something that you can kind of game as well. What I really care about is market share and overall profitability. So I think it's lost a little bit of its luster there, but it's still super valuable.
Speaker 3 (26:04):
Ross, what do you think of I roas, I've heard a lot more people talking about ROAS in that way. Is that something that you're talking to your clients about and using as metric,
Speaker 2 (26:12):
Let's go down that route. I've seen that a lot. It's super common and I think they've done a great job branding the term, even though it's kind of like a bog standard concept.
Speaker 1 (26:27):
I'll be the dumb person here. Could you define I in iro?
Speaker 2 (26:32):
The I and IRO ROAS is incremental roas. Oh,
Speaker 1 (26:35):
Thank you. Sorry.
Speaker 2 (26:35):
If you're already spending $10, if I spend $15, what's the incremental sales lift that I should expect on top of what I normally sell? And then what's the incremental ROAS that's attached to it? And it's often calculated as a function, as an opportunity function, which is to say, if I spend more in this area, I have a much higher likelihood of driving incremental sales than another. So you use it as a value to evaluate where you should be spending to grow incrementally. And I'm a big fan of the concept. I don't call it I oss, but I'm a big, big fan of the concept.
Speaker 1 (27:13):
Why do you not call it your came
Speaker 2 (27:15):
Out? It came out a month or two ago ago. Someone very clever put that one together. And I guess I'm kicking myself for not having thought of it through.
Speaker 1 (27:26):
So is it trademarked by someone? Anyway, we can talk about this. It
Speaker 2 (27:29):
Feels like it at this
Speaker 1 (27:30):
Point. Oh God, yeah. Okay. So what do you call it?
Speaker 2 (27:34):
I mean, I would just call it your incremental roas. We're brilliant at shortening it and sometimes that's all it takes, right?
Speaker 1 (27:43):
ECommerce acronym, you can call it mental roas. Maybe use the other part.
Speaker 2 (27:49):
I mean the thing people are always asking, how do I grow my channel? And I think what a lot of a brands have experienced over the course of the last three or four years is working with in-house teams or agencies or Amazon themselves, where the answer was spend more money to grow your market share, and they didn't. And so now a lot of 'em have, they're wary of investing in the channel because they've been burned, we doubled their spend and our sales didn't grow. And I'm like, that's where I start to salivate. I know that the problem is a media allocation problem, a creative problem, an assortment problem, not a budgetary problem. So if you gave me the same budget, I'm certain that I could deliver better results probably because someone felt handcuffed or hampered by what their performance had been in the past. They weren't focused on incremental roas, they were focused on some other metric. So it's a really important concept for anyone to keep in mind because retail marketplaces are growing and continue to grow, and they continue to have a higher share of ad spend for any brand. I think they're expected to grow like 10 to 14% this year compared to Facebook and Google, which are likely to plateau or even shrink as a percentage of overall ad spend. So in that environment, it's really important to know how to grow with your dollars.
Speaker 1 (29:11):
So to sum up, if you're having trouble growing on retail marketplaces, you have to call Ross. Is that your,
Speaker 2 (29:18):
I should tattoo that on my forehead.
Speaker 1 (29:22):
I look forward to seeing that shortly. I love it on LinkedIn. So I don't know if you've heard of this newfangled thing called ai, and I'm just wondering, do you have any prognostication about where the capabilities of AMC are going to go over the next year? Say what people can look forward to? Yeah, any inside scoop.
Speaker 2 (29:49):
There's a couple of interesting things when it comes to AI in the world of AMC. So initially, I think the previous iteration of, I think it was GTA three, not GTA four, which is essentially the version of AI that's available through chatbot GPT, so that the previous iteration could not understand the SQL code used, used by Amazon to write or read queries. What I had read is that the newer version was better at it, which is to say you could ask chat bot GPT to write you a query. Now that would actually work because it understood the code better. And now I'm not, internal testing has not proven that to be the case just yet. But what I imagine becoming the future is either Amazon incorporates an AI function into AMC so that someone in natural language can write a question and the AI can interpret it and spit out a query that will give you the answer. Amazon has just invested a lot in a new ai, not a new AI startup, but like another AI company. And I imagine they'll be looking for use cases to put it into all of their products from seller central and vendor central to the ad console to DSP and AMC as well, just to, again, as you say, remove human touchpoints wherever possible.
Speaker 1 (31:16):
So if you think of our listeners who are, well, they're probably pretty far into their 2024 planning and budgeting and all of that. Thinking about driving top line growth and incorporating AMC into the mix, do you have advice that you would have someone keep in mind as they're figuring out the next year's strategy?
Speaker 2 (31:44):
Yeah, so I think you definitely want to have the capabilities to understand the unique data that comes out of AMC for what you're already doing. So you need to have a set of monthly queries set up to track your neuter brand for your campaign. So you can have that insight when you're analyzing performance. You need to have a query set up that will give you your top products for neuter, brand, customer generation, so you know where you can place Betts to potentially grow into new markets or new niches that you're not dominating in already in the future. You need to have a query set up that will tell you whether your customer path to purchases. So if it makes sense to invest more in sponsored brands or DSP or awareness tactics. So you've got to have those baseline things in play so that you can make, I think those are table stakes now for Amazon. So if you're not doing those things, it's likely you're leaving some either low hanging fruit on the table or some great opportunities on the table, because those are not hard insights to get at. They're not hard insights to understand and they should just be part of your regular monthly review of your account. So I would say you've got to have a partner or someone in-House who can do that to make 2024 a great year.
Speaker 1 (33:01):
That is great advice. And there's so much firstname.lastname@example.org. I know if you go over to the site, retail media marketplaces and then look for Amazon Marketing Cloud, there's more there. We'll put a direct link in the show notes. And Ross, thank you so much for bringing the AMC info and certainly mental ROAS is right in my brain now.
Speaker 2 (33:27):
Speaker 3 (33:28):
Pleasure. New trend. There you go. Thanks
Speaker 1 (33:31):
For having me. Appreciate it. Yeah, no, of course. Happy to.
Speaker 3 (33:33):
Speaker 1 (33:34):
Thanks again to Ross for all his deep data on AMC. There's more where that came from. So swing on over to digital shelf institute.org and become a member to stay on top of it all. Thanks for being part of our community.