In this episode of Revenue Boost: A Marketing Podcast, titled From Amazon to Instacart: Optimizing Multi-Network Retail Media Programs, host Kerry Curran sits down with retail media expert Misha Cohn to explore the complex world of multi-network retail media. As brands increasingly shift budgets from traditional platforms like Google and Meta to retail media networks (RMNs), managing campaigns across platforms like Amazon, Walmart, and Instacart has become both a challenge and an opportunity. Misha shares his insights from years of working with Fortune 1000 brands, discussing everything from the evolution of RMNs and the consolidation of ad tech to the strategic use of data for optimizing ad spend and improving market share. Tune in to learn actionable strategies for navigating multiple networks, enhancing bid efficiency, and harnessing AI-driven tools to boost performance and streamline processes.

Podcast transcript

 Misha Cohn (00:01.25)

Thanks for having me on the call today, Kerry. I really appreciate it. My name is Misha Cohn. I've been working in the industry for the past eight to 10 years. I currently work for Skai, working with Fortune 100 brands in a sales capacity, learning a lot about their goals, what they're working on, and how we can help. In the past, I'd also worked a little bit at Flywheel Digital, now part of Omnicom, as well as packview, another ad tech company for a few years.

Kerry Curran, RBMA (00:34.21)

Excellent. Thank you, Misha. Yeah. So Misha and I met many, many years ago through Packview. And I know Misha, you and I have talked about and seen since we've known each other too, just the rapid evolution of the e-commerce space, the evolution of retail media and all of the retailers having their own media networks. What, like talk a bit about what kind of what you've been witnessing first hand over the last eight to 10 years.

Misha Cohn (01:04.406)

Yeah, it's been very interesting. would say years ago, the focus was really Amazon, some Walmart, Instacart got started. Now there are so many retail media networks. There's also Citrusad, which is now called Epsilon, Criteo, Kroger, Chewy, Sam's Club, eBay, and the list goes on. And so that's where I've seen more of a consolidation, where a lot of brands are looking to consolidate different ad technology they're using, different agencies, so they're not all using one agency for Amazon, one for Walmart, one for Target. And then even if they can consolidate globally, keeping that consistency, that similar way of reporting, keeping it easy for CMOs, CEOs, VPs of e-commerce to analyze their business.

Kerry Curran, RBMA (01:55.859)

I mean, there's so much. Like you said, back when I was at Group and you were with Packview, it wasn't even that long ago we had a panel at Advertising Week.

I think it was 2019 and it was the only panel at Advertising Week on even e-comm or Amazon. And now you look at Advertising Week, look at Can, everything's retail media network. It's been, you know, we both kind of had a front row seat to that evolution, which has been really fascinating, but it creates a lot of challenges for brands. Like you were saying, like it's now multiple networks, it's product feeds to multiple networks, it's budgets and, you know, bid strategies for different networks and which means more people, more resources, more analytics and data. So it's been a lot. And so when you're talking to brands today, I know you've, you know, have taken a very consultative approach to kind of how you're helping them.

When you're talking, first talk a bit about the brands that you are supporting and helping in your current role. like when they're, you're taking your first calls with them, what are the challenges? Like why, why are they reaching out to you?

Misha Cohn (03:25.9)

Sure, yes, I work with a lot of industry leaders and whether it's in food and beverage, toys, apparel, consumer electronics. And what I've noticed is although every company is different and they may have certain different things, they need to connect their data to maybe Skai's data in a different way. A lot of times there are similar trends where one of the big ones is how can the company better connect the dots? So whether that's between marketing teams and sales teams, if that's connecting the dots between their internal systems and then bringing in profitability information or inventory. What I think has been helpful, like from that Skai perspective, Skai has that open architecture. So it can work both ways. If you want to bring in Nielsen data into Skai, see your results. If you want to bring Skai's data into your different systems, you can do that. That's been very helpful.

Kerry Curran, RBMA (03:58.155)

Yeah. Mm-mm.

Misha Cohn (04:25.282)

Also connecting the dots to and thinking about what is that point of diminishing returns and what sales are truly incremental that that brand would not have had if they hadn't been doing advertising. And with that retail media advertising growing at 17% year over year versus Google only growing 3% year over year, there is such an emphasis. 

Kerry Curran, RBMA (04:34.208)

Hm-mm.

Misha Cohn (04:50.412)

Now companies are truly understanding the importance of retail media and even shifting some of that budget from Google or social media to that retail media budget.

Yeah. So are you seeing trends in that shift coming from like Meta or Google to the retail media networks?

I have been seeing that, yes.

Kerry Curran, RBMA (05:10.41)

Yeah, it's really interesting. So you're saying a lot of the times top brands are trying to get a better kind of view of the competitive landscape and market share. Are you seeing, how are you seeing the trends in the brands that are investing with that cross media strategy?

Misha Cohn (05:30.914)

Sure, yeah, I've seen a lot of brands working with companies that can get them that data. So such as with Profitero, they can be able to show you what is that estimated sales data, what is that market share data for those competitors. I've seen a lot more differing trends where different strategies used to be more siloed versus even focusing on that competitor advantage. Now I've seen those be a lot more connected. So as an example with … 

Kerry Curran, RBMA (05:57.43)

Mm-hmm.

Misha Cohn (05:59.29)

… certain large clients that are using both Skai as well as Profitero. Now what those clients are able to do is they can actually optimize, increase or decrease their bids or their ad budgets based on competitor pricing, competitor availability, competitor buy box data. So it's taking that to a whole new level and making sure that that advertising spend is even more efficient to drive the most sales possible.

Kerry Curran, RBMA (06:26.004)

Yeah, so it's, are you seeing, it's definitely impactful on the business results when they are taking that, using that data and incorporating it into their strategy. Yeah. And so how are brands kind of tackling multiple networks? Cause from my perspective, I just keep thinking about, all right, it was, you know, we had Amazon and you had ...

Misha Cohn (06:37.695)

Exactly.

Kerry Curran, RBMA (06:51.286)

… Google, Amazon, Meta, and then you had, round down and then you had Walmart. And now it seems every big retailer has their own retail media network and it just seems like a lot. So talk about like, how, are you helping or how are you seeing brands tackle that?

Misha Cohn (07:09.046)

Yeah, I think this space has been really interesting where it's kind of similar if you think about linear TV, how that's more connected TV. Where nowadays, you know, people might have a YouTube TV, they have Netflix, have Hulu, they have Peacock. There's so many options. And what a lot of these enterprise brands are looking for is a one unified, one stop shop dashboard …

Kerry Curran, RBMA (07:34.516)

Mm-mm.

Misha Cohn (07:34.536)

… a marketing suite where they can have all the data in one place and can see trends. And so in that case, if they did want to have that budget fluidity and to take funds from, let's say, if it's an Amazon to an Instacart or an Instacart to Epsilon based on performance, they can do so. And then there's also certain companies like Skai that also have places where you can do different scenario planning. So if you wanted to test out … 

Kerry Curran, RBMA (07:43.884)

Okay. Hm-mm.

Misha Cohn (08:01.474)

… okay, I want to see if I increase my budget by this amount. Then what is that impact from both an ROI and a sales standpoint? And so that way, what you can be able to do is actually bring in your last two years of historical data. And then you can plan out more than a quarter in advance, being able to see a very accurate view, even with a 90 % confidence interval of how you're going to perform. And so I think the more you can do that,incorporate brands as their own forecasts as well, test to see which is more accurate or a combination of both working together and getting very integrated with those different brands too.

Kerry Curran, RBMA (08:42.752)

Yeah. So no, it's, there's definitely a lot. It's a lot of data and it sounds like they, the ability to collect that data from multiple sources and forecast, think is helping marketers get smarter for sure. Are there other kinds of aspects or business challenges that brands have been talking about or that you're hearing or seeing?

Misha Cohn (09:06.54)

Yeah, I would say a couple of them. One would be an example is that it used to be that people were optimizing towards ROAs. Now people are trying to look more into incrementality. So incremental sales and incremental ROAs. With that, there can be some challenges. I know from what I hear for certain companies, they might have a certain solution, but man, it might be more of an inverse share voice times a ROAs number.

Kerry Curran, RBMA (09:21.302)

Mm-hmm.

Misha Cohn (09:34.828)

So it might be a good fit for three out of four companies, but then one out of four of those biggest brands are like, eh, this isn't really taking my performance all the way that it needs to be. And so I think if you are able to look at more of that performance, getting very nitty gritty on a campaign and keyword level results or even product level incrementality. And then I know with some brands, what they are working on right now is more of a custom solution is that … 

Kerry Curran, RBMA (09:45.292)

Mm-mm.

Misha Cohn (10:03.178)

… what you're able to do is even optimize more towards an eye profit versus instead of that whole eye ROAs. And I think with that, so a lot of brands may not feel fully comfortable giving all their certain profitability data to those retail partners they're working with because they could create their own private label, do it a little bit cheaper, may not have to pay all those advertising costs versus using a third party like a Skai. If you're able to integrate … 

Kerry Curran, RBMA (10:08.258)

Okay. Mm-hmm.

Misha Cohn (10:31.424)

… that data, then that is something that you're able to more optimize towards. And whether you have that data more already in your networks and then piping it into the Skai or even through a Profitero and then having those data pipes come into the Skai. And then you can see your performance. You can optimize towards it. And I am also thinking about, you know, what's next is thinking about how brands optimize towards market share versus just share voice and I think that's something I've heard of that hasn't really existed in the market. I do see that as a possibility for different data.


So with us again, as a partner of Profitero access to that sales and share data that I do see that as a possibility and be able to optimize towards that. And that would help a lot with brands. So it's not just sharing a voice where maybe you have 200 or 300 keywords. You're only getting a small view of that performance versus if you're looking at market share as a whole, that's going to be a lot more accurate versus that share voice data.

Kerry Curran, RBMA (11:35.211)

No, that's actually really interesting. I think to your point, an important KPI to have as part of the campaign. It'll be interesting to see too if brands, as brands shift to that a bit, does that, how does market share tie back to revenue? And you mentioned before the point of diminishing returns. So the combination of optimizing to greater market share, does that and the profitability, does that continue to drive revenue and get more profitable or is there a point where kind of you're saturated, you're good and it's kind of to continue that positioning.

Misha Cohn (12:23.06)

I think that's a good point. I think with certain brands, they're more focused just on the revenue. But I think with a lot of the industry leaders, they might have more of that long-term approach where they're thinking, okay, even if I am taking a little bit of a hit from a revenue, I'm not just doing retargeting or targeting my same keywords of my own branded terms to get results. If I'm able to capture that market share from competitors, I will win in the long-term.

Kerry Curran, RBMA (12:33.334)

Mm-hmm.

Misha Cohn (12:50.038)

My brand will have that great reputation and continue bringing on newer and new to brand clients.

Kerry Curran, RBMA (12:56.471)

Yeah, no, and to your point, I think that's not going to work for every brand at every stage of maturity. But I think you're spot on with that being, that's an additional value that retail media investment will bring if it's not just focused on sales and direct response and moving more up the funnel. Because that's where when I hear brands getting obsessed about incrementality, that's where I kind of get nervous for them because to your point, it's that market share that's going to help the longer term strategy. So brand growth and the conquesting the competition. So yeah. And so I love that, you know, you and your teams are looking at that holistically and working with brands are looking at that holistically because it needs to be a mix. Do you think that those trends will continue?

Misha Cohn (13:56.99)

I think that'll definitely continue. And I think another one too, is a focus on different AI where in the past, it might've been more of a focus with automated actions or optimization rules. And while I still think that is a great solution for now, I think the end goal is how can you make an AI as accurate as possible? And I think it's still going to take time. Like if you think of, let's say chat GPT where … 

Kerry Curran, RBMA (14:22.166)

Mm-mm.

Misha Cohn (14:25.758)

… it can be pretty accurate in certain easier tasks. It can make it even easier. Like if you are an assistant principal and you need to write a welcome letter to freshmen students coming into the school, they might write you a really, really great email you can send out right away. I think sometimes if you think about it, if you're like, hey, I want to write a letter, an email to a prospective client and here's the industry I'm in that doesn't have all that data, it might type in inaccurate things. And I think … 

Kerry Curran, RBMA (14:42.403)

Mm-mm.

Misha Cohn (14:54.934)

… you know, over these next few years that's what we're going to see is AI take off. Now certain companies have a focus on it. Other ones I might've worked for in the past, not as much where it's more on that automated actions type of capabilities. But I do see that it is powerful, I think in some ways, especially more medium sized versus enterprise brands. They are having a lot of success with even bidding AI, seeing increased sales, increased ROAs, and increased conversion rates. And then also from a gen creative AI capabilities, that's becoming more popular now where different companies are thinking, okay, instead of me doing a photoshoot, I'm having to spend hours and many dollars to create all these different photos. If I'm able to create 10 good ones, can I create another hundred that look similar but different to that photo?

Or even working as part of Amazon's creative beta before where it can generate some things. But again, those may or may not be perfect where I've known some leading CPG brands, they were working with Amazon. man, the font of our name on this product in that creative way was way too small. They have to give that feedback back to Amazon. So it's not a perfect system, but at the same time, it's getting closer. And even with certain things like for Skai, …

Kerry Curran, RBMA (16:15.287)

Mm-mm.

Misha Cohn (16:17.794)

Celeste AI can give a lot of instructions both for easy things like how do I set up day parting or intraday bidding? And then also what are my top 10 performing campaigns? What are my top and bottom performing keywords? How do I increase performance and give those steps? But I think the next step for all companies that I haven't really seen done yet is taking it a really even a step deeper, not just here's 10 steps to anyone can use to increase sales, but here are specific for you how to increase those sales from what you've been noticing as part of that too.

Kerry Curran, RBMA (16:52.704)

Yeah, no, I see that being a huge benefit. The AI for creativity is something we're definitely going to see more and more. And I love the consideration of once you have all of this data, getting those customized recommendations, both on the bid management, but also on the creative strategies. So lots, lots to come from the AI impact, for sure. And so.

I know you work, you said you work with a lot of different challenger brands and the larger fortune 500, 100 brands. What do you think? Do you have recommendations for a brand that maybe hasn't worked with a bid management or retail media management platform? Like Skai, what would be your recommendation for one, determining if they are mature enough for partnering, but to like how to get started?

Misha Cohn (17:50.178)

Sure. Yeah, I think that for a company that's very small, maybe they're spending less than $15,000 or $20,000 a month in ad spend, you can probably do that a lot more manually. You're not going to have as many keywords or products. But after that standpoint, and especially after like $50,000 in ad spend, I do think it's really important to use some level of an ad tech company because you're going to get a lot … 

Kerry Curran, RBMA (18:02.591)

Mm-mm.

Misha Cohn (18:16.502)

… better results, you're going to be able to do keyword research besides just Amazon's auto campaigns. Like in Skai, what you can do is you can see what are the top performing keywords on even a Google on a Meta, you know, Instagram, Facebook, Tiktok that then you can utilize in your campaigns. So if you think about it, like with a top coconut water company I've worked with, they noticed a trend on Tiktok and YouTube for natural cereal.

So then it was saying that instead of milk, people were using their coconut water, is that milk? And then instead of cereal, people were using those different berries. And so then when applying that certain term then into more of their campaigns on a retail media side that helped them boost those different sales, conversions and more. And so that's where if people can incorporate those types of trends, I think that can be even more powerful versus a lot of folks maybe looking at things a little bit more siloed in that case.

Kerry Curran, RBMA (19:20.726)

That's a terrific example. And I love the incorporation of social and TikTok trends into your retail media strategies. So lots of benefits and value when the brands have the budget and the maturity. So this is great. Thank you, Misha. Thank you so much for joining us today. This has been awesome and always good to catch up with you to get more insight on the rapid evolution of the e-comm retail media space. So thank you for joining us today.

Misha Cohn (19:52.7)

Thanks so much for your time. I really appreciate it.