Season 2 | Episode 3

Dynamic and Personalized Creative - Changing the Landscape of Paid Media

As consumers and marketers, we’ve all been frustrated when we are served generic or irrelevant ads on the Internet.

Join us in Episode 3 of Season 2 of Paid Media Coffee, where we discuss many of the platforms now available for dynamic and highly personalized creative content.

Host Kelly Mancuso along with Casey Baker, Nick Toomey, and Alexa Munoz, tell you why dynamic content is a strategic tactic you need to leverage. Get a cup of your favorite hot beverage and follow along to learn more.

Kelly: Welcome to Paid Media Coffee. I'm Kelly Mancuso, and today we're talking about dynamic creative. My guests for today are Casey Baker, a paid media associate at Nebo. Welcome back to the show.

Casey: I'm excited to be back.

Kelly: Also joining us again is Nick Toomey, a paid media associate at Nebo.

Nick: Happy to be here.

Kelly: And finally we have Alexa Muñoz. She's a senior media buyer at Nebo.

Alexa: I'm excited. Thanks for having me.

Kelly: Welcome. Thanks for all being here. As I mentioned, we're talking about dynamic, creative, and. I think all of us know study after study shows that users want more personalized ads. It's something that we've also talked about on the show and you know, in our day to day lives, it's super annoying when we see ads that are irrelevant to us, or, you know, ads that don't fit where we are in life, or, you know, in our quote-unquote buyer journey.

But the good news is that there are so many new technologies and opportunities for a more personalized approach to creative. That's what I'd like to talk about today. Just to start off, Casey, will you help give us a quick overview or definition of what dynamic creative is.

Casey: In simple terms, it's basically just personalized content.

So, showing different creative to different people. For example, using dynamic creative, someone's searching on Target's website, they might be later shown a generic retargeting ad, whereas using dynamic creative would allow you to retarget someone with content that's more personalized to them. An ad for a pair of shoes that they had previously put in their cart, or a  pair of pants that they were searching on the site for.

Kelly: Awesome. Thank you. And we're really seeing this capability emerge or you know, be a possibility across a lot of different channels. Where are you guys seeing new technologies emerge or platforms evolve in order to allow advertisers to incorporate dynamic or personalized creative into their strategy?

Nick: I’ll start here. Honestly, I feel like almost every platform and channel is having to go this route. Creating those personalized one-to-one ads is almost key in order to keep up with today's digital society.  One technology that I know that we're looking into for a lot of our clients is a CMP or creative management platform.

Basically just a way to house all the creatives that you're making and be able to serve them again on that one-to-one basis, making it very personalized. So, for example, you're able to serve ads based on temperature or location, any audiences that you create. One of our clients we're working with has over 800 communities that we want to create ads for specific for each of those communities.

Instead of using all this manpower and resources in order to create those ads, we're able to use that CMP to automate it, right? We can create ads at scale. And then serve those across any platforms that we really want to serve it on.

Alexa: And then sometimes the CMP may not be financially in grasp at the moment for some clients.

There's a couple of other options. There's a lot of display partners right now that will almost offer it as added value. SteelHouse is one of them. One of the bigger players in the game does this. And you can do feed based or pixel-based SteelHouse specifically does pixel-based. I prefer pixel-based.

The difference is feed. It's usually an updated spreadsheet that is pulling from, that's housing. All the available products on your site, pixel-based, it's going to scan the whole entire website and automatically update. It's a little bit more reliable.

Nick: Feed based is more manual, basically. Okay.

Casey: And another big platform that's recently moved to that is, Facebook and Facebook's dynamic creative tool. You essentially just create an ad template and it automatically uses image and ad copy and generates headlines for you.

Kelly: And Casey, you talking about Facebook's response to bad's also, brings to mind Google's responsive ad formats as well, even in search.

This is something that we're able to, you know, upload and develop different headlines and description lines. It's not just something that we're seeing from an image-based creative perspective but also in search ads as well.

Nick: And what's nice about that too is you're able to test different creatives really quickly, right?

It allows the platforms to optimize. I know Facebook, for example, we run a couple of campaigns using this dynamic creative tool. And it allows Facebook the opportunity to optimize, right? The best image, the best headlines, the best, you know, call to actions, things like that. It kind of takes a little bit off of our plate, but you know, does it in a very nice way.

Casey: And I know we'll jump into some of the challenges, some more of the challenges later. but I think, I definitely think that if the clients are willing to kind of give up some of the reins of that control, it's definitely where the industry is headed.

Kelly: For sure. I think when this type of capability first started emerging and becoming popular, everyone was a little bit hesitant to move towards it.

It was like, okay, well we want to know what's happening and who's seeing what. But at the end of the day, data is showing that it's driving better results.

Alexa: And maybe you're newer to dynamic creative and you don't want to give up all the reigns at once. You can allow the whole entire template to be dynamic.

But say you just aren't ready to trust just yet for the whole entire ad, you can just pick the image, you can pick the product, you can just make the text dynamic. You can just make the CTA dynamic. It's up to you to kind of mix and match and test.

Kelly: That's a good transition. Let's go into specific strategies and ways people are using this.

What are some of the interesting ways that advertisers are applying dynamic creative strategies?

Alexa: I thought this was pretty interesting. I think AI is going to be really prevalent this year. Watson advertising, they rolled out, dynamic creative optimization utilizing their Watson AI. What was really cool about that is that before they even went live, it could understand and predict which combinations would perform best.

They didn't even have to waste any ad dollars trying to figure out what mix and matching would perform best for which audiences.

Nick: What Watson AI is able to do. Well, if we don't have Watson AI, what we're having to do is work on sequential messaging. And figuring out what that messaging looks like throughout the funnel.

Whether it's the first time that we hit them or the second time or the third time, and we're able to hit them with different messaging. And that's something that I think is a really good strategy within dynamic creative is being able to change that messaging throughout the funnel for users.

Alexa: Definitely. There are so many different ways to use dynamic creative. I mean, the main one's going to be product retargeting just because. That is exactly what you were looking for. You don't really have to guess, but there's plenty of other opportunities out there. You can give location, behavior, device, demographic, the context on the site they're looking at. Hotel and airline industry - they can do it based on where you're living. Like if you've shown interest in traveling to other places, hotels are the same way. I mean, really the possibilities are limitless.

Casey: You brought up location, and that's one thing that I definitely utilize for one of my clients.

They have several locations across the country, and being able to localize that content and show people ads that are in that specific area is something that has been huge for them.

Kelly: It kind of allows you to target a broader audience and have to manage less and then use that dynamic capability, whether that's, you know, an ad customizer in search or leveraging a, you know, a display partner or CMP to tailor that creative specifically, and kind of manage things at scale a lot easier.

Something going back to Facebook that I just learned today, and I don't know if it's new or if I just missed it completely, but they are offering dynamic language optimization capabilities. You can create essentially one ad in your native language and then target it to people across different languages and Facebook will automatically translate the ad.

I don't know how good it is or how accurate it is, but I’m really interested to look into that and see if anybody has any case studies. If anybody listening has any experience doing that, please email us and let us know or definitely interested to hear.

Alexa: That's interesting. I think that's actually one of the times where I would be not trusting to run dynamic.

Casey: Because what if there's one word that's awkward.

Alexa: Google translate is never exactly how it's used. I don't know, something like that actually would scare me. I'd want to see a case study.

Kelly: Just to take an ad and run it through Google translate, that would be terrible.

But they also have, you know, asset customization for placements too. If you have one variation of an image, you can upload it and then their technology will, you know, adjust the sizing and the ratios and whatnot to make it eligible to show in different placements as much as possible. They're doing a lot to enable advertisers to personalize their creative based on, you know, the user and also the environment with which they're viewing ads.

What about some best practices for advertisers? What do we need to keep in mind when leveraging dynamic creative?

Casey: I think one of the big things is to continuously be monitoring and testing your ads. I think that a lot of times people have this idea that once everything's set up, they just let it go and let it run, but you definitely need to be cautious of this and make sure that you're constantly monitoring and optimizing. Making sure that you're not, kind of blowing through that budget faster than you'd like to.

Alexa: Speaking of optimizing, I think it's always also good to go beyond just last product viewed if you want to mix it up or also put in some of your best sellers.

There's ways where it can pull in like, similar shoes, similar clothes, whatever the products you viewed. You don't always just have to do the last product viewed.

Kelly: There are ways to mix it up or complementary products to upsell. Right?

Nick: I mean, creative is fantastic and it's really important part of the media mix, but still the audience is key, right?

The whole point of dynamic creative is making sure that you are giving the best creative to the user that you want to hit. If you don't know who that user is or you don't know where they are in the funnel, it almost kind of defeats the whole purpose of making the dynamic creative so specific and different.

Kelly: I think that's where data comes in. Data is. What we have to rely on to be able to make these strategies work.

Alexa: That's probably the main challenge, especially if you're relying...I mean, you definitely want to rely heavily on your first-party data. You just have to make sure it's clean and correct because you don't, it's almost the opposite.

Like it's good, whenever I see an ad that speaks to me and I resonate with the brand. They can also make me dislike a brand if I see an ad that I can tell is totally not geared toward me, it makes them feel like they're, they don't know their audience at all. I think making sure that your CRM data, your first-party data is all clean and accurate.

You're talking to the right audience with your creatives.

Kelly: For sure.

Casey: And I would also say making sure you keep that personalized messaging from the ad over and make sure it translates over to the landing page because if those messages aren't consistent or your, or look totally different, then that's also going to be something that is a turn off to people.

Kelly: That's a really good point. I read a resource from IAB that was actually focused on data-driven video, but I think it's pretty applicable to any type of creative that you're running. And I really liked it. They were talking about when you're approaching this kind of strategy, start by determining what kind of data that you have and organizing that data.

Maybe it's a product feed, maybe it's an API connection to some kind of third party source, like a weather API. Maybe it's your own first-party data. Maybe it's multiple things, but figure that out. Organize it. Then think about the channel that you want to execute this through. You know, where your ad is going to be seen and what are the user behaviors associated with being on that channel, because somebody on a mobile device, it's going to be a lot different than somebody viewing a TV ad or an OTT ad, versus someone on social. And then once you have that laid out, then you can start thinking about the creative execution. What is your messaging or your actual imagery look like?

And that can kind of help you start at square one because it can seem really daunting when you try to think about this and you're thinking about 800 communities. For example, going back to your example, Nick.

Alexa: I think another challenge that can sometimes arise is going past personalization. I think there's a fine line of being creepy, which you have to walk that line because you want to speak to your audience and you want to talk to them about what they're looking for, what their behaviors and contextual data are.

But at the same time, It would freak me out if I saw it. Now that's like, it knows where I live, and how many dogs I have, how long have been married. They might know all that stuff, but I don't want them necessarily blasted where I see it.

Nick: Especially now with all the privacy acts coming out. It's a very touchy subject.

Absolutely. So I'm with you. If I got an ad, a display ad that said, Hey, we know you just saw our CTV ad and be like, Whoa, I do not like you as a brand.

Alexa: It's like we are, you know, you just saw our CTV ad, right? But let's pretend like we didn't know that. And it's just good timing.

Kelly: You want to balance relevance and personalization with, avoiding somebody seeing the ad and being like, Oh Whoa, they know way too much about me, I'm going to opt-out immediately. And then we don't have their data at all.

Casey: Exactly. Like that also brings up another point with if you have too many different variations and test too many different things and get too personal to somebody, I think that that also could be an issue with too many different creatives and getting actual insights.

Cause if you have all this data, sometimes you have more data than you can actually kind of digest.

Kelly: I think starting with quality over quantity is good advice for people because I mean, you can run hundreds of different ad variations and test all of this stuff, but as I mentioned, it can be really overwhelming.

Start smaller with strong options and then kind of scale from there when you feel more comfortable with it. Definitely.

Alexa: I think. Some best practices are, start with a segment. Don't have to go all-in at once, get some learnings, and also possibly start with a partner that has these tools on hand that doesn't have to pay extra for right off the bat.

If you see those performing well for those segments, with that partner, and you want to invest heavily, then from there you can go on an investment personalization tool.

 Kelly: What about some other challenges? I know we talked about a few of these already, but is there anything else that, we're going to need to overcome or just at least think through when diving into personalized and dynamic creative?

Nick: One thing is that CMPs or creative management platforms are pretty expensive in order to run. So, like Alexa was just saying, start small, start in a segment. Don't jump right into a CMP cause it might be a little more expensive and a little too robust in terms of what you're trying to do.

Whereas if you take it slow, take it segment by segment, that'll kind of reduce that challenge at least for a little bit.

Kelly: And from our experience, the pricing structures are also a little bit confusing. 

Casey: Something I've seen with some of my clients, is just giving up that control. I think they're kind of hesitant to hand over the reins. At first, it can be nerve-wracking if they don't know that they're approving every single variation of the different ads that are going out. But I think kind of if we can get over that hump, it's definitely, something that will pay off in the long run.

Kelly: And speaking about control. I think another challenge is figuring out who manages this stuff, especially with a CMP.  Creating the decision tree, uploading all of the assets, organizing all of those things. There are a lot of processes involved and you know, does that belong to the creative team?

Does that belong to the media buyers? Where is that line drawn? Trying to figure that out. You definitely need to have buy-in across all departments and figure out what the process is internally for managing it.

Nick: Absolutely. And then even adding onto that, right is getting all partners and vendors able to integrate with, if you, for example, want to use a CMP or using a data management platform and things like that.

I know, Alexa, you have had experience in this before, but I don't know if you want to expand on that a little bit more and making sure that data's all in the right spot and everyone can tap into it.

Alexa: There are a lot of moving pieces that you have to make sure in the right spot before moving forward. One of them is your data, making sure your data is correct and that it will ingest into the CMP you pick. That's a big one. And also whatever CMP you work with can push out to whatever ad server you're using. I mean, everyone's different in their needs, especially brands and marketers. I think finding a CMP that makes the most sense for what you're trying to achieve.

Kelly: And even using similar data from different data providers can be a challenge. We were running a campaign recently where we're targeting people based on household income and dynamically changing the creative based on their household income brackets, and we were doing that within Google, but also with a programmatic partner who is handling the dynamic aspect.

And then through Google, we had a little bit more manual retargeting segments set up. But you know, something that we thought about beforehand was, wait, what if Google's household income classifications are different than what the programmatic partners are? We had to get on the phone with them and, you know, figure out where all of their data is coming from.

And luckily, the programmatic partner was able to say that they can tap into Google's data, first and foremost, so that, you know, we weren't serving different ads to people based on how they were bucketed.

Casey: That's something I've noticed with some of like vendors that we'll ask them if they can ingest certain types of data or do certain types of things and they'll say, we haven't, but we can. And I'm like, well, how are you going to do that?

Kelly: Definitely. I mean, I feel like the sky's the limit with the. Options out there, but you don't want to be the first one to try something. Writing with a, with a partner or even, you know, internally with the client probably doesn't want that either.

Casey: No, they were, they definitely have hesitations with that.

Kelly: All right. Awesome. Well, I think we've covered a lot of really good stuff here, but just to wrap up, what final thoughts or recommendations do you guys have for advertisers who are working or preparing to work with dynamic creative?

Alexa: I think a really good call out that people don't talk about often is implementing a burn pixel.

Most people don't utilize this, but I think it really can make or break your dynamic retargeting ads because there's nothing more annoying than say, buying a pair of shoes. And seeing that same ad for the same pair of shoes 30 days later, like, yeah, I already bought these. Shouldn't you know? Like it's, you're ad and your website.

Kelly: Or seeing something on sale after you've already bought it.

Alexa: It's definitely infuriating. 

Kelly: I get so mad. It's digital advertising 101, right? But so many people don't do it. Drives me nuts.

Alexa: Implementing a burn pixel basically just makes sure that once you've purchased something, it will not serve you that for that product anymore.

Nick: Kelly, like you were saying earlier, well two things, like you were saying, quality over quantity and the sky's the limit kind of meeting Pat in the middle. basically take your time to test and make sure that you're doing a good test, right? Make sure it's the right fit. Like you said, the sky is the limit.

There's so much you can do with this, but just like any creative testing is going to take time, it's going to be a trial and error, right? So, start with a segment, like Alexa said. Start with an audience that, you know, it's gonna resonate with and build from there. You know, don't make it too complicated right off the bat, cause then there are so many moving pieces you don't even know what you're testing anymore. 

Alexa: When you're doing multiple tests at once, the readouts are never conclusive. 

Casey: And I would say if you're talking about onboarding a CMP or a creative management platform specifically, do your research and just make sure you truly understand the different capabilities of these partners.

Kelly: Awesome. Well, great advice. Thank you all so much for joining me. Our next episode is going to be a deep dive into connected TV. Please tune in to that, and if you as a listener, have any questions at all, or if there's anything specific that you'd like us to talk about in one of our episodes, please email us at

Follow us on Twitter @paidmediacoffee. Then also like subscribe rate, review the podcast and we'll see you next time. Thanks.

Posted by Paid Media Coffee on February 4, 2020


Add A Comment
Posted by
Paid Media Coffee