← Next in Media
Next in Media
How Georgia Pacific Modernized Its Marketing Mix Modeling
00:22:16
Georgia Pacific’s Javier Bustillos breaks down the hard realities of modern media, revealing how the consumer goods giant tackles walled-garden measurement gaps, modernizes in-house Marketing Mix Modeling, and tests AI capabilities without sacrificing human oversight.
Loading summary
Transcript58 lines
- [00:00]
A
Yes, it is a challenge. I mean and we are doing the best that we can with the access to the data that we have. But as you know, I mean there are walled gardens that don't share the log level data. There are challenges on measurement, truly getting a cross channel measurement because of the walled garden not sharing that data. So yeah, we are doing the best that we can with the data that we have, but it's not perfect.
- [00:24]
B
We all know TV has become fragmented and we all know the feeling of spending more time deciding what to watch than actually watching or worse, knowing exactly what you want to watch and still not being able to find it. Vizio OS was built to solve this problem. Viewers spend 30 plus minutes a day on Vizio's home screen searching, discovering and getting inspiration to guide their entertainment and shopping journey. For advertisers, whether you're promoting a new show, restaurant or car or product, this is the best and only place to capture attention before they dive into an ad free environment. And analysts project Vizio will become the number one TV operating system by 2027, positioning it as the front door of TV. If you want to connect with consumers at the moment of discovery and let Vizio turn attention into action, reach out to Vizio adsizio.com this week on Next in Media I talked with Javier Bustillas. He's the Vice president of Integrated media and Brand analytics at Georgia Pacific where he oversees media spent for brands such as Bronnie and Angelsoft. We talked about how the brand that essentially wants to reach everyone employs data and targeting and some of the shortfalls in TV ad tech and measurement. Prestigious also talks about the importance of MMMs and why some media companies are just aren't wild about them and even gave a reality check on the process of in housing lots to get into. So let's get started. Hi everybody. Welcome to NEXT to Media. I am Mike Shields. My guest this week is Javier Bristillos. He is the Vice president of Integrated media and Brand analytics at Georgia Pacific. Hey Javier, thanks for being here.
- [01:39]
A
Thank you Mike for having me. Looking forward to a great conversation.
- [01:42]
B
Yeah, I'm excited to talk to you. You know I think your company is one of those huge brands that maybe not everybody knows. Tell people about what you work on at Georgia Pacific, your role and everything.
- [01:54]
A
Yeah, so at Georgia Pacific I'm responsible for integrated media planning strategy and activation of digital media in house as well as marketing analytics for all of the Georgia Pacific retail brands. So Georgia Pacific is a major manufacturer of paper products. We have retail brands in Toilet paper, paper napkins, paper plates and paper towels in the retail business.
- [02:17]
B
Okay, so in another time that was an easy job. Right. Just reach everybody. Just like everybody uses paper towels. Blasted out the message. Why didn't you integrate anything?
- [02:27]
A
Yeah, I mean things are changing. Things are changing. So I hope it was that simple. And I think, yes, there is still an element of, of trying to reach as many category buyers as we can, but doing it in a smart way and leveraging the. Yeah. Targeting capabilities that are available now in media channels.
- [02:45]
B
Well, so let's talk about that for your company. Obviously the TV business is changing rapidly. Everyone's trying to make it more data driven, figuring out how outcomes based buying works in this medium. What does that mean for you guys when you have broad brands that you don't always sell directly?
- [03:04]
A
Yeah, I think the important point, Mike, is to think about our holistic media plan. Right. So I prefer not to talk about kind of the targeting approach that we take on an individual tactic or channel. And it's more about the targeting approach that we use across our entire integrated plan. And we use a range of different type of audiences from broad audiences depending on the channel to drive that mass awareness and reach of our brands. But we also use kind of hyper targeting when we feel that it is effective and we use audiences somewhere in between a mix of purchase based audiences, contextual targeting, behavioral targeting. And then when we look at it holistically, we see great results because we are trying to maximize the reach among the category buyers that we want to target.
- [03:53]
B
What's in your. To do all those things you're describing? What's in your way? Like is it. Are you able to do the integration that you'd like is the challenge? I guess pulling everything together or being as precise as you'd like to be,
- [04:05]
A
I would say is putting them all together is probably kind of the biggest challenge. And making sure that any new audience that we add to our mix is driving incremental reach and not overlapping and just driving frequency with the same type of buyers.
- [04:19]
B
Is it hard to get that information? I mean, it sounds obvious, but that seems like a walled garden challenge. We've borrowed from the web to television, correct?
- [04:28]
A
Yes, it is a challenge. I mean, and we are doing the best that we can with the access to the data that we have. But as you know, I mean, there are walled gardens that don't share the log level data. There are challenges on measurement, truly getting a cross channel measurement because of the walled garden not sharing that data. So yeah, we are doing the best that we can with the data that we have, but it's not perfect.
- [04:52]
B
Yeah, we're right in the middle of the upfront season, I guess. How much does that change for you guys since you even you've been at the company for more than a decade, I imagine it's radically different than when you started. I hear there's almost like two upfronts. There's like the sports scarcity stuff and then everything else. Is it sort of what it looked like for you?
- [05:11]
A
Yes. So upfronts are still kind of an important part of our media planning process and the video investment. But yeah, it has changed. Right. So the video consumption has changed. There are kind of new players that are in the upfronts. YouTube being one. Right. Netflix now also participating in the upfront. So we have evolved our video mix based on media consumption trends, but it is still a significant portion of our, of our investment planning.
- [05:40]
B
Yeah. Can you, can you talk about. We talk a lot more about addressable tv. That used to mean something very specific, kind of out of the cable, two minutes an hour kind of world we used to live in. What does addressable TV mean to you right now?
- [05:54]
A
Yeah, I mean addressable TV at the end is, is about kind of serving ads to different households or different individuals based on their media consumption habits. So we have experimented with addressable tv. We have experimented with data driven linear. I would say our approach when we're using linear TV is more of broad targeting approach with the use of linear TV or cable. And we use other channels, connected TV, YouTube, online video, to reach kind of more targeted audiences with digital video rather than through linear tv.
- [06:29]
B
Do you worry about going like. Because TV is often. It's changing. People watch TV individually now more than I think they used to, but it's still communal a lot of the time. Do you worry about going too far to trying to target specific individuals?
- [06:43]
A
Yeah, again, our approach on linear TV is more like broad type of targeting. So that's the way that we have used it. We, as I said, we have experimented with a more targeted approach on linear tv, but we haven't seen the results that justify kind of paying the premium for addressable TV or for data driven linear. So we have taken a step back and using the channel more for kind of that mass awareness and mass reach.
- [07:04]
B
That makes sense. Okay, getting back to. You hinted at this. The players have changed in the upfronts for sure for a while there. Does YouTube belong in the conversation now? It's sort of like, do we put creators in the upfront conversation? Are they their own thing. You know, do they zep along in social video or are they big enough now that they should be thought of as another reach vehicle or as big as television properties?
- [07:30]
A
It's a great question, I think about creators are everywhere, right. So they are not only on YouTube or social media, they are expanding into connected TV. So we don't think about creators as a channel. We think about creators as a different type of content or ad format that can drive kind of a different type of engagement with the media. So we plan our kind of investments more at the channel and tactic level. And then we think what is the best type of content format or content type to deliver kind of the objective that we want? So we plan kind of to use creators in many different tactics, including social media, including YouTube, TikTok and other platforms.
- [08:15]
B
Yeah. So that it's really going to depend, I think probably on the tactic and the size of the career that we're talking about.
- [08:20]
A
That's correct, yes.
- [08:22]
B
Okay, I want to ask you about. There's been a lot of, I would say excitement or renewed enthusiasm for MMMs over the last couple years. Why is that? Why are they a bigger deal than ever and why, how has that changed things for you?
- [08:35]
A
Well, I think, I mean, we were talking about the challenges on, on truly having a cross channel measurement. Right. So. And in the absence of having kind of that unified measurement approach and a standardized measurement approach, then marketing mix modeling is still kind of a robust way of measuring different marketing tactics from national media to retail media to coupons and consumer promotions. So we do rely on MMM as probably that tool that give us kind of the broadest view of all of the marketing activity and the marketing investment that we're doing and the ability to compare between different tactics. Having said that, MMM is a statistical model. Right. So it's not perfect. So we don't use only mmm. We use MMM as one data point among other type of measurement that complements and give us kind of a full view of the marketing performance for the different investments that we're doing now.
- [09:29]
B
It used to be that you would use your MMM like once a year maybe. And now you hear they're more flexible and they're malleable. Do you use them all the time or is it something you're making decisions with on a regular basis or is it still sort of a macro level?
- [09:44]
A
No, we have evolved our approach to marketing mix modeling. We do it in house. So we manage our MMMs in house. When we transition MMM to do it kind of in house as part of My team, we had the ability to accelerate how quickly we can deliver the results. Okay, so before it was once a year, maybe six to eight months after the year ended. And now we're able to do it like three to four months after the year ended. We have also, in the last couple of years we have been able to do it more frequently. So now we're doing it twice a year. And we're moving in a direction where we feel that we can do it quarterly. We are incorporating more channels and more tactics that we can model in mmm and we're doing it more granular as well. So we have kind of data breakout by audience, by device, by publishers or partners, by ad format. So kind of all of those changes, more frequent, faster and more granular, allow us to make better and faster decisions using MMM data.
- [10:43]
B
Now, are there things that the. You hear different opinions from people, like certain buyers, like, oh, my client loves their MMM and they go. But you hear things like maybe they're not as good at measuring certain media, like retail media creators, things like that are not a lot of history or data. Are you finding that's changing?
- [11:02]
A
No, I would agree with that. As I said, kind of. MMM is probably not the best tool for every single tactic, but it's the best tool to get a holistic view of the entire marketing plan. Now we supplement MMM with either sales lift studies or brand lift or AB testing, market testing to supplement what data we're missing from MMM and get a holistic view of kind of the entire performance. So as I said, it's just one data point. It's probably the more robust data point that we have. But it's important to have multiple data points, to have an informed point of view of what's working and what's not. Right.
- [11:39]
B
You referred to have all these different pieces you have to pull together, studies, MMM data, things like what's missing, like what do you want more of from your TV measurement providers, for example?
- [11:50]
A
Well, I would say is first of all sales list or outcome based measurement. Right. So I think we need to continue to push the industry to move towards business. KPIs making sure that we measure the performance of our marketing investments based on business outcomes and not just based on media metrics. So that's one. And the second one I mentioned a bit ago and it's a cross channel measurement. Right. So it is an industry challenge. Right. So with the walled gardens, we are not able to get a holistic view of all of their media activity. Right. So either we get video from TV measurement providers that are measuring linear TV and connected tv. But then we're missing kind of a big component of the video ecosystem with YouTube. Right. Or with social media. So I would say that that cross channel measurement between TV activity as well as digital activity is, is kind of a big piece that we're missing in the industry.
- [12:46]
B
And once when it comes to the sales lifts, you know, I've heard different opinions on this. Is that, is that the, should that be the burden of the media company, the seller? They'll say, well, you know, TV doesn't drive outcomes immediately like other things. We need to be judged differently. That's the kind of on the brand to look at that. But should that be part of the package?
- [13:07]
A
I think absolutely should be part of the package. Right. So because at the end everything that we're doing is to drive brand growth. Okay. And the best way to measure brand growth is there are leading indicators and there are lagging indicators. Right. So yeah, we can agree that there are other metrics outside of, of sales lift that are important. Brand health metrics, brand equity, KPIs. So absolutely we're tracking that as well and that's important. But at the end of the day, in order to justify investments in TV or any type of marketing investment, we need to prove to the organization that is driving a business outcome either in the short term or we should have a point of view about what is the impact that is having over the long term.
- [13:48]
B
Is that way different than when you started? That kind of burden of proof? Has it really changed over time?
- [13:54]
A
I think it has always been there. I think as marketing investments are more diversified and there are more channels and channels that are harder to measure then I think it's getting kind of more and more attention of, okay, how do you demonstrate that this investment is driving a positive return? So I think it's getting a little bit more focused because of the diversification of media and new media tactics that are basically unproven or have limited data. But I think that the need for companies to demonstrate business results has always been there.
- [14:26]
B
It's not like they just didn't care in the old days. Spent money. Exactly. Okay, I want to ask you a couple of AI questions. Of course, right now, are you doing anything that would be considered agentic yet? Is that still very theoretical? There's so much excitement around that in the ad tech world. I don't know what's actually really happening out there.
- [14:43]
A
Yeah. What I can tell you is that we are using machine learning and AI tools that are already available, especially in digital Media. Right. So when we're talking about bidding optimizations, when we are talking about supply path optimizations, we are using AI tools to help us with that. Now, are we using agentic capabilities? Not yet. We're having conversations with different ad tech vendors that are developing capabilities. We are intrigued. Probably when we feel that the capability is ready, we would take more like a testing approach before we actually scale it. But no, we are not yet using any agentic capability for media buying.
- [15:21]
B
My sense is most brands are, tell me if you agree. Like they're ready to use agents for grunt work, but maybe not making decisions without oversight. Is that sort of the right.
- [15:32]
A
And I think that's, that's probably the way that we feel right now. Right. So I do believe that agentic capabilities are going to continue to evolve. I think they're going to get to a point where it's going to be more important for media buying and for marketing analytics. At this point, again, I think it's in its infant phases. So we're not ready yet to trust 100% of an agent to make a decision for us, especially when we're talking about millions of dollars of investments that we're making in these channels. Right.
- [15:58]
B
So you're still showing up for work every day?
- [16:00]
A
Absolutely. And we have a team that we think it is important that has kind of the best knowledge of our brands, the best knowledge of our consumer targets, the best knowledge of the industry to be able, kind of to guide kind of those important decisions.
- [16:13]
B
Okay. So on that note, you have more knowledge probably than you ever did about your customers and your media spend. How much is AI driving your media buying and planning and optimization? Because you know, there's a lot of thinking, well, eventually the media plans will run themselves. You don't need agencies, you can do everything yourself. Where do you see that playing out?
- [16:33]
A
Yeah. So I can tell you again, we are using already AI capabilities in digital media. We manage all of our digital media buying in house. We have a group of programmatic traders, social media managers, search marketing managers, and they're already using AI tools for optimizations on kind of day to day campaigns. Okay. The same way that agencies are leveraging or developing AI capabilities to help with media planning and buying in the future. Okay. I don't think that with the use of AI and the evolution of AI capabilities, agencies are going to go away. I think they need to evolve their model. They need to find ways to add incremental value, marginal value over what the machine is going to be able to do. But my point of view and the way that we're working today with our media agencies, they supplement what we do in house. And I think as we both start to use AI capabilities, I think we both are going to evolve where we can create value. I don't think that the model, the agencies are going to go away. I don't think that the in house teams or program administrators are going to go away. We just need to add value in a different way.
- [17:41]
B
What about, you know, the, the platforms kind of talk about just give me your dollars and we'll, our machine learning, we'll figure out what to, how to, how best to spend dollars. Are you wary of that or does it depend?
- [17:52]
A
I would say it depends. And we test, we validate before making a decision about leveraging those kind of AI tools from the different platforms. Of course they have a vested interest on driving dollars to their platform and that's why we are, I would say always more careful about making those decisions. We do believe that technology and AI can help, but we need to make sure that we validate the incremental value that the AI tool can provide before we actually lean in.
- [18:21]
B
You kind of hinted at this about like the need to prove the value of investments to the organization is, you know, maybe getting more scrutiny than it used to. You hear a lot now like the CMOs are butting heads with a CFO or they need to figure out how to talk to them or they're getting in trouble. How does that without single singling out any individual cfo? Like how, how do you see that in the, in the industry right now? How do you communicate the value of stuff that maybe isn't immediately trackable? Like, like a CFO might be used to.
- [18:49]
A
Yeah, I mean I think it is important for senior leaders regardless of their position if it is a cfo, if it is a CEO or if it is a, the coo. Right. So to understand kind of the, the impact of, of marketing in order to drive, to build brands. Right. And to drive brand growth over the long term. So it starts with kind of that education and, and they need to bind into kind of that concept. Right. So if senior leadership buys into the concept of marketing is a lever to drive brand growth, then at the end is then having the right measurement approach in place to show them what is the short term impact of that marketing activity and what is the long term impact of that marketing activity. But it starts with kind of the understanding of that marketing plays a role in the short term and in the long term. And then we need to have kind of different measurement approaches to, to provide the support for that.
- [19:42]
B
It's not, it's not a one size fits all. It's not, it's not something that is only justifiable in the short term. You have to just be able to break those pieces apart. I guess. What do people not maybe understand about brands bringing things in house? What, that's what's involved there, why that strategy makes sense for you guys, why what kind of the challenges they might face.
- [20:01]
A
I would say probably kind of a misconception is that is very easy to do that you can build an in house team very quickly and you are going to get savings just because the talent cost is cheaper to hire in house versus paying an agency. I think that's a huge misconception. It took us about three to four years at Georgia Pacific to build an in house capability and scale that in house capability. We started at a small scale with one brand, with one type of media tactic. Okay. And from there we improve the results not only in cost savings from talent fees, but we had cost efficiencies in the media cost. We had a lower CPM when we had more control over campaign optimizations. We had higher effectiveness because we had better knowledge and better connectivity with the rest of kind of the media plan. So we proved that out early on. And then from there we added another tactic and we added another brand. And I would say it takes time.
- [20:59]
B
It's not just hire a couple people and you're, you're good.
- [21:01]
A
Exactly. I would say another one is building the relationships with those ad tech partners is difficult and getting the, the rates to really achieve that CPM efficiency, it is difficult. If you don't have the scale, getting
- [21:17]
B
the talent in place automatically transfer over. You got to forge those things.
- [21:21]
A
That's why, I mean there is a role for the agencies. They have the skill. Right. So in order for you to get those kind of same CPM rates or data fees, it's going to take time and it's going to take kind of a strong relationship with some of those partners. Also the talent, I mean, getting talent and retaining that talent, it is hard. It's not that easy. Right.
- [21:41]
B
So especially at the beginning. Right. That's a leap of faith for someone to get someone to come in and say we're going to, you're going to build this with us.
- [21:46]
A
Absolutely. So I would say are there benefits to in housing? Absolutely. But it takes time, it takes effort and it's not easy to maintain.
- [21:54]
B
Yeah. All right, Javier, awesome conversation. Thank you so much for taking your time out and let's talk again down the road here.
- [22:00]
A
Well, thank you, Mike. I really appreciate it.
- [22:03]
B
Thanks again to my guest this week, Georgia Pacific's Javier Burkillos and my partners at Vizio. If you like this week's episode, please take a moment to rate and leave a review. We have lots more to bring you, so please hit that subscribe button. We'll see you next time for more on what's next in media. Thanks for listening.