How Digitas is navigating search’s shift to the AI era, OpenAI's latest talent poach and Netflix's push for personalization

Summary

The Digiday Podcast: Navigating the AI Era in Search, Talent Moves at OpenAI, and Netflix's Personalization Push
Release Date: May 13, 2025

In this insightful episode of The Digiday Podcast, hosts Kamiko McCoy and Tim Peterson delve into critical shifts shaping the digital landscape for brands, agencies, and publishers. Covering the decline of traditional search engines in the AI era, strategic talent acquisition at OpenAI, and Netflix's groundbreaking personalization efforts, the episode offers a comprehensive analysis of the evolving media and marketing environment.

---

1. Recap of the Programmatic Marketing Summit (DPMS)

The episode opens with Kamiko and Tim reflecting on their recent attendance at the DJ Programmatic Marketing Summit (DPMS) in Palm Springs, California.

Tim Peterson (01:15):
"Heading into it, I was expecting a lot of conversation about the third-party cookie. There was virtually no conversation about it."

Despite anticipating discussions around third-party cookies, the summit surprisingly lacked focus on this topic. Instead, conversations gravitated towards other pressing issues like transparency in the Connected TV (CTV) ad market and the persistent "ad tech tax" that continues to burden agencies and publishers.

Kamiko McCoy (03:12):
"It almost feels like if we put the text of what was said during DPMS in May 2025, it might as well have been May 2019."

The hosts highlighted the stagnation in addressing long-standing issues, underscoring the need for more progressive solutions in the digital marketing sphere.

---

2. The Decline of Google Search and the Rise of AI-Powered Search

A significant portion of the discussion centers on Google's waning dominance in the search engine market, particularly within Safari browsers, and the broader implications of AI on traditional search behaviors.

Tim Peterson (05:33):
"In April 2025, Google search query volume within Safari declined for the first time in its history, which is a big thing because that seems like the latest indication that AI is probably cutting into traditional search behavior."

The hosts dissected Google's response, which emphasized overall query growth but avoided addressing the specific decline within Safari. This decline is partly attributed to AI-infused search tools like ChatGPT and Gemini, which are reshaping how users seek information.

Kamiko McCoy (07:36):
"AI has really started eating away at how we traditionally have searched for things. And that seems to be something that brands and publishers are taking notice of."

Alaina McGurn, SVP of Search at Digitas, is featured later in the episode, elaborating on the shift towards AI-powered search and the necessity for new strategies in search engine optimization (SEO).

---

3. OpenAI's Strategic Talent Acquisition: Instacart's CEO Joins

OpenAI's recent move to poach Fiji Simo, the former CEO of Instacart, marks a significant strategic shift towards enhancing its business operations and advertising capabilities.

Tim Peterson (10:37):
"OpenAI is hiring its 'Sheryl Sandberg,' which, for all the faults of that comparison, Sheryl Sandberg did a lot to build Facebook's and eventually Meta's business."

Fiji Simo's appointment as CEO of Applications at OpenAI signals the company's intent to diversify its revenue streams beyond subscriptions and enterprise sales. With her extensive background in ad tech and retail media, Simo is poised to spearhead initiatives that could position OpenAI as a formidable player against tech giants like Alphabet and Meta.

Kamiko McCoy (12:48):
"Instacart is one of the biggest players in retail media, right? If you're talking top five, Instacart is in there right behind Walmart and Amazon."

The hosts speculate on the potential impacts of this leadership change, suggesting that OpenAI is gearing up to compete more aggressively in the advertising and social media spaces.

---

4. Netflix's Major Home Screen Redesign and Enhanced Personalization

Netflix's latest overhaul of its home screen, the first since 2013, aims to revolutionize user experience through real-time personalization and more dynamic content presentation.

Tim Peterson (14:37):
"With this redesign, there's going to be fewer titles, but more video and animation. It's going to reflect search behavior in real-time."

The redesign focuses on reducing the clutter of titles and incorporating more engaging visual elements. A key feature is the immediate reflection of user search behavior in home screen recommendations, facilitating quicker content discovery and potentially increasing user retention.

Kamiko McCoy (18:08):
"If you're able to drill down the personalization, Netflix has a better chance of keeping you in the app."

This enhancement not only improves user experience but also presents lucrative opportunities for advertisers by enabling more precise audience targeting and synchronized viewing events, akin to the Super Bowl's advertising model.

---

5. Featured Segment: AI as the New Search Engine with Alaina McGurn

A highlight of the episode is a live conversation with Alaina McGurn, SVP of Search at Digitas, where she explores the transformative role of AI in search and the emergence of AI-driven search engine optimization (AI SEO).

Alaina McGurn (22:10):
"AI platforms get to decide what gets to be in those answers. If you're not in the answer, if your brand, if your product is not featured in those answers, you're essentially invisible to a big chunk of your audience."

McGurn discusses how AI has shifted the competitive landscape for brands reliant on traffic for revenue. Traditional SEO tactics are evolving as AI platforms curate personalized, contextually relevant responses, necessitating a reengineering of content strategies to become the preferred sources for AI-generated answers.

Tim Peterson (24:34):
"With Google Search, there was the PageRank algorithm... Does that even exist with AI?"

Alaina McGurn (25:09):
"We are very much focused on optimizing that consumer experience for context in all forms of relevance... brand relevance, audience relevance, topical relevance."

She emphasizes the importance of integrating various data sets—brand voice, product content, and commerce strategies—to align with AI's evaluation criteria. This approach ensures that brands can maintain visibility and effectively compete within AI-curated search results.

Measurement and Analytics Challenges

McGurn highlights the complexities in measuring AI-driven search visibility, noting the lack of universal analytics platforms and the necessity to align measurement strategies with specific business KPIs.

Alaina McGurn (28:10):
"We have to start with the end result first and then looking at the ways that you can measure that... How is external AI search visibility affecting my performance across different retailer platforms?"

She also discusses the critical role of direct data feeds to AI platforms, which enhances real-time accuracy and visibility in AI responses, thereby boosting traffic and sales.

Tim Peterson (36:04):
"Does search AI become adopted enough that we are able to kind of feed the beast here?"

Alaina McGurn (36:33):
"The more data we provide in a way that is consistent and helpful... the better the message is, the better the experience is for consumers."

---

6. Concluding Insights

The episode wraps up with a reflection on the ongoing transformation within the search landscape and the necessity for brands and marketers to adapt swiftly to maintain visibility and competitiveness in an AI-dominated environment.

Tim Peterson (33:25):
"We are becoming more product managers because we're synthesizing and integrating so many different data sets to provide those personalized curated experiences to consumers."

The hosts underscore the imperative for strategic integration of data and content to thrive in the new AI-powered search paradigm.

---

Key Takeaways

• AI's Encroachment on Traditional Search: AI platforms like ChatGPT and Gemini are significantly altering search behaviors, leading to traditional search engines like Google experiencing declines in certain areas.

• Strategic Talent Moves: OpenAI's hiring of Fiji Simo signifies a strategic push towards enhancing advertising and business operations, potentially increasing competition against tech giants.

• Personalization in Streaming: Netflix's home screen redesign enhances real-time personalization, improving user experience and creating new advertising opportunities.

• Evolving SEO Practices: The rise of AI-driven search necessitates new SEO strategies focused on context, relevance, and direct data integration to maintain brand visibility.

• Measurement Challenges: The shift to AI-powered search presents complex measurement and analytics challenges, requiring brands to define clear KPIs and leverage multiple data sources for effective evaluation.

This episode of The Digiday Podcast provides a thorough exploration of how AI is reshaping the digital marketing and media landscape, offering valuable insights for brands, agencies, and publishers navigating these transformative changes.

Transcript

[Kamiko McCoy]

Foreign hello, hello and welcome to another episode of the Digiday Podcast, a show about the business of media and marketing. I'm Kamiko McCoy, senior marketing reporter here at Digiday.

[Tim Peterson]

And I'm Tim Peterson, executive editor of video and Audio Digiday Media. What's up, Kimiko?

[Kamiko McCoy]

How are you, Tim? Are you good?

[Tim Peterson]

I'm good.

[Kamiko McCoy]

I'm covered.

[Tim Peterson]

A little bit rested. Yeah. After. So you and I were both at the DJ Programmatic Marketing Summit this past week in Palm Springs, California, which was not as hot as I think I had like pitched to you a few weeks back on the podcast and it's even rainy at times, but I thought it was like really nice weather. Like, it was the best weather I've had in Palm Springs. As someone who wants nothing to do with any anything even, you know, in.

[Kamiko McCoy]

The mid-80s, it was a lovely, lovely little treat. I will say the, the ride back for me back to Atlanta, there were no direct flights, so me and everybody that was going to New York suffered in Salt Lake City for a couple of hours before we made it back to our respective homes. But overall, it was a really, really good summit. What did you find most interesting?

[Tim Peterson]

Yeah, I mean, one thing I found most interesting, like heading into it, I was expecting there to be a lot of conversation about the third party cookie. There virtually no conversation about the third party cookie. Like we had these behind closed door town hall sessions with the agency execs in attendance. We figured we'd open that up by talking about third party cookie. You know, Google, okay, is going to keep third party cookie around in Chrome, but it's still not available really anywhere else. Safari, Firefox, CTV mobile apps, which feels messy to me, but no one wanted to engage in that conversation.

[Kamiko McCoy]

It was almost like you could hear like an audible sigh or like a pin drop in the room. You just like, I'm watching you announce like, oh, what's going on with cookies? Let's talk about it. And everybody's eyes start slowly gravitating toward the door.

[Tim Peterson]

Yeah, yeah, so. But there was plenty of other stuff to talk about. Like transparency in the CTV ad market was a big one. Also transparency when it comes to the ad tech tax, which neither of these are new issues. And so like, one of the interesting things to me is that they continue to be such pervasive issues. All these years later. People are talking about like all the fees that various DSPs are taken from them and how the money isn't, you know, not as much money is making it to the publishers. And you know agencies needing to justify to their clients why they're having to spend however much they're having to spend or bid whatever they're having to bid. So again, it just, it almost feels like if we put like the text of what was said during DPMS in May 2025, it might as well have been May 2019.

[Kamiko McCoy]

I think that's a very good way to put it. And fortunately for everybody that's listening on our site will have overheard at DPMS up on Monday. So keep an eye out for that if you want to read said transcript. And then also there was an interesting come to Jesus moment at the end of this town hall, which I think we had last year as well, where there was a, I guess a we want to talk about this more transparently amongst ourselves. And you are going to be hosting a conversation for this. No?

[Tim Peterson]

Yeah, we're gonna do a virtual town hall on May 30 at 1pm Eastern. It'll be limited to just agency executives. That was a big request from the group, which makes sense. They want to be able to talk about this stuff amongst themselves as opposed to like in front of vendors or publishers or even clients. So look out for a post going out on the Digiday LinkedIn account where, where people will be able to register for that virtual town hall. We will be going through registrations to make sure that only people who are at agencies will get the link to the zoom. But then we'll have that conversation again on May 30th at 1pm Eastern. So I'm looking forward to that. I'm also, as much as I've made the joke, this DPMS felt like it took place in 2019. Not every part of it. Like for our featured segment for this week's episode, we have a live podcast recording that you and I did with Alaina McGurn, who is the SVP of Search at Digitas, where we talked about a very new thing in the industry, which is this idea of AI becoming the new search and what does that mean for search engine optimization tactics, search engine marketing, you know, what is AI engine optimization? So I thought it was a really fascinating conversation and obviously a great topic.

[Kamiko McCoy]

It was, it was. And I'm excited to get to that. But first, we've got a couple of juicy scoops, as I often say, that we're going to cover, including, to your point just now, search, Google search declining. And we've also got on the docket OpenAI, who has poached itself new CEO and Netflix redesigned its Home screen. So we've got some juicy bits in there before we get to our conversation with Elena. But first let's start about Google search in declining.

[Tim Peterson]

Yeah, yeah. So the of the two antitrust cases that Google is in the midst of, so it lost the search antitrust case last year. Now they're having hearings to figure out, okay, what should the remedies be for this? And so as part of these hearings, Apple executive Eddie Q testified. And during his testimony drops the BombShell that in April 2025, Google search query volume within Safari declined for the first time in its history, which is a big thing because that does seem like the latest indication that AI is probably cutting into traditional search behavior. Google has responded to this and I'll quote from, from their blog post, we continue to see overall query growth in search that includes an increase in total queries coming from Apple's devices and platforms. So Google's saying no, not true, naturally. But Google's also saying like total queries coming from Apple's devices and platforms, not calling out Safari specifically. So it does seem like Google is finding a way to thread the needle here. And that, you know, does seem like Eddie Q. I don't know why he would lie under oath like that. Seems like. No, nope, bad idea. So this is a notable thing also in Alphabet's earnings. Google's parent company, it had disclosed a week or two ago that paid click growth for Google had decelerated to 2% year over year in Q1 2025 versus 5% growth in Q1 2024. So, you know, the number of times people clicked on ads in Google search, it still grew. Yeah, in the first quarter, but it was still a deceleration at a time where search, traditional search does seem to be ceding more ground to AI chatbots, which is effectively like AI powered search.

[Kamiko McCoy]

You know, I was having a conversation with Ronan Shields, our ad tech, our ad tech reporter, excuse me, on the way back from DPMs and after like hearing about Google and whatnot, I said, do we think that this era of like tech, mostly Google, because they're getting walloped on all sides right now. Right. Is done. And he said, no, but OpenAI is probably the first platform or AI is the first platforms, right. To really give Google a run for its money here and now we're also hearing that Apple is planning to add some AI based search capabilities as well. So like their fighting chance at this point is just getting harder and harder.

[Tim Peterson]

Yeah, yeah. Although you know, at the same time there was a story within again, the Past week or two of, I think it was Sundar Pichai, the CEO of Alphabet, and may have been during their earnings call or during one of the DOJ hearings, talked about having talks with Apple about getting Gemini set up on iPhones in the same way that ChatGPT is already set up on iPhones. And Apple has said as much that it, you know, while it initially started with ChatGPT being kind of hooked into Siri a bit more natively on iPhones that it would be adding these other large language models to it. So Google's by no means out of work. There's just, to your point, a lot more competition for Google than there's ever been. Like OpenAI is not Bing.

[Kamiko McCoy]

Correct? Correct. But to your point, AI has really started eating away at how we traditionally have searched for things. And that seems to be something that brands and publishers are taking notice of. Like to the point about the conversation that we had with Elena about how they're pulling data in to be able to feed the beast and get these ads out there on these platforms as they continue to roll out opportunities and learn more. But you also brought up, and this will come up in the conversation, that there are limitations in this space as well for what, what we're able to do.

[Tim Peterson]

Yeah, I mean, there's just, you know, a lot less control. It's not even like control, but like it's harder to influence the large language models if you're a company versus traditional search, where still it was, you know, just influence. It wasn't control. But there historically have been like tactics you could use, even ways that you could design, you know, website pages to tailor content for the search crawlers that would be crawling the websites. The Wall Street Journal had a really interesting story this week talking about the same topic we talked about with Elena, which will come up later in the show, but like about AI engine optimization. And in it, you know, mailchimp, you know, talked about how it's seen a traffic decline that it attributes to AI. And so what it's been doing is redesigning the code on its homepages to be a bit more AI crawler friendly, which I imagine everyone's going to be doing if they're not already in process of doing.

[Kamiko McCoy]

Yeah, I think that's, that's fair to say. And other big news, staying on the AI topic, OpenAI is poaching Instacart's CEO. I know OpenAI has got some real big ambitions of adding ads at some point and there's been rumors of it launching a social platform and whatnot. And given what Fiji Simo has been able to do at instacart, I think the rationale here is building that into a proper business. It'll be the same at OpenAI.

[Tim Peterson]

Yeah, yeah, the like comp here and it is by no means a perfect comp and even a little reductive, but the idea is that this is OpenAI hiring its Sheryl Sandberg, which again, for all the faults of that comparison, Sheryl Sandberg did a lot to build Facebook's and eventually Meta's business and really maturing that company, turning it into the Goliath that it's become. Fiji Simo seems like the kind of person you would want to hire at OpenAI to do much the same. So Sam Altman is still going to be there. He's still going to be CEO. Fiji Simo is going to be reporting to him, but she is going to have a CEO title. She's going to be CEO of applications. Her oversight is going to include the business, operations, marketing, sales, finance, so all of the money makers and where the money is, because that's the thing that OpenAI is going to need to figure out. Now OpenAI has canceled its plans to turn into a for profit company, so it's going to stay nonprofit, but it still needs to be bringing in money in order to offset the like astronomical costs of powering its large language model. And ads is probably going to be a part, a big part of that. Fiji Simo has a pretty great track record when it comes to ads. She was, you know, at Facebook before where she was very integral into its Facebook's push into native video and eventually took over oversight of the newsfeed and Facebook as head of Facebook before she then went to Instacart and really like seemed to steer instacart into the retail media game. And, and Kimiko, this is your world. Instacart's one of the biggest players in retail media, right?

[Kamiko McCoy]

It is, it is. As a matter of fact, if you ask anybody to list like their top five retail media networks, which I've like drilled this into everybody's head, there's more than 250 of them at this point. So if you're talking top five, Instacart is in there right behind Walmart and Amazon. It can be interchangeable with like Target and whatnot, but it's up there, which speaks to a lot of what Fiji has been able to do with that ad business.

[Tim Peterson]

Yeah, so it seems like the playbook is there for how OpenAI can generate revenue beyond like what it's already doing when it comes to subscriptions and enterprise sales. But in terms of consumer revenue, like advertising, affiliate revenue, which they've started testing out affiliate links within ChatGPT, they get some money there. So it seems like this is a really good person to be overseeing the building of those revenue streams. And like we were saying with Google, like that's going to create competition for Google from an ad sales side of things when it comes to search and even, you know, social. You mentioned there's. It's been reported that OpenAI is considering launching some sort of social platform that could eat into Meta's ad revenue as well. Basically, like OpenAI seems to have pick someone to put it into that Alphabet or Google Meta, Amazon Triopoly to create, you know what a quadruply. Is that going to be the term we're going to have to adopt soon? The Four Horsemen, maybe War Horseman, Apocalypse?

[Kamiko McCoy]

I think that's a really good way to put that. For sure. For sure. And we'll probably circle back around when there's more news, when she finally steps into the role officially later this year.

[Tim Peterson]

Absolutely, yeah. And I mean, as much as we talk about those Big four, there's still a lot of big companies on the TV and streaming front, like Netflix, which is this week going to be rolling out the first redesign, significant redesign of its home screen on connected TVs since 2013. And it's a fairly significant redesign, if only because when every other company launched their streaming service post Netflix's last redesign, they seem to just ape the design that Netflix had. And so I would expect probably a similar thing to happen here. So with this redesign, the main thing is there's going to be fewer titles, but more video and animation. According to the New York Times, there's going to be, if you search for something on Netflix, I guess you know, historically that search behavior wouldn't be reflected in the recommendations you get on the home screen until like a day later. Now it's going to like be reflected in real time. So it seems like a lot of this is for Netflix to be able to push you to find something a lot quicker, which also seems like it's going to be very helpful when it comes to Netflix having, creating appointment viewing things like live events like the Christmas Day NFL games. But also I would imagine when it has a new show premiering, being able to push people straight into that, which will probably help Netflix, if it can establish that that is the case, will probably help Netflix when it comes to its dealings with advertisers which that's what this week is. This is Upfront's week. I think the day this episode goes out, that night Netflix will be having its upfront presentation. I'll be there. I'm very impressed. Interested to hear what they have to announce there. It's. They've been doing a lot on the ad tech front, but now they're doing a lot on the product front too.

[Kamiko McCoy]

That's gonna be really, really interesting just in how much. Cause you think about Netflix when you go. Well, actually, let me pitch this question to you. When you open Netflix, like, do you go just to browse? Are you looking for something specific?

[Tim Peterson]

Sometimes I go to browse. A lot of times I go to like, my list. Cause sometimes I'll be on Netflix and just adding things to my list of like, okay, I'm gonna wanna watch that eventually. And at this point my list has become like my DVD library in the past, where I don't even remember everything that I have in my list. Plus things leave Netflix and come back to Netflix on such a regular basis that a lot of times I'll go through my list and it's like, oh, that's on Netflix now. Cool. Like, this happened to me a few weeks ago. I was like, okay, I got, you know, a couple of hours is the end of the workday. I had some time to chill. Let me see what's in my list. And Heat was back on Netflix. Now Heat is a three hour movie. I did not necessarily have three hours to watch a movie, but Heat was back on Netflix. I hadn't seen Heat in at least a year. So of course I fired up Heat and watched the whole damn thing. So. But something that's plagued streaming services in general is it just takes people too long to find something to watch.

[Kamiko McCoy]

Well, I mean, if there's a personal. If they're able to drill down the personalization. Right. As soon as you get to the landing page one, you're sucked in. There's. There's no getting out of it. You are. The TV's watching you at that point. But I mean, to your point earlier, it'd be really good for advertisers because one of the things that we talked about at DPMS also is just this push for as much insight as they can get from CTV channels.

[Tim Peterson]

Yeah. Because it's also that like, balance between personalization, but also group think. Yeah, if you will. But like, especially when it comes to the upfronts and TV and streaming advertisers. Yes. They want to reach the right Audience at the right time with the right ad, like all, you know, that usual shtick. But they also want to reach a large group of people at the same time. That's why the super bowl is where gets the most money from advertisers. And so if Netflix can be pulling the levers and doing a better job of getting as many people as possible to be watching one thing at the same time, that's going to be really appealing to advertisers, especially the big brand advertisers that spend in the upfront.

[Kamiko McCoy]

Yeah, absolutely. And also I would imagine cuts down on some of that churn where you know, if you scroll, scroll, scroll, can't find anything and then you leave the app looking for something else. If you've got more detailed personalization, Netflix got a better chance of you keeping you in the app.

[Tim Peterson]

Yeah, I would think so. Although Netflix had been doing fine in the past year or two when it comes to churn and it doesn't even report subscriber numbers anymore, so we would have no idea either way.

[Kamiko McCoy]

We'll just have to believe them.

[Tim Peterson]

I guess one thing that I'll be interested in just because we've just been talking about search and AI and so when you're hammer, everything is a nail. The idea of Netflix doing a better job when it comes to taking searches and having that reflect in the home screen recommendations, at what point does it really start to implement AI into this? And like in the same way that like YouTube and Amazon became two of the biggest search engines around, is there a potential for something like a Netflix to become one of the biggest search engines around? I guess, you know, that kind of depends on how important movies and TV shows as the library to be searching is. But I'm sure there's some potential for even if you're searching for something on Netflix, that Netflix could start running ads in those search results. That seems like horrible from a user standpoint, but at the same time potentially there's money to be made there of if you're searching for a family movie, kind of that idea of then you get a, I don't know, a pizza delivery coupon ad or something along those lines.

[Kamiko McCoy]

Well, if it ever did come to fruition, we know exactly who we would go talk to, which would be Elena, who we already had a conversation with and we'll hear more about that in our feature segment.

[Tim Peterson]

Yeah, yeah, this is a really fun conversation. I think there's so much to talk about when it comes to the future of search in this AI era.

[Kamiko McCoy]

So no further ado let's get to it.

[Tim Peterson]

So the idea for this session kind of came out of a conversation I had with the CEO of Digitas, Stern CES earlier this year. There's a lot of talk at CES about the agentic era of AI and AI in general. And what she put me onto is this idea of, okay, a lot of people have drawn parallels between AI chatbots being kind of the new search engines. And what she said is that's going to mean we're going to need to create a version of search engine optimization for these AI chatbots. You oversee search at Digitas. So I imagine a lot of this falls under you. How are you thinking about the parallels between traditional search and search engine optimization and search engine marketing tactics and to what extent those apply to these AI chatbots? Like a ChatGPT Claude Gemini Absolutely.

[Alaina McGurn]

AI has become such a game changer for us as consumers, the way we find information. But in a way, it's also become a gatekeeper for brands who rely on traffic to drive revenue, who rely on traffic for monetization. There's been such a seismic shift in how we find information. We're not getting the Templar links, we're getting those curated, perfect, personalized experiences, the answers. And AI platforms get to decide what gets to be in those answers. So if you're not in the answer, if your brand, if your product is not featured in those answers, you're essentially invisible to a big chunk of your audience. That's 80% of US consumers now rely on AI platforms. Whether it's Gemini Perplexity, I don't know what you guys are using ChatGPT for product inspiration. They rely for brand comparison, for shopping as they find and discover that information. So obviously that has big, huge implications to monetize. Value lost, ad revenue value lost. Any large publisher, any large content publisher that is reliant on site traffic has seen major losses in the past year. Depending on the vertical, we're seeing anywhere from 25 to 75 of reduction in site traffic. That changes the game. They're no longer competing with other publishers, other large websites. They're competing with AI itself. So the question that we are asking ourselves as an agency is not so much anymore for how do we help our clients be found? It's how do we help our clients become the answer, that trusted source of information that AI draws from when they present that curated experience, personalized targeting experience to consumers with options to purchase. And in the process of that, we are finding ourselves re engineering, re architecting a lot of the old strategies Introducing new strategies because we are working with platforms, content platforms that are not just the source of content, but they're also the reviewer, the editor and the recommender, all packaged in one.

[Tim Peterson]

I'm going to date myself with this one. But I remember with Google Search There was the PageRank algorithm, which was the way for anyone running a website to be able to figure out how is Google going to be evaluating my page? A lot of that was like linking to other pages. And so that's why you saw like the blog roles. That search has obviously got insanely complicated since then, but it started out with something of a clear way to figure out how to not game the system, but to market to the algorithm. Does that even exist with AI?

[Alaina McGurn]

Yes, yes, absolutely. So if you look at how. Right. Like I think the key word there was really the how AI search platforms are looking at content, how they're presenting that information, we have to look at obviously in what context we're showing up. So gone are the days when we are looking at very specific keywords to insert in titles and descriptions or put into our ad placements. We really are very much focused on optimizing that consumer experience for context in all forms of relevance. So that's brand relevance, audience relevance, making sure that we are aligned with what we want to put out there, aligned with what the audience is expecting to see, but also capitalizing on the topical relevance, capturing the most recent conversational trends that are happening within that space. I mean, if the goal is to drive some form of revenue, whether it's ad revenue, of product revenue from your website experience, the content that you have out there, your media plan, media strategy absolutely has to account for brand voice, product content and commerce strategy to be aligned with how these AI platforms are evaluating that content.

[Kamiko McCoy]

If there's more context that's happening around these, there's no longer just keyword search queries.

[Alaina McGurn]

Right.

[Kamiko McCoy]

You've got more. And I've heard the word brand happen a couple of times here. So at what point do we determine is it a brand play or performance play, or as I mentioned yesterday, a brand Forman's play?

[Alaina McGurn]

So that's a great question. Right. Obviously your visibility and engagement right. Earlier on in the funnel, how you are driving that inspiration around your brand ultimately affects how much you're converting.

[Tim Peterson]

Right.

[Alaina McGurn]

As well. So we definitely want to be looking at that from both the brand perspective. Right. As well as the product perspective. So where I like to start is the end result, ultimately, what is it that we're trying to achieve? If you're looking at selling more products, right. If you're looking for maybe your Amazon links showing up more prominently in those AI search results, then you have to start building that inspiration collaboration in the commerce space with the content that gets surfaced in that context. And that's exactly how we are also approaching measurement around these specific tasks. If you are looking for more product visibility, you're then looking for how is that external AI search visibility then affecting my performance across maybe different retailer platforms, driving traffic to my product listings on the website, right. Selling more products or more ad revenue.

[Kamiko McCoy]

Actually that's a very good point you bring up there. And I'm always asking about measurement because I feel like that's always the big question, right? Is you know, how does measurement differ when it was traditional search versus AI powered search.

[Alaina McGurn]

There's so many options that we're now seeing starting to pop up in this space. There's not one universal analytics platform that really helps us assess share voice visibility across multiple platforms. That's why I really like to start with what is the end goal that we're trying to achieve here so that we are measuring impact in terms of business. KPI's right. Unis sold. Some of our clients look at AI visibility as a form of risk mitigation, brand reputation management. Others rely on that for something as maybe tangential as investors, investor relations, right? Meeting investors, answering their questions before they even pop up and become asked in anticipation maybe of a specific event. Making sure that you have the right content to answer those questions because you know they're going to be asked, right? So whether it's your brand reputation management or selling more products and boosting your commerce presence in AI search results, you have to start with the end result first and then looking at the ways that you can measure that. The interesting thing with the commerce side of it is we're now seeing retail platforms actually stepping up and helping us measure a lot of that engagement on their end as well. They care about how much play their commerce content, product listings, even Amazon listings, how much play they get in AI search results, how much visibility that brings into those retail sites in turn. So something like Amazon Attribution does track how much play your organic listings, your sponsored listings got in perplexity in ChatGPT, Gemini overviews search results, how much traffic they brought in and then those are the listings that you're going to see pop up up first, sometimes organic, over sponsored because they were more popular, more visible in those AI generated responses.

[Kamiko McCoy]

It seems like there's a lot of moving parts there. So the last thing that I want to ask about measurement is, is it an Apples to Apples conversation when you're talking about the different platforms and the different ways that you can measure these things, can you make it an Apples to Apples conversation or do you kind of have to hodgepodge the measurement together to make sense of it?

[Alaina McGurn]

So we are starting to see a lot more options now with some of our older partners that came out and helping us track this AI visibility for each specific platform. What we're doing right now is we're trying to syndicate that across as many key AI platforms as we can. Right. So that we can say, all right, here's the share of voice that we're seeing for this particular client as it relates to this category or these specific types of questions that we care about consumers asking. Right. Looking at again, sentiment analysis, so the answers that are being surfaced that are targeting specific consumers, what types of categories of that sentiment analysis are we seeing? What trends can we predict and how can we optimize our future content to meet that demand so that we are providing those curated consumer experiences at just the right moments in time?

[Tim Peterson]

So I want to make sure. So when I'm having a conversation with ChatGPT about recommend me whatever shoes or something like that, you all are able to get that Data out of ChatGPT?

[Alaina McGurn]

Absolutely. So you can see how often you're showing up or how often your products or your brand mentions are now showing up in ChatGPT. Perplexity, Gemini overviews. Gemini just itself just in like a.

[Tim Peterson]

Generic conversation, not with Amazon links or anything like that.

[Alaina McGurn]

Exactly, exactly. So you can see the types of mentions that are popping up, how often those mentions are popping up, and also the sentiment around those mentions. Right. Quite often you get that curated response and there is a lot more in there, related questions. Right. Sources of information that come maybe with the source of data that is directly from your website or from your product listing what's around it. So that sentiment analysis also really helps us create the right context. So if, let's say you want your message to be consistent with the brand voice, the why behind the content that you're putting out there, you want to make sure that that context and why is is loud and clear and consistent across all of your data sources appearing in those AI search orders.

[Tim Peterson]

So these chats aren't private. In that case.

[Alaina McGurn]

We have so much third party data from search platforms directly that tell us the types of questions that are being asked, the volume of interest, the trends around that to a point where we can now predict similar types of Conversations around the trend and also use that data to inform media activations, content production, marketing.

[Tim Peterson]

Okay, so this is all just coming through their APIs. Is that how this. I wasn't even aware that all of this information was good. I'm going to have very different conversations.

[Alaina McGurn]

Yes. And you know what's interesting is that in it works both ways, right? We can see that data, but one of the cool competitive advantages that brands now have is actually providing our information directly to AI platforms again via either direct API, the public facing API, AI platform specific APIs or partner maybe specific APIs. Right? So this is huge for commerce. If let's say you have your product catalog, I know a lot of brands now feed that data directly to retailers, to Amazons, Walmarts, right? Dick's Sporting Goods, whatever the retailer might be that you're working with. A lot of brands are using Shopify, Salsify, those types of data feeds to provide that product data directly to those platforms. We can now do the same thing with AI soon search. So you're starting to see Perplexity coming out with their merchant center. You're starting to see the integration of OpenAI and Shopify, right? Obviously Google Shopping now has a direct feed where they do present that shopping data anytime you are searching for related information. That gives us more tools in our toolbox to provide that product data directly to to the platforms. And right now, while AI is becoming mainstream and becoming table stakes, that's still a huge competitive advantage. We're still seeing a lot of white space. Not a lot of brands are doing this right now. So any effort to provide your data directly to AI platforms is still a huge competitive advantage. We've seen companies that are providing the data directly to to a platform like perplexity instantly enjoying 20, 30% more visibility in those AI responses anytime the product is searched. Anything related to the products that's being researched at that time. And then when it comes to organic feeds to more traditional platforms, right, like Google Gemini overviews in the automotive vertical, we've seen a four digit per percentage increase in both organic traffic as well as revenue sales that then comes from that influx of traffic. Simply by opening up and making your data available to these platforms, they want to index your content, they want to understand it, they want to ingest it so that they as businesses provide a better experience.

[Tim Peterson]

And so how often is this data being uploaded? Because like pricing, we have a running gag on the show about how the price of salmon has changed so much in the past couple months because with ChatGPT, GPT specifically its training data set I think is up to 2024 at this point. I don't know if it's still September 2023, but there's always been a lag between the data that it has available. Are you able to get around that?

[Alaina McGurn]

So again, opening up your data via feed solves that, right? Because then you are feeding the data and becomes a bit more real time. You have the options to update it more effectively, more efficiently. And that goes for both AI platforms as well as how they find your information across other websites as well. So for example, if you have a tomato sauce, you're selling tomato sauce, you might want to create a partner APIs for restaurants websites, right? So then they bring your product data and that then becomes available to AI platforms. And if you're feeding data directly to Perplexity, that's where you have the same level of control as if you were feeding your data directly to Amazon or Walmart. It gets updated, you get more visibility in those product search results, you get better brand associations. And what's interesting is that you're also investing in your products being more shoppable tomorrow. Right, because we're already seeing a lot of these AI search platforms being very competitive with well established existing marketplaces, platforms that we rely on today. So Perplexity Pro is essentially the answer to Prime. You can buy those products right there and then if you're shopping for something and if the data is provided directly to AI platforms and it becomes more shoppable, not just at whatever that final destination is, but in the platform itself.

[Tim Peterson]

Okay. And so we've talked about like how companies are able to provide their own data to these AI chatbots to inform their responses. How do you deal though, with all the other data that's available to these LLMs that can be informing. Like Google obviously has the Reddit deals, so that like really is a factor. But then there's also, I know I've had ChatGPT recommend products to me and this happened recently. I said, I've tried that product. That product actually sucks. And so I don't know if like that now informs the model, but like there's all this data you can't control, a company can't control that is also informing these results.

[Alaina McGurn]

It's fascinating how much I think the world of search has evolved. In a sense, we're becoming more product managers because we're synthesizing and integrating so many different data sets to provide those personalized curated experience to consumers throughout their entire search journey. No matter which platform they're looking for information on. So you're looking at integrating third party data. Ideally, you're integrating your first party data to make that targeting more effective. You're integrating obviously everything you want to communicate from the brand perspective, the desired brand voice, the why behind what you are trying to sell or promote or put out there as a message from the brand perspective. The more of these data sets we're bringing together, the better the message is, the better the experience is for consumer. And obviously AI platforms are incentivized to provide that experience as well. The more we integrate those data sets, the better the outcome.

[Tim Peterson]

And then I guess if you get in that listening data on the back end, that company whose product I said sucked probably found that out. I guess then.

[Kamiko McCoy]

Yeah. I think the last thing that I wanted to ask is kind of feeding off of what you were saying, Tim, is like how, how much data do we have here? Has it. Has search AI become adopted enough that we are able to kind of feed the beast here?

[Alaina McGurn]

You mean providing data directly to AI platforms? Yes, yes. This is where the more data we provide in a way that again is consistent, is helpful to create the right consumer experiences, the better. Again, the goal is the same. AI search platforms want to be there at the right moments in time in a consumer journey. As they're looking for information, they want to curate that answer and provide the best possible experience. Right. So if we're providing the audience data, we're providing the brand data, we're following the topical relevance and capitalizing on the trendiest conversations that we know will result in high impact and also provide consumers what they're interested in today. That's where we can count on the results and moving the needle.

[Tim Peterson]

Unfortunately, we are out of time. But maybe I could throw this clip into Notebook LM and continue the conversation with AI. But Elena, this is.

[Alaina McGurn]

Thanks for having me.

[Tim Peterson]

Thanks for doing this. Thank you so much. Thanks for listening to this episode of the Digiday podcast. If you enjoyed it, please leave us a rating and review on Apple Podcasts, Spotify or wherever you're listening. Get more from Digiday with our daily newsletters sent out each weekday morning. Visit digidaire.comnewsletters to sign up.