The 2026 Performance Model:
Your Playbook for
Marketing Growth

Catch the key takeaways

In our latest webinar, Amsive’s experts unpacked how performance growth is evolving in 2026, and why traditional, channel-led models are no longer enough. As AI becomes increasingly embedded in frameworks and platforms, attribution becomes muddier, and customer journeys are increasingly fragmented, brands need to rethink how performance is designed, measured, and scaled.

Marketing growth hinges on leading with audiences, treating AI as infrastructure, activating first-party data, expanding visibility beyond search engines, and orchestrating efforts across the full lifecycle

Here are the top five takeaways from the conversation:

1. Performance growth starts with audiences, not channels

The most effective performance models aren’t siloed by media channel. They’re organized by audience behavior across the full journey. When audiences become foundational to the strategy, marketers can test incrementality, suppress waste, and understand what actually impacts behavior, rather than what generates clicks.

Leading with audiences shifts teams away from debates over attribution and toward more meaningful insights about impact and lift.

2. AI is infrastructure, not a testing ground

In 2026, AI is moving from being an experimental strategy to a piece of marketing infrastructure. It increasingly embedded into how platforms make decisions, scale media, and optimize outcomes. This opportunity is can lessen the distance between consumer intent and decision.

AI doesn’t fix messy strategy. It scales it. Clear audiences, success metrics, and guardrails are required to maintain control in an AI-driven ecosystem.

3. First-party data is the backbone of performance marketing

First-party data is foundational for scalable growth. High-value consumer intent signals guide platform learning, fuel smarter expansion, and reduce wasted investment.

First-party data gives AI the signal quality it needs to optimize for real business outcomes, rather than surface-level engagement.

4. Search visibility doesn’t start and stop on search engines

While SEO is a cornerstone of visibility, brands need to consider their visibility across AI-driven answer engines and generative platforms. Large language models (LLMs) pull from a brand’s site content, off-site mentions, user-generated content, product data, and brand authority to surface recommendations.

Success in AI search is measured less by clicks and more by a brand’s presence, share of voice, and influence in AI-powered conversations across the funnel. SEO analytics need to evolve to reflect visibility and impact, not just traffic.

5. Orchestration beats isolated optimization

Optimizing individual channels improves parts of the system. Orchestrating outcomes improves the system itself. As consumer journeys fragment across platforms, teams have to coordinate strategy, data, and messaging across disciplines.

Orchestration ensures consistent consumer experiences, stronger performance, and less wasted investment. It’s the defining performance advantage for 2026.

FAQs: The 2026 Performance Model

What does it mean to be audience-led?

Being audience-led means organizing strategy, measurement, and investment around how people behave across the journey, not around individual platforms. It enables better incrementality testing, smarter exclusions, and more accurate assessments of what drives growth.

How should brands think about AI in 2026?

AI should be treated as infrastructure. It’s embedded in platforms by default and scales whatever strategy you put in place. Strong inputs, exclusion guardrails, and human oversight are required to ensure AI optimizes toward meaningful business outcomes.

Why is first-party data critical to a marketing strategy?

First-party data teaches platforms who your best customers are and who should be excluded. It reduces wasted investment, improves targeting, and creates a closed-loop system where performance improves over time instead of becoming noisier.

How is SEO evolving with AI and AEO?

SEO is still a key foundation for search visibility, but success now depends on visibility within AI answers and LLM recommendations. Brands need to move beyond measuring rankings and site traffic to measuring influence, share of voice, and presence in AI answers.

What role does orchestration play in performance growth?

Orchestration aligns teams, channels, and data around shared outcomes. It prevents conflicting messages, eliminates redundant investment, and ensures audiences are engaged consistently across touchpoints. In an AI-influenced marketing environment, orchestration is no longer optional.

Dive into the transcript

The 2026 Marketing Landscape: From Uncertainty to Strategic Opportunity

Michael Coppola (00:17):

Okay. I think we have our attendees rolling in. First, just want to welcome everyone to the 2026. We’re in 2026 performance model playbook for growth. I’m Mike Coppola, CEO of Amsive, and just really appreciate everybody taking the time today. I know calendars are full. There’s a lot going on. We’re at that time of year where we’re ironing out budgets, putting together our playbooks and plans for the year, and really appreciate spending some time. So before I get in, I think reflecting a little bit on last year, it really was a year as marketers. It was very, very unpredictable.

And in many ways, we were on uneven ground as kind of marketers where there were some positives like inflation slowing tariff tensions and political pressures made things definitely more uneven. So when you factor that with the impact of AI and particularly the hype phase of what we’re experiencing with new tools, we almost had this reactionary marketing sort of response.

It’s almost a marketing reaction to decision-making. And all that, when you put it together, if you look at last year, whether brands had good years or bad years for growth as marketers, it really created a very short-term decision-making process in many ways. There were some pockets where there was some strong verticals of growth or brands, but there was a lot of short-term thinking. So when you move into 2026 and we think about where we are this year, it’s definitely feeling like more of a stabilized year, more stabilization.

The Winterberry Group has come out with their prediction in terms of our marketing economy growing about nine and a half percent. That’s strong. Now that includes political and about half of that would be our bump from the midterms this year, but that’s still going to be a five to 6% growth in which outpaces our expected GDP.

So all things point towards a more healthier year, less, I think, reactionary or unpredictability or sort of settled into this new norm. And I think what it brings up is really a little bit more where we could start thinking about mid and long term past the reactionary phase of AI. So really in the moment today, when we put that on balance and we talk to our marketers, they’re being asked a lot of the same thing.

We want you to do more with less, optimize harder, but most of all, prove ROI with confidence. And I think this year gives us the idea to think a little bit broader, think a little bit more long-term. And we know from listening to our marketers and our colleagues and friends, a lot of this might sound familiar, but yeah, the reality is attribution. It’s got a lot of noise.

It’s not becoming clearer. And those lower funnel channels that we’ve relied on, they’re not more efficient. They’re inherently more expensive. We see it in the click costs and the CPMs. And really the importance has sort of risen. And we’re going to talk a lot about that today with SEO and AEO and GEO and AI disruption, but it was hard to justify an in- year investment. How do I justify it? And really, literally, AI is everywhere.

Everywhere we turn, there seems to be another tool, another product, and it’s supposed to make decisions go faster, but in many ways feels slower. And I can tell you if I’m running a performance marketing agency like day-to-day, it’s a lot. The amount of tools that we license, not just for marketing, but operational tools, every single tool’s got a pitch where there’s an embedded AI that needs you to pay a little bit more.

The reality is most of that is going to be foundational to these products in the future. So we’re having to deal with that noise and spending a lot of time making sure that we’re able to evaluate tools. So if the unlock word of the day is, let’s just get some of that time back and put it to marketing. So bottom line is what we hear is there’s definitely movement, people doing things. I think the problem is an effort. The problem is really around how we view this performance model.

So today we’re not going to talk about a new tactic or a new channel. It’s really about the operating model. What really sustains growth in the future? And that’s a model that is one that’s audience-led instead of leading by channel and one that really sees SEO and AEO as foundational and a very holistic opportunity for brands, not just a line item that needs to be checked.

And definitely one that’s really prioritizing orchestration across the business and moving away from isolated optimization. So that’s what we’re calling this 2026 performance model. And today we’re going to break some content on that, talk about what’s no longer working, where leading brands and where we’re seeing success in our portfolio and really how this performance model compounds.

So I’m really excited to have joined me to absolute rockstars. I believe rockstars in the industry, both Lily Ray and Inna Zeyger, really representing two critically important areas of the marketing ecosystem in this performance playbook. So Lily, Inna, thanks for joining. I know you’re excited about this year, but any thoughts upfront, Lily, when you think about the year ahead of us?

SEO Evolves: The Rise of Answer Engine Optimization (AEO)

Lily Ray (06:24):

Yeah, absolutely. And I’m super excited to be here and talk about all things SEO and what we call AEO, so answer engine optimization. Our team has been working really hard over the last couple of years to really pivot into understanding how large language models work, how search surfaces are shifting more towards highlighting AI-generated content and generative content in general.

So we have a lot of new ways of thinking about organic search, especially as it relates to how consumers are really taking in information across these new platforms and how we really drive visibility, awareness, and then measure success with AI search as well. So I’m excited to be here.

Michael Coppola (07:02):

Thanks. And it’s all at the forefront of the conversation, so it’s definitely exciting for that. Inna, what are you looking forward to this year? And I know no snow tomorrow, but…

Inna Zeyger (07:16):

Definitely no snow and definitely not football, but that’s a webinar topic for another time. And I’m really excited to be talking to both of you, some of my favorite people. But what really gets me excited about 2026 is that marketing fundamentals are really at the forefront right now.

Strategy is really leading again and tactics are being evaluated in the context of the bigger picture. So how does what we’re doing in each of these platforms really ladder up to real business outcomes? So it’s not just clicks, impressions, or conversions. And in digital specifically, that shows up how we think across channels. So making decisions that reflect the full customer journey. And I know we’ll get to quite a bit of that in a little bit.

Michael Coppola (08:05):

Yeah, I think you’re right. And for me too, I’d echo that. What I’m excited about is there’s a lot of noise. When you go through this very rapid change period that as performance marketers we’re going through, we have to make that purposeful time. It’s like to be above the tools and the technology, it’s exciting to be a marketer today, more so than ever before. And it’s exciting for Lily in terms of there is no time other than today in the present to be an effective SEO marketer, the attention is there.

So what do we do with it as marketers to maybe make that stronger leap in investment into more of a holistic brand? So yeah, I’m super excited. I’m pumped and let’s get to it. And I think these are kind of what we’re going to talk about. Foundationally today, I think there’s five critical themes here.

One is we’re going to talk about being audience-led. That’s definitely that more holistic marketing thinking and a little bit about where AI has moved more to infrastructure beyond experimentation. Critical importance of first-party data connects to both of those in a very meaningful way, happens to be sitting in the middle of this slide, and then really how this visibility and opportunity extends beyond search and how AI impacts that, and then bringing it all together and the importance of orchestration.

So I think these are our lanes to guide the discussion today and maybe to kick it off, Inna, when we talk about the performance model being audience-led, not particularly channel, I guess what does it actually mean in terms of organizing performance around audiences instead of channels?

The Audience-First Framework: Rethinking Performance Beyond Channels

Inna Zeyger (10:12):

And this is probably one of my favorite topics of discussion. So when we talk about organizing performance around audiences instead of channels, this is really when that model changes, especially when you think about mid-funnel. And just to sort of ground this for a second, when I talk about the mid-funnel here, I’m not talking about things that are very easy to see. So low funnel for most of us is pretty clear for most brands that’s for marketing, it’s branded search, it’s already people looking for you by name.

That stuff tends to perform really well because the intent is already there. So when I say mid-funnel, I’m talking about audiences who aren’t raising their hand yet, people who are problem aware, category aware, maybe comparing options, but not actively searching for you. So that’s the part of the journey that notoriously hard to prove. And I think we’ve all see that.

The reason isn’t that it doesn’t work, it’s just that you can’t really optimize a mid-funnel the way you optimize a channel. So you can’t just turn on and off and say like, “Okay, now I’m doing mid-funnel.” So you actually have to design it. And that only happens when audience behavior, not channel metrics become the organizing principle for us. So for example, if you’re channel led, you end up debating attribution.

If you’re audience-led, you can ask a much better question, which is, does this group of people behave differently because of what we did or would they have done this anyway? And for us, that’s where the incrementality comes in. So audience first gives you the ability to test impact, right? So you can do that through suppression, holdout sequencing. It’s not just reporting on performance. And a really simple way to think about it is if you talk to everyone, you’re going to see results.

But if you intentionally don’t talk to a small portion of that audience and results stay the same, you’ve learned something incredibly important and that’s applicable to a lot of channels. You can’t do that clearly on the channel level, but you can only do it when the audience is a unit of measurement for you.

Michael Coppola (12:23):

I mean, audiences, it’s really the core planning unit, where I think the way we plan before, it’s like we think, all right, we got to do this channel, that channel. But if you start with that audience, now I think I like what you said before, there’s a bigger insight that you’re going to get, right?

Inna Zeyger (12:43):

Yeah. And the other piece is that there’s this other shift today. Optimization in general today is much more about exclusion than inclusion. So this same applies to the audiences. So the biggest gains today from any of the marketing dollars we’re putting behind platforms and channels is actually going to be more about deciding who not to include.

So we’ve always understood this in search like, hey, negative keywords, geographies, but audience strategy can work the same way. And this is where first party data starts to matter. Using it as a control layer, and it’s what lets you enforce those audience decisions across AI-driven systems. So who gets exposed, who gets suppressed, when someone moves forward in your journey and when they don’t need to be chased anymore. So this is where that orchestration really starts to show up more operationally and not just philosophically.

Michael Coppola (13:40):

Yeah. And I think we’re going to talk about both those a little bit more, like the importance of first party data and just how to organize. But an audience-led strategy versus channel, it sort of ensures that regardless of where you are, we’re trying to speak to that customer consistently, persistently across a variety of different formats. Because if you don’t organize that, you might be speaking to one customer and said platform as if they were unaware of the brand and another platform as if they’re already well aware in evaluating features and benefits.

So you start to organize around the customer journey better rather than within channel. Now that doesn’t mean within channel, we aren’t doing things that are very specific to that channel or in channel optimization behaviors. You have to have that, I think what you say is that holistic view if you want to be more successful.

Inna Zeyger (14:46):

Totally.

AI Transitions from Experiment to Essential Infrastructure

Michael Coppola (14:47):

Yeah. And I think where to kind of tie that off, where teams usually struggle, I’d say it’s trying to make that shift because of how they’re even organized internally. One team handling this, another team, that’s why it’s important to put audience at the center works against those silos. So a hundred percent. I think the other thing that we’re seeing is definitely that AI no longer a testing strategy.

If you look at the data here, AI adoption and marketing strategy from 2023 to today, two-thirds of all marketers absolutely see some AI adoption in their strategy, let alone are they using a tactical piece of machine learning or AI. But we definitely moved from this experimentation phase to more of a embedded infrastructure. So I guess, Inna, how do you see that kind of happening in infrastructure right now?

Inna Zeyger (15:56):

Yeah. So big shout out to the webinar we did back in July. For those of you that joined us for that one, some of this might sound a little bit familiar, but a great place to start. I think what’s changed isn’t the idea of AI, it’s just the speed. So back then, just in July, well, six months ago, we were talking a lot about Google’s AI Max and some of their AI-driven products and frameworks we were putting out there to really be able to start to enable that testing.

And at that time, AI Max was rolling out more broadly across accounts and we were very much in that testing mode, figuring out where it worked, where it didn’t, and what guardrails we needed. So fast forward to today, it’s no longer really an option. So that AI Max piece is table stakes, performance max became table stakes, and Microsoft is actively building its own version of the same system.

So the question is, shouldn’t we test AI anymore? It’s how do we operate in a world where AI is already embedded in how those decisions get made? So what’s interesting is in those frameworks we talked about in July, they actually matter more now than less. So back then, testing felt experimental. Now testing is how you maintain control. And AI is incredibly good. I think some of you who have tested some of those tools.

Inna Zeyger (17:22):

It’s really, really good at scaling whatever you put in front of it, and that’s powerful, but it’s also risky. But look, with AI, AI doesn’t really fix the mess, it scales it. So if your audiences are unclear, your success metrics are unclear. If you don’t design for incrementality, AI will optimizeery, very confidently, but not always meaningfully. So this is what-

Michael Coppola (17:44):

Yeah, it’s a great comment. It’ll scale the mess, right? It’ll scale the messy strategy. So I think we understand, all right, AI can be a task helper, but moving it to more of infrastructure, yeah, that’s where you can unlock a lot of time, but if you don’t have the right structure to that, or dare I say the human in the loop or the oversight, things-

Inna Zeyger (18:18):

Or manual automation.

Michael Coppola (18:20):

Yeah. Automating for the sake of automating is not going to provide any marketing value. So I think one of the litmus tests is when you get to it being infrastructure, if there are things that are still optional, that’s not infrastructure. So it’s like as building a really good performance marketing practice when you’re embedding AI and workflows, it needs to be part of the delivery where there’s some really good human oversight or strategy, or it could spin out of control, or if it’s optional, you’re not getting the most out of the scale for that. Yeah, 100%.

And I’d say, Lily, when you think about where it’s embedded, I mean, even think about the solution of SEO and AEO, it’s really emerged. It’s now no longer optional. I guess that was that statement. So it’s actually foundational to how you need to think of things. So AI has really moved as far as from a product set into really a different place today. So how would you speak to that?

Lily Ray (19:34):

Yeah, definitely. And I was really excited that our team was able to launch some new offerings around AEO early last year, so answer engine optimization. The way we see it and the way we’ve really been trying to get to the understanding of how AI search is fundamentally different than SEO, we put out this Venn diagram to help visualize a lot of those changes. One thing that’s important to understand about AI search in general is that there’s absolutely a lot of overlap with SEO from a tactical perspective. And the reason for that is because large language models like ChatGPT and Gemini still use web search.

They still go to search engines to get a lot of up-to-date factual information. So in that sense, SEO fundamentals are still essential. There’s been a lot of talk around, is SEO still important? And what we’re seeing in the data and the way these things work is that it’s absolutely as important as ever.

It really helps to position yourself within answer engines. But as far as looking at the far right side of what our AEO offering looks like, there’s been a lot of evolutions as far as how brands and companies appear in AI search, very fundamentally different than the traditional ranking and clicks and backlinks and these kind of traditional SEO factors that we’ve thought about for a very long time.

So now we’re thinking about visibility within different large language models, thinking about your share of voice compared to competitors. Is your website, is your brand appearing in these answers? So we’re using different tools to surface that information. And then also just thinking about the fundamental nature of how people use large language models and AI search is very different than traditional keyword searching in search engines.

So a lot of the way that we’re pivoting is to understand these more conversational prompts and even out loud conversations that people are having with large language models, and how can we ensure that for all the very granular questions that they’re asking at different stages of the funnel, how can we make sure that we have the best content to answer those questions both on our site as well as on external sites as well?

Michael Coppola (21:36):

Yeah. And it’s still a owned content versus influenced content play. So I think we’re going to talk a lot more about how do we win in it and what are the things we need to do. But the idea that this is a more holistic solution, it’s no longer SEO, it’s SEO, AEO, and it’s more multidisciplined right now. And I think you’ve been doing this a long time. Years ago, the nomenclature in the industry has tried to change. Is it SEO? Are we doing visibility management? Are we doing these terms? And it was always getting at the same thing, is that it isn’t always just about what’s going on in the rankings within search.

It’s really about your overall brand presence and your visibility. And I think this has been a good moment where that’s getting the most awareness that it’s ever been. So the seat at the table that SEO has to really should have in an organization isn’t in this very small silo anymore.

It should be across multiple disciplines, and I think it’s just a great opportunity and you’ve heard it all. So I think in terms of this AI infrastructure, just to double tap a little bit on this, it’s very clear that AI is giving scale to what we would normally do on a task level or an operational level, but it’s the assisted decisioning. That’s a tremendous value, and I think that’s a huge shift that it isn’t just about executing tasks faster.

It’s really, I think, shortening the distance between the signal or the behavior and the decision that I need to make. And that’s what is exciting about marketers. It’s all this embedded AI within the tools, not just the generative AI assistant or thinking partner that you have. It’s how it actually allows us to make decisions much faster in the day-to-day to get back above the business, what Ina was talking about earlier on, being excited about being a marketer. So it stops being tools that people try and start to be process and behavior.

So I’m definitely excited about how AI can move us above the business as marketers. But there’s one fundamental very clear challenge sometimes, bad data and bad data out. But when we think about first-party data in a broader sense, not just for AI foundation for performance, it’s a big part of the playbook. So I guess, Inna, why obviously we see some performance lifts here. We know when we include really good data in our audience models, we’ll see 30 to 40% lift brands stop vying for potential customers that are no longer … There’s a lot that goes into it, but I guess why is it the foundation for the playbook?

First-Party Data: The Foundation That Powers Everything

Inna Zeyger (25:11):

And this is a big one because this is something that’s accessible to most brands because they have their own data. It’s available to them. But the difference is this is where first party data stops being abstract. It’s not just a thing that exists and has to start becoming very practical. So at a high level, I want to talk about first party data seating. It’s about using what you already know to control how you grow next.

So instead of starting with channels, which has historically been a big way that advertisers focus on, whether it’s Google or Meta or broad targeting, you start with your best customers. So the people already showing you through real behavior, what value actually looks like for your business, and then that data becomes the seed. So when I say seating, I don’t mean just uploading a list into Google or Meta or any other platform.

I mean, using high intent segments. So based on lifecycle, stage, frequency, value, or engagement to teach platforms, who they should go find more of. And at the same time, first-party data is also how you optimize through exclusion. It lets you filter out users who’ve already converted, you suppress audiences that aren’t incremental and avoid chasing people who look good on paper who don’t actually move the business.

So instead of trying to include everyone who might be interested, you start with the people that you know matter and let the platforms expand from that signal. So from that starting point with your data and uploading and identifying your high value signals, platforms and audience tools can build lookalikes and predictive models. That expansion happens and guardrails defined by your data. And then you can overlay third party or additional contextual signals after the fact, after that sort of backbone is set.

And so then you’re creating this sort of closed loop model, right? So as new customers come in, their behavior feeds back into the system and then strengthens that seed over time. So that’s where the media actually gets smarter instead of noisier. And this is, by the way, very flexible by industry. For some examples, right?

For eComm, the seed is often like high LTV or VP purchasers. And then if you do any sort of incrementality studies, because again, your audience first, it shows you which channels are creating that demand versus just harvesting it for B2B or SaaS, it’s customers who onboard successfully or expand usage, not just leads or-

Michael Coppola (27:45):

Or show up to an appointment.

Inna Zeyger (27:47):

Versus. Exactly. Or for heavily regulated businesses too, it’s their lifecycle stage and frequency, which lets you suppress some of that waste and overspending on people who have converted anyway. So it’s different verticals, but the same principle, like that first-party data is your backbone. AI is the accelerator of it and incrementality is how you know it’s actually working.

Michael Coppola (28:15):

And you mentioned the guardrails, right? There’s just this very…I think there’s two things you said in there. One is if I repackage AI, excuse me, the first party data, it’s kind of foundational for the AI to learn from really well, but it’s also, it’s our guardrails for those AIs. And I think when you’re focused on each individual channel and you’re not audience-led, it’s so easy to not do things like exclude audiences or include or set those guardrails because you are just chasing that performance metric.

I know we looked at potential, not a client today, but someone that we were looking at was having a lot of success absolutely because of the creative and the product. But when you started to get into the actual performance metrics, you saw that there was no segmentation. So all these metrics were getting cluttered by repeat purchasers and really what was driving net new ads.

Then you start getting purposeful around, what’s my next strategy? So I think it comes back to, to your point, it’s foundational and it’s also like how you set some guardrails for sure. And the regulated industries, we happen to work with many, it’s sort of quietly ahead of it on a few ways, because when you have a business that’s built on consent and trust and identity, you start to structure or manage your data internally in a way that allows that ecosystem channels or partners to start accessing it and using it.

So in many ways, they’re a little bit ahead, maybe still restricted in how you can use that data, but there’s at least those guardrails that you can draw from. So if anybody is not starting with their data or doing basic segmentation, it’s a huge value driver for performance. Isolate your customers from as simple as that.

So a lot to do there. So Lily, just getting back to SEO and AEO, this visibility beyond search engines, but I guess how does that actually happen in AI? Talk a little bit about how the search behavior is.

Measuring What Matters in Search: New Metrics for AI-Driven Visibility

Lily Ray (31:16):

Yeah. Well, I think a lot of people, as they’ve realized that large language models like ChatGPT are increasingly effective at answering questions throughout all stages of the consumer search journey and the funnel essentially. Whereas before, maybe it was a lot more upper funnel information seeking type questions. Now we’re seeing that AI engines like ChatGPT and Gemini and Perplexity are doing a much better job of surfacing products, transactional recommendations.

And so a lot of people are staying within that ecosystem to have these full funnel conversations and to get these questions answered. So we’re seeing a lot more people getting product and brand recommendations directly from ChatGPT, for example. And it’s important to remember, I share this statistic a lot of the time when I give talks, while large language model usage is absolutely adoption is growing in usage rapidly, we’re also seeing that people continue to use search engines like Google.

And actually Google’s shared some data around the fact that these AI tools are actually additive. They’re almost expansionary on top of traditional search. And not to mention that Google, for example, and other search engines are now embedding AI answers directly on the search results. So that’s another way that people are engaging with AI. So I think the future, what we’re projecting out to see over the next few years is more and more of these not only discovery-oriented conversations, but very much transactional conversations will take place directly in AI platforms.

Michael Coppola (32:45):

Yeah. And I think it’s also just changed in terms of you’re less focused on, you’re getting the answers and you’re getting the recommendations before the interaction in terms of the click. So I think the way you have to think about success is very different. Talk a little bit about that.

Lily Ray (33:13):

Yeah, absolutely. I think in the past with SEO, we were very used to measuring performance based on things like rankings and search engines, share voice within search engines, and then of course traffic and clicks. It was very measurable, it was much easier, especially when we get the information directly in places like Google Search Console and Bing Webmaster tools, we’re able to easily measure the performance of our efforts in terms of traffic. What’s really tricky now is number one, we don’t get that level of granularity from Google as it stands as it relates to AI products.

We don’t know how AIO reviews are sending traffic. We don’t know how AI mode is sending traffic. There’s a lot of challenges around tracking actual traffic and visibility from things like ChatGPT. And what makes it even trickier is that a lot of these tools are basically designed to not send traffic.

A lot of them are designed to keep the user in the interface answering those questions, but that doesn’t mean that being there’s not important. It just means that measuring it presents a whole set of new challenges. So one thing that we’re really interested in doing with AEO is emulating what are users seeing when they’re having these conversations, which brands and products are being recommended, and how can we measure how visible a certain brand is compared to competitors across different AI surfaces?

Michael Coppola (34:23):

Yeah. So I mean, clearly measurement’s got to go beyond just click counts and post-click conversion. But practically, what does that look like? Because you need a different set of metrics.

Lily Ray (34:37):

Yep, exactly. So we are redefining metrics as we know them in the AEO space. I mentioned before, we were much more reliant on rankings and clicks. Now we have different tools that we’re using. For example, we have Profound, we have Semrush, we have some new AI features within our tools. And what they do is, again, they emulate lots of different prompts that are relevant to our clients and to see how frequently our clients and brands are being mentioned relative to competitors and quantifying that. That’s one way of doing it, but also just looking at how frequently large language models are visiting our websites, which pages are they visiting.

So we’re working with our web development team to put together some reporting about log file analysis. So basically redefining the experience that users are having because now it’s not just about how much traffic is coming to our site, it’s really about how embedded we are in the conversations in these different LLMs. So just being able to report back on that as much as possible is what we’re really working on.

Michael Coppola (35:31):

And it’s always, I think SEOs have always had to be really good at presenting business cases when it comes to budgets at clients and internally marketers. I know it’s important, but I can’t unlock the funding. So now with all this disruption, it’s a big moment, but if I kind of read between lines, if you’re thinking it’s click growth alone, that’s not going to unlock the investment. It needs to be a broader set of metrics. It’s kind of what I’m hearing you say.

Lily Ray (36:12):

Yeah. I mean, one thing that’s important to note is there’s been a lot of new studies in our industry. And one thing we’re absolutely seeing across the board is ChatGPT, for example, does not send a lot of traffic. If you look at the metrics and pretty much any analytics account, you’re going to see ChatGPT anywhere from 0.1% to maybe 2% of your overall website traffic compared to something like Google Organic that might be 30, 40, 50%, but that’s not the only thing we need to think about. We need to think about what are these conversations looking like in something like ChatGPT?

For example, if ChatGPT recommends your product or recommends your brand and then somebody goes to Google and searches for it, that’s absolutely still an important consideration. So a lot more conversions and those deeper funnel, lower funnel conversations are happening directly in AI platforms, even if we don’t see the traffic account for that.

Michael Coppola (36:58):

100%. I think we got a topic here is winning. If we go back to the previous slide, how do you win though? I guess how are we going to be successful? I think the measurement, like we said, has changed, but how are we going to be successful in terms of building that larger influence rather than click traffic stream?

Winning in AI Search: Building Influence Across Platforms

Lily Ray (37:29):

Yeah, I mean the prioritization of what we work on as organic search marketers is changing. I would say we’re much more integrated now with our social media and influencer and content teams, for example, because a lot of how you win is being the most mentioned and recommended brand across all the different places where it matters.

A lot of that comes from social media, a lot of that comes from forums and from what people are talking about online. So it’s not just about what you’re doing on your website, it’s really becoming increasingly important what people are saying about you off your website as well.

Michael Coppola (38:00):

And in our conversation separately, you mentioned just the need for that to be sort of holistic because you could very easily say, we now need more digital PR and we need more social, but everybody still needs to organize around what a common goal is. And that’s something that you’ve pitched and embedded in all of our team here is it’s really having that common goal of here’s what that influence needs to look like. So you’re working across tons of teams now in brands.

Lily Ray (38:40):

Yeah. I mean, I would say with AEO, a lot of what we’re doing is identifying the opportunities, using data to understand, for example, where are all the relevant publications where this brand is being, or its competitors are being recommended and we are not there.

And then it becomes a conversation with our social team, with our PR team to say, “We know that it’s really important. We need to be recommended here or we need to take ownership of this conversation that’s happening about our brand and chime into the conversation.” So it’s identifying those opportunities and then working cross-channel to have those conversations.

Michael Coppola (39:11):

It’s like bringing that opportunity and saying, “Hey, we need a better Reddit presence, but specifically with these types of goals in mind. So how do we create a common work plan to get us there?” And I think the role that SEOs by themselves in is then owning the metrics and the orchestration of that objective. So it’s just a great time to be in the SEO space and I think we’re still going to trip over AEO, GEO. Call it what it is. It’s really brand presence visibility, so 100%. And also AI conversations increasing or becoming shoppable. We’re seeing the ads, Inna, give us a take in what you see.

The Early Days of Shoppable AI: What Marketers Need to Know Now

Inna Zeyger (40:05):

Yeah, I think we internally have an almost twice weekly update going on in the team to see if there’s been any progress or momentum with ChatGPT or Perplexity or any of these platforms of actually creating shoppable ads there. So it’s been a big topic of conversation.

And I think this really glad that Lily talked about what she did about AEO, GEO just ahead of that because what we’re clearly moving towards with AI conversations becoming transactional, but it’s really important, honest to talk about where we are. So it’s still something very early and the experiences are very uneven. I think Mike, you and I actually had a whole conversation about not the best experience of you getting some results.

Michael Coppola (40:56):

Not so good sometimes.

Inna Zeyger (40:59):

Yeah. It was a little rough. We won’t get too deep into it, but-

Michael Coppola (41:04):

Yeah. And I think it’s like where, hey, you’re bringing in all this shoppable imagery from your search index. Great. Whoa, where those ads come from because that’s not the experience I want. So it’s a early ship leaving the harbor, as I say.

Inna Zeyger (41:21):

Yeah. And I would say where I think advertisers and brands should be focused on is the data integrity and the quality and the information they have on their site. So what’s really changing and what will I believe eventually happen is that it’s not just like having ads in those AI conversations. It’s again, going back to that decision making process getting compressed.

So AI is already recommending, narrowing, and this is sort of the last stage of eventually transacting, which shifts the competition from who bids the most to who the system trusts enough to suggest. Trust enough. So it’s based on two things. It’s like all the AEO/GEO work that Lily’s talking about, but also from our perspective, do you have product feeds, having clean product data, data integrity, consistent signals.

Michael Coppola (42:18):

And that’s not easy, right? No. Yeah. Seemingly it seems easy, product feed from an eComm site, but it can be … So you’re saying it’s definitely a place to start. Get those trust signals up and get your content in a much better place to allow you to be more favored when we do get to this more shoppable commerce part.

Inna Zeyger (42:47):

Yeah. Get the SEO, AEO, GEO, whatever we want to call it. I mean, I think maybe the closest approximation to think about it is something like with Amazon where if you’re advertising products on Amazon, you’re doing sponsored product ads or sponsored display ads, whatever it might be, you’re targeting keywords, sure, but a lot of the algorithm also relies on how well optimized your product pages are and your brand pages are. And that’s really more of the start-

Michael Coppola (43:18):

Because they want to make the sale, right?

Inna Zeyger (43:20):

Yeah. And especially if you’re shifting away from manual bidding in those platforms and really focusing on automated category. So think of it the same way, but ultimately with any sort of product feed outside of just any other sort of optimizations to make you more visible within AI, it would be just make sure your data’s right.

Michael Coppola (43:45):

Yeah. Yeah. It’s like data’s foundational. And I think the headline is it’s early still. There’s always this idea of first mover advantage, and we will test and try things as they come, but I think a lot of the trust signals are going to play in your content, how it’s organized is going to play bigger role. I know when I love to cook and I find it a hobby to build recipes and ChatGPT drives my brother nuts because he doesn’t like that, but I fact check everything by doing a Google search and saying, “Are you sure it’s a quarter cup of sugar that goes into that? “

You can go horribly wrong. So there’s still this early territory of fully in paid signals feels a little aggressive today, could change in six months from now, for sure. So bring it back to, I think a common theme is whether it was with the AEO and SEO work that really needs to orchestrate across teams and disciplines to how we think about an audience-led strategy, orchestration really is a foundational for this next year’s playbook and plan.

So maybe for both of you, maybe start with Inna, how does it contribute to performance in 2026 and why is it so important?

Orchestration: The Key to Scalable Performance Growth

Inna Zeyger (45:28):

Yeah, and I’ll try to keep this one short because we’ve talked quite a bit about it, but when you put all of it together, so audience first design, incrementality, first party data, and then AI, orchestration stops being theoretical. So we talk so much about optimization in our space, but I think the distinction is, so optimization asks, “How do I get more out of this channel?”

Whereas when you start shifting towards orchestration, orchestration asks what actually moved this audience and what didn’t. So I guess the way to look at it is when you look at optimization, optimization improves parts and orchestration really just improves systems. And that’s what you’re trying to create as a marketer rather than something that lives in a silo. And so because AI has become a big default execution layer, we have to start thinking systematically and it’s not optional anymore. So that for me is the performance advantage in 2026.

Michael Coppola (46:33):

Yeah, the technology enables orchestration, but it doesn’t create it. The people need to pull that together. And I think fundamentally being audience-led forces some of those silos. I think there was always a debate in how you organize marketing teams and/or agencies around specialists in this area, especially you have this channel here, that channel there.

And as much as there were gains from that system, I think the market’s been pretty clear that that really fights against what really is driving performance. We need more holistic ownership across the system, and it definitely will unlock a lot of performance. Lily, any thoughts on this in terms of orchestration and its importance?

Lily Ray (47:33):

Yeah, I mean, I think for a long time, performance marketing was about optimizing individual channels. So SEO teams were hyper-focused on rankings and email teams were hyper-focused on open rates, for example, but now it’s like the consumer journey doesn’t really only live in one channel. Optimizing for one channel ignores how consumer decisions are actually made nowadays. So we talked before about needing to work cross-channel, and I think that that’s just becoming increasingly important because people are discovering brands across many different points of the search journey, across many different channels and platforms.

And we’re definitely finding with our AEO offerings, we’re working much more cross-departmental than we used to be just because we can’t think in isolation. It’s not just about one specific place where people are learning about our clients. There’s many different stages of that process. And so we need to make sure that we’re working with not only content teams, but also paid teams to get a lot of search query behavior analytics and data because we don’t necessarily have that as much on the organic side.

And we’re also working with influencer and PR teams to get our message out there more. So I think orchestrations become increasingly important.

The 2026 Performance Playbook: Putting It All Together

Michael Coppola (48:42):

Yeah. And definitely, and you covered that a lot, but to be effective in the role and where SEO and AEO is going, it’s forcing this ability to collaborate and coordinate. Doesn’t mean we always have to own everything, but we have to own the orchestration for the goals of that. It wasn’t long ago where SEOs were trying to get a seat at the table at their early part of building a website.

Now everybody knows that or most people … It’s like you would be, of course, but there was a fight to be part of that conversation. And it feels like, I wouldn’t say it’s gloves off now, it’s different, but it’s a similar struggle to ensure that that happens. And I think a lot of what’s supportive of that is making sure that there’s leadership alignment.

Your partners out there, your agencies, your external partners need to not work and feed them, but they should be put together with separate conversations. But it’s definitely from leadership alignment to, I would also say cultural accountability. We’re no longer defined in this age of AI-driven marketing of, at least this is more definitely a personal thing that I’ve been speaking about. You’re no longer defined by how many people report to you and what your span of control is.

And it’s more about how you affect the outcomes. And so AEO, role of AEO affects a tremendous outcome, but touches a lot of things. And for Inna, the performance part, it’s operating and having different channels operate separately and be measured differently. It’s not going to unlock that next level of performance. It has to orchestrate together or we’re going to waste a lot of money. Our consumers, like Lily said, it’s a complex journey.

They’re in many different places and many of them similar. It’s a Meta user versus a Reddit user. Many of them overlap in both. So we want to speak to them consistently, persistently. So orchestration, it makes the individual channels performance better, done right, but it’s definitely not easy. It’s hard, and it’s a skill that marketing teams need to learn, and part of it is getting that alignment up top, but breaking down features and making sure that people are rallied around common goals. So huge, huge ability for it to work well. So to wrap up playbooks for really the year, what did we talk about?

Definitely leading with audiences starts to break down those silos right away fundamentally, and also is where our clients and our potential prospects out there, they want to be spoken to like that. BDAI is infrastructure, not just task enablement, and absolutely get the first party data. And if we are not thinking in our evolution of SEO, if we’re thinking still just site clicks and site performance behaviors, it’s beyond that.

It’s more holistic. And in order to do it, orchestrate around the full lifecycle of either your AEO goals or your performance marketing objectives, I think these are the foundations to be really successful next year. Any closing thoughts, Inna or Lily? I think we got a couple of questions.

Staying Nimble in a Fast-Moving Landscape

Lily Ray (52:46):

I mean, I would say as far as this whole space, things are moving really, really fast, obviously. Things are changing every day. We don’t always have the answers because things are moving so quickly. So for example, in the eCommerce space, now we have a whole new set of considerations from Google. Google just announced what’s called the Universal Commerce Protocol. So there’s different types of feeds that e-commerce sites need to start thinking about.

We didn’t even know about that two weeks ago. So things are just moving so quickly. So I do think it’s important to stay nimble because every day we have new developments, new things we need to action on. We also don’t know how the lay of the land is going to look as far as which LLMs people are using in six months or a year. So just stay on your toes because what’s true today might not be true in six months.

Q&A

Michael Coppola (53:32):

Absolutely. I got a couple of questions. One on AEO, it’s a very competitive market out there. I’m going to summarize this question where our business financial service competing against some much, much larger institutions where they have more scale. So for Lily, if I shorten that too, if you had to prioritize the most important things as it relates to content or strategy to be successful, what would those things be? And particularly thinking that through someone that might not be the market leader more of a challenger brand or mid-size. Yeah.

Lily Ray (54:28):

I mean, I noticed this question talks about if you are a community bank serving three counties in Illinois, how do you compete with the big players? So what’s interesting and actually useful about the way that LLMs like ChatGPT are trending over time and also Google’s AI products is that they’re becoming increasingly personalized even when you don’t ask for a specific location or you don’t ask for maybe a specific need that you have because of the context and the memory that these large language models have about you, generally speaking, they’re becoming much more personalized.

So that includes location and that includes the customer’s specific needs. So by answering specific questions that your ideal customer has as clearly as possible on your website and all the different considerations about your products, everything that makes you unique, who’s your leadership team, why should people trust you?

Having all of that laid out very clearly can help encourage LLMs to pick you when they know that somebody’s really in market for your specific offerings in that particular micro geography. Yeah, especially for a location. So the fact that you are a community bank with three counties is actually helpful because somebody who’s really a potential customer for you in that region is much more likely to see you as a recommended bank.

Michael Coppola (55:44):

Yeah. Maybe getting into the national mortgage market, not so much, but when it comes to local, and I’d say Lilly, also the influence that the local graph, and I’d say particularly with Google’s AI, but that local graph and local listing citations is becoming even more critical and even more regionalized, which could be a whole other webinar, but that’s a huge place of focus.

Lily Ray (56:14):

Yeah. If you ask ChatGPT, assuming you’re logged in and you have memory turned on, if you ask ChatGPT a generic question, in many cases, it’s going to answer by saying, these are your options in Brooklyn, New York, even if you didn’t ask about Brooklyn. So I think that’s a trend that we’ll see being more and more true across AI search over time.

Michael Coppola (56:32):

Yeah. And one more question, we’re coming up on the hour. Ina first-party data, I understand it’s critical, but ours is not perfect. What’s the first place somebody should start so they can get something actionable and some performance improvements?

Inna Zeyger (56:59):

Yeah. And I’ll try to keep this on brief because I think we get stuck in analysis paralysis often with, is this great? Is this perfect? And I would just say don’t wait for perfect data. Start with things that reduce waste. So anything like, for example, as simple as if you have a branded keyword search active and you know that you have a retention strategy in place, maybe exclude people who are visiting, who are searching, who’ve already engaged with you in some way, upload that data.

Or if we’re sequencing, so people have engaged with maybe a certain promotion if you’re eComm, maybe put them on a different audience list and serve them with something different. So these are all forms of exclusion and often exclusion is the easiest way to start using first data.

Michael Coppola (57:57):

I think that’s a great suggestion. And we’ve talked about this a lot, getting to do audits and things of other accounts. And you’re shocked at the excluding existing customers or excluding not past customers with low net promoter score. Unless there’s a conscious win back strategy, why would we want to continue to serve them advertising?

So these are basic marketing principles, but the idea of just taking that list and feeding it to all the different channels and platforms reduces waste overnight. So there are some small motions that start with customer, not customer, or bad prior customer. It can make a big difference before we get the really good segmentation or building models off of who we’ve understood as our best LTV-based customers for this particular product set over time.

Inna Zeyger (59:02):

Yeah. So usable data today will beat out any imperfect data that just never gets activated.

Michael Coppola (59:11):

I love it. Great quote to close our webinar. So appreciate everybody spending the time. We are available. If there was one or two more questions that came in we didn’t get to, happy to follow up, but there’s a QR code here. Reach out. We’d love to talk with whatever marketers have challenges, whether it’s AEO, orchestration, we just want to provide some value. So appreciate everybody spending the hour. Hopefully you got something out of it and talk to you soon. Inna, Lily, thank you.

Inna Zeyger (59:50):

Thank you all so much.

We’re living in a choice overload economy. Learn how you can cut through the noise and earn meaningful attention in our recent webinar with thinqinsights, or let’s talk about how Amsive can help you future-proof your marketing strategy.