Performance Max Optimization: Structure, Signals & Intent Done Right

Episode Description

Frederick Vallaeys, CEO & Co-Founder of Optmyzr, sits down with Bia Camargo, Head of Google at Ecom Nation, to discuss something many advertisers overlook: most Google Ads accounts aren’t ready for AI.

Performance Max isn’t magic. Smart bidding isn’t broken. And the real problem is usually messy conversion tracking, overcomplicated structure, or unclear intent signals that prevent the algorithm from doing its job.

In this episode, Bia walks through her comprehensive framework for making Google Ads accounts AI-ready, a framework she’s already successfully applied to dozens of e-commerce businesses. She explains how to build what she calls “the house”: starting with a solid foundation of conversion tracking, adding proper wiring through audience signals and first-party data, and then constructing the right campaign structure on top.

Bia learned from working alongside Mike Rhodes at WebSavvy in Australia, where she discovered that knowing which buttons to push matters far less than understanding why you’re pushing them. That mindset shift, from tactical execution to strategic thinking runs throughout her entire approach to account optimization.

Here’s what you’ll learn:

• Most Google Ads accounts aren’t ready for AI
• Why conversion tracking cleanup gives the algorithm clear signals
• How to use audiences as intent signals, not restrictions
• Why your Google Ads structure should follow your website structure
• How pushing assets to “excellent” actually moves results
• How AI automates the entire framework with one prompt
• Why understanding intent matters more than keywords
• How to break down silos between PPC, SEO, and other channels
• Why stopping to think before acting saves time

Episode Takeaways

Bia Camargo has spent 12 years building her career in digital marketing, starting from the bottom when she moved from Brazil to Australia. But everything changed when she joined Mike Rhodes’ agency, WebSavvy.

Working with Mike taught her something more valuable than platform mechanics. He taught her how to think like Google, how to understand the why behind every decision, not just the what. That foundation shapes everything about Bia’s current work as Head of Google at Ecom Nation, where she’s developed a systematic framework for making accounts AI-ready in 2026.

The framework treats account optimization like building a house: foundation first, then wiring, then structure. Skip the foundation, and everything built on top becomes unstable. Rush to advanced tactics without cleaning up the basics, and the AI has nothing solid to work with.

Throughout the conversation, Bia returns to a core theme: AI isn’t the problem. The accounts feeding it bad data are the problem. Clean that up, simplify the structure, and give the algorithm clear intent signals, and suddenly, automation starts performing the way it should.

Most Google Ads accounts aren’t ready for AI

When Bia audits a new account, she repeatedly sees the same problems. Conversion tracking is a mess. Campaign structure is overcomplicated. Audiences exist but aren’t applied. And the foundation is cracked before anyone even thinks about Performance Max or smart bidding.

“Even in 2026, people are still having 20 conversion actions, and they are all misaligned, and they are not giving you the right signals,” Bia explained. “This is the first thing we do at the framework. We review your structure, your basic structure.”

The AI needs clean, clear signals to optimize effectively. When you have three primary conversion actions for purchases, you’re at minimum double-counting results. When you have page views that aren’t applied to any campaigns, you’re cluttering the signal without adding value. These aren’t advanced problems. They’re housekeeping.

But housekeeping matters more than ever because machine learning optimization depends entirely on the quality of data you feed it. Garbage in, garbage out; except now the garbage goes into an algorithm that makes thousands of micro-decisions per day based on that bad data.

Bia’s first step in her framework is always conversion action cleanup. Make sure the Google Tag or Tag Manager is working perfectly, eliminate duplicate tracking, set clear primary actions, and use secondary conversions for comparison, not optimization.

Why conversion tracking cleanup gives the algorithm clear signals

The goal isn’t just to have conversions tracked. The goal is to have them tracked in a way that tells the algorithm exactly what success looks like for this specific business.

Bia finds clients with conversion actions coming from Shopify, Google Analytics 4, and custom implementations, all counting the same purchase event. “I say, ‘Oh, you are at least double-counting your results. So let’s clean this up.’”

Her preferred setup: use the Google tag sitewide, then connect the native platform (Shopify, for example) directly to Google Ads as the primary conversion source. Use other sources as secondary for comparison and validation, but make sure they’re using the same settings so discrepancies become visible.

This matters because the algorithm uses conversion data to understand what’s working. If one campaign shows 100 conversions but another source says 150, which signal should it trust? Clean data removes that ambiguity.

Fred asked about the challenge of perfect conversion tracking in an age of privacy restrictions. Bia’s answer was pragmatic: do the best you can with what’s available, but prioritize clarity over perfection. A slightly undercounted but consistent signal beats a theoretically complete but contradictory signal.

She also addressed profit versus revenue optimization. When clients have profit data available through tools like Profitpick, Bia optimizes toward profit.

“I always try to drive to the profit side, even in the campaign structures like most profit products, less profit products,” Bia explains.

But if that data isn’t reliably available, she lets the algorithm follow what the website sends rather than applying fixed values based on guesswork. The risk of feeding the AI wrong numbers outweighs the benefit of theoretical profit optimization.

How to use audiences as intent signals, not restrictions

One of the biggest misconceptions about Performance Max is that audiences restrict who sees your ads. That’s not how they work. Audiences in PMax are signals that guide the algorithm toward certain behaviors without creating hard boundaries.

“I don’t think the audiences are limiting PMax. I’m using them as a guide for PMax,” Bia said. “What I do is a mix of affinity, in-market, and more intent-related. So people who browse that website that I want, people who browse a competitor website, or an interest website.”

The key is building audiences around intent. What question is this campaign trying to answer for searchers? Who is actively looking for the solution this product provides? That’s different from traditional demographic targeting.

Bia starts by cleaning up existing audience lists. Many accounts have 300 audience segments that are outdated, too small to be useful, or simply never applied to campaigns. Cleaning that up creates space to build audiences that actually matter.

Then she creates audiences tailored to each campaign’s specific intent. Not generic “80-day website visitors” applied everywhere, but audiences that answer: who is searching for this specific solution right now?

She combines these intent-based audiences with search themes to guide the algorithm in the right direction. The creative assets then need to match that intent. If the audience is “people looking for shampoo solutions for dry hair,” the asset group should have a dedicated creative and copy for that category.

This is the shift Bia keeps emphasizing: from keyword-driven campaigns to intent-driven campaigns. The work isn’t about building massive keyword lists anymore. It’s about understanding search intent and guiding the machine toward the right people.

Why your Google Ads structure should follow your website structure

When Bia tells clients to simplify their account structure, she gets pushback. People are used to the old SKAG (Single Keyword Ad Group) model, where granularity meant control. Hundreds of ad groups with thousands of keywords felt like optimization.

“What I’m finding the most is that sometimes people are overcomplicating things,” Bia explained. “Your Google account structure should follow your website structure. So the way you organize your website is the way you can organize your account structure.”

If you’re an e-commerce brand selling beauty products organized by category, like shampoo, conditioner, and styling products, your Performance Max asset groups should mirror that. Not 12 asset groups for variations of the same product when you don’t have the budget to support them.

The mistake Bia sees constantly: three asset groups with identical assets, differentiated only by audience signals. That’s not strategic segmentation. That’s just fragmentation.

Her structure recommendation is straightforward. Build the full funnel: Demand Gen and YouTube for top-of-funnel, non-brand search for middle, Performance Max shopping and brand for bottom. Not every client has a budget for all three layers, but the structure should be ready when they’re ready to scale.

Within campaigns, consolidate ruthlessly. If multiple ad groups point to the same landing page with the same themes, they belong in the same ad group. “You don’t need 300 ad groups anymore.”

Fred agreed, pointing out that this simplification isn’t about being lazy. It’s about giving the algorithm sufficient signal volume to learn effectively. Spread your budget across 300 micro-segments, and none of them gets enough data to optimize. Consolidate to 15 well-defined segments, and the AI has room to work.

How pushing assets to “excellent” actually moves results

Bia has a reputation in her team as the “Marie Kondo of Google Ads” because of her obsession with asset ratings. She pushes every asset group to “excellent” status—maximum headlines, descriptions, images, videos—and sees proven results every time.

“I always push all assets to excellence. People say that I’m a little bit crazy about that,” she admitted. “But actually I can see proven results every time we push asset groups to excellent, our ads to excellent, because it’s actually the easiest way to do what Google is asking from us without damaging our campaigns.”

This is important because Google’s recommendations vary wildly in quality. Some hurt performance. But the asset excellence recommendation consistently helps because it gives the algorithm more raw material to test and optimize.

Bia doesn’t just add assets blindly, though. She continuously reviews performance. Which creatives are actually working? Which ones are dead weight? The goal isn’t 15 excellent headlines for the sake of hitting “excellent.” It’s learning from what performs and applying those insights elsewhere.

She also watches Performance Max search terms religiously. “It’s not one thing set and done. You need to check PMax search terms at least on a weekly basis because this is our work now,” she remarked.

The work isn’t building massive keyword lists. The work is giving the machine the freedom to explore, then going into the backend to manage placement exclusions and negative search terms, and ensuring the algorithm goes to the right places. Even if you’re adding brand terms to Performance Max, you need to monitor how they’re performing—whether they’re getting all the attention or none.

How AI automates the entire framework with one prompt

Bia’s framework works well for small to medium businesses. She can apply it quickly and see results from day one.

“Since I started at Ecom Nation, we already did five or six account restructures, and from the day we did, we saw the uplift. I never had a case where the account got worse after applying those.”

But she hit a wall with larger, more complex accounts. How could she scale the framework to massive accounts with hundreds of campaigns without spending weeks on manual analysis?

The answer came from working with Mike Rhodes and his 8020 Brain tool. Mike built a connector that links ChatGPT to Google Ads accounts, giving the AI full visibility into everything happening in the account.

Bia took that connector and loaded her entire framework into it. “I put all this information and everything that I do—it was more than 300 questions and answers. So I get the results in the way I want.”

Now she has what she calls a “fetch framework” prompt. She types the command, adds the Google Ads account ID, and the AI generates a complete analysis: conversion action audit, audience review, account structure assessment, performance over the last 12 months, what needs to be killed, what needs consolidation, Merchant Center feed health.

“I just run one prompt, and I have all this framework implemented for your account,” Bia added.

It’s still not fully automated. A human still needs to guide decisions, validate recommendations, and make strategic calls. But what used to take days of manual work now takes minutes, freeing Bia to focus on the thinking work rather than the data collection work.

She’s also integrating Optmyzr into the process. “I say like, go to Optmyzr website and run that, read the information in there for me on that page, and from there I have this document that is just the starter document,” added Bia.

Fred mentioned that Optmyzr has an MCP (Model Context Protocol) integration that could further streamline it. Bia immediately asked to be counted in.

Why understanding intent matters more than keywords

The biggest mindset shift Bia asks of her clients is moving from keywords to intent. That sounds simple, but it fundamentally changes how you build campaigns.

Keywords assume you know exactly what people will type. Intent assumes you understand what people are trying to accomplish, even if the specific query varies.

“One of the things I love about Google Ads is that it’s always changing,” Bia said. “So it’s always something for you to learn. When AI started coming up, I was one of the first ones to say, ‘Hey, I want to test it.’”

She was an early adopter of Performance Max when it was still rough around the edges. Same with Demand Gen. She doesn’t wait for things to be perfect. She tests, learns, adapts.

That experimental mindset led her to notice something interesting: BookTok was driving Google Shopping behavior. The book community on TikTok had gotten so large that it was spilling over into purchase behavior—not just for books, but for book-related products.

“We have companies like Popsocket in their third collection about book-inspired phone cases. We have Black Milk doing book-inspired clothing collections because they are not waiting for the keyword to pick up. They are looking for the intent where the intent is.”

This is what Bia means by thinking outside the box. The demand exists. It’s just not where you’re looking if you’re only checking keyword search volume.

She uses Google Trends extensively to spot rising queries related to her clients’ products or brands. Those rising terms become search themes in Performance Max campaigns. But more importantly, they inform the broader intent strategy.

And critically, that language needs to be consistent across Google Ads, SEO, and the website itself.

Bia explained, “We all need to have the same communication, the same language, so AI can understand all of this and have just one concise information about my business.”

How to break down silos between PPC, SEO, and other channels

One of the challenges of the AI era is that siloed teams produce siloed data. If Google Ads is optimizing for one message, Meta is optimizing for another, and SEO is targeting a different language entirely, the overall business suffers.

Bia’s agency is fortunate to have direct Slack channels with clients that include their CEO team, CRM team, Google team, and Meta team. “We can align where it’s important, so it’s one message across all channels and all the business,” explained Bia.

That cross-functional collaboration unlocks new opportunities. If a client doesn’t have a budget for top-of-funnel work in Google Ads but is already running Meta campaigns, Bia uses Meta as a creative lab. Whatever’s working there gets adapted for Google.

“Let’s use Meta as our Google Ads creative lab. So whatever is working on Meta—okay, it’s working really well, and we have the foundation—okay, let’s bring that to Google,” added Bia.

This kind of thinking requires moving beyond “I own Google Ads and that’s my territory.” In the AI era, with automation handling more tactical work, the strategic value comes from seeing patterns across channels and applying learnings wherever they’re relevant.

As manual tasks—like reviewing Performance Max search terms, running budget pacing reports—get automated, marketers have more room to plan and strategize. That’s where the real wins come from.

Why stopping to think before acting saves time

Fred asked Bia what skill is currently underrated in paid ads. “I think we need to think outside the box,” she said, circling back to the BookTok example. “The demand is there. It’s just not where you are looking at.”

The shift she advocates is understanding intent and thinking about where your people actually are. If your audience is on Microsoft Ads, being great at Google Ads won’t help. If your customers are discovering products through TikTok trends, keyword research alone won’t capture that demand.

But more fundamentally, Bia emphasized that business skills and context matter more than platform expertise. Understanding the client’s actual business—their margins, their competitive position, how their sales process works—makes you far more valuable than knowing every obscure Google Ads feature.

Fred agreed, pointing out that most wins happen outside the Google Ads account. “Whether that be the landing page, the offer, how the person answers the phone in the legal office, or whatever it is, it’s outside of Google Ads. That’s where most of the wins can be had,” explained Fred.

Bia’s framework reflects that understanding. It’s not about clever hacks or advanced tactics. It’s about building a solid foundation, giving the algorithm clean signals, and then continuously monitoring to ensure it’s heading in the right direction.

The mindset shift from control to guidance is hard for many advertisers. They’re used to exact match keywords, manual bidding, and granular ad group structures. Letting go of that control feels risky.

But as Bia puts it, “Your work is not going to be a lot of keywords or a lot of bidding strategies for keywords and everything, but how you can guide machine learning to do what you need to do.”

That’s the real skill in 2026: understanding what the algorithm needs to succeed, providing it, and then verifying it’s doing what you expected.

Episode Transcript

Frederick Vallaeys: Hello and welcome to this episode of PPC Town Hall. My name is Fred Vallaeys. I’m your host. I’m also CEO and co-founder of Optmyzr, a PPC management tool. So for today’s episode, we have Bia Camargo. She is someone you may not have heard about, but she has been doing PPC for quite a long time, and she’s actually worked with one of the most influential people in PPC, Mike Rhodes. So she got her start with Mike Rhodes.

Now working with an expert like that, she has done some amazing work, and she’s come up with a framework for how to make accounts ready to make the most out of all the new capabilities that exist in PPC in the age of AI. So it’s a framework, and I can’t wait to talk to Bia about this framework and hopefully teach you as well how you can set up your accounts for success and replicate what Bia has done. So with that, let’s get rolling with this episode of PPC Town Hall. Bia, welcome to the show.

Bia Camargo: Hi, thank you. Thank you for having me.

Frederick Vallaeys: Yeah, of course. You were an Optmyzr customer, and that’s sort of how we got reintroduced after you worked with Mike Rhodes at his agency. But yeah, I’m so happy you did because you’re doing some amazing things, and we wanted to share that with the world. And yeah, I’m very grateful that you’re willing to share what you’ve learned and what’s working for you. So before we dive into it, tell us a little bit more about who you are, how you got to be a PPC practitioner and expert for 15 plus years now. So yeah, give us the background on Bia.

Bia Camargo: Yeah, so I’m Brazilian, and I moved to Australia 12 years ago. I had a marketing career back in Brazil, but when I came to Australia, I had to start all over again. So I started from the bottom and was building my career in digital marketing. I was doing a little bit of everything. I have a graphic designer background as well. And then at some point, I started at this agency, and I was hired just to focus on Google Ads, on PPC. And I said, “Yes, that’s fine. I was doing it before. I know what I’m doing. I will be fine.”

And then I started working with Mike Rhodes at WebSavvy, his agency. And in the first two weeks, I realized I know nothing about Google Ads. Like everything that I thought I knew, I didn’t. So I was so blessed that I could work with him and learn from him. And the most thing that got me is not just learning the practical things of the platform, but he’ll teach us the mindset or how to think Google, how to think why we are choosing this bid strategy, why we are doing these adjustments. And then I started learning from him, and from there, just got better.

So I worked at WebSavvy for five years, and recently I joined Ecom Nation as head of Google, and it’s been wonderful to help so many clients at this point to get their accounts ready for 2026.

Frederick Vallaeys: That’s amazing. And I love the point that you’re making about teaching how to think about the goals and to think about marketing. So it’s not just about knowing which buttons to push, but it’s knowing about why we are pushing these buttons. Because I think what’s fundamentally changing in 2026 and beyond is that user interfaces and pages where we go to push on buttons to do things, that’s going away.

We’re coming into an age where you’re just going to talk to the AI and tell it what you’re trying to have achieved, and it will figure out what the buttons are that need to be pushed, and it’ll set it all up, and it’ll do it for you. But that can also be scary, right? So kind of talk to us about that age of AI and how you think about how work gets done and what’s the most important thing for PPC managers and in your case the head of Google to think about.

Bia Camargo: Yeah. So one of the things I love about Google Ads is that it’s always changing. So it’s always something for you to learn. So I’ve always been an early adopter of everything that was coming. So, “Oh, we have PMax now. Yes, let’s test it. Let’s start. Oh, it doesn’t work that well. That’s fine. We will learn from that and adapt.”

So when AI started coming up, I was one of the first ones to say, “Hey, I want to test it.” And even with Demand Gen, I was one of the first ones in Australia to actually test it, and I have a huge case study with success in that. So I’m always open to learn, even if it’s not 100% yet. I’m ready for it.

So but as you said, it brings complications. So on my day-to-day life, it’s not just easily to explain to a client or explain to my team that we need to adapt to AI. I think the most difficult thing is the mindset shift that you have to have going to Google Ads in 2026. A lot of PPC experts are saying you need to let go of control and you need to move to guidance.

So your work is not going to be a lot of keywords or a lot of bidding strategies for keywords and everything, but how you can guide machine learning to do what you need to do. And that is the most difficult thing because a lot of businesses are still using old structures, or they are holding on to things in the past. So I’m gently trying to get them to the AI era, like let’s get your account ready so you can work and AI can work at the same as your business.

Frederick Vallaeys: Yeah, that’s amazing. So talk us through what are some of these things, because I think that gets into the framework now, right? How do you prepare the account to do really well in an AI world? So what would be the number one thing you would start with?

Bia Camargo: So my framework is like you’re building a house. So first we need to start with foundation. So it’s your first-party data and your housecleaning that needs to be done. So even in 2026, people are still having 20 conversion actions, and they are all misaligned, and they are not giving you the right signals. So this is the first thing we do at the framework. We review your structure, your basic structure.

And from there, you’re going to assess and move to the foundation of the house that we call the wiring. So you can start building something from a solid foundation. So you’re never going to put up your walls if you don’t have a solid foundation. So in this part of the framework, we have a look at conversion actions and audiences.

If you have all your first-party data connected, if you’re having the right signals, the right tools connected to Google Ads. And from there, we are going to analyze the account structure according to your business.

And what I’m finding the most is that sometimes people are overcomplicating things. So what I need to do in this part is actually tell the business and say, “Hey, have a look on your website. Your Google account structure should follow your website structure.”

So the way you organize your website is the way you can organize your account structure. Of course, not one size fits all depending on the business, but you can simplify your Google account. And this is where I receive the most not rejection, but people need a little bit more convincing when I say, “Oh, you don’t need 12 asset groups in this PMax campaign because you don’t have enough budget for them.”

And this is the mindset that I’m trying—a lot of accounts that I’m dealing with still with SCAG structures. So I have 300 ad groups with thousands of keywords. And to make the client understand that that’s not going to work anymore is really hard, right? Because they come from a place where that has worked so well in the past.

Frederick Vallaeys: So let’s unpack what you said, right? Because I think there was a lot there. Let’s go back to the conversion actions. And so it sounds like you do that audit of the conversion actions. What do you typically look for?

Like what is a good conversion action setup? And also tell us, is it e-commerce, lead gen? So help people understand what you are trying to achieve with conversion actions.

Bia Camargo: Yeah, my focus now is e-commerce, but the structure works for both. You need to have a clear structure. So mainly you need to have conversions opted in, but in a good structure. So if you’re using Google Tag Manager or if you’re just using the Google tag, this needs to be working perfectly.

And a lot of things that I see, for example, I have clients with three primary conversion actions for purchases, and I say, “Oh, you are at least double counting your results. So let’s clean this up.” I have like page views, but they are not actually applying to any campaigns. So these kinds of small changes that you can fix like quick wins that get your data more clean to give the machine clear signals, right?

Frederick Vallaeys: So that’s the housekeeping part, and that makes sense. But now in the case of a client who’s got three primary conversion actions, in some cases there is some intent behind it, right? There’s different ways of measuring, maybe the Google way, maybe there’s a different way to measure it based on the business internal data.

So do you find that there’s still value in having those things but maybe set them to secondary action so it doesn’t count towards optimization, but you still have that data flow so you can compare it to whatever other baseline you have?

Bia Camargo: Yes, I do. So normally I have, for example, the primary is coming from Shopify or the website directly, and we have the same conversions coming from GA4 but as a secondary. But we need just to make sure that they’re using the same settings because this way you can compare the numbers and see if you have any gaps or any discrepancy.

So this is important also for your add to cart, begin checkouts, and everything. You need to make sure that the data—not just having them—but you are applying as observation, and you are actually checking the numbers.

Frederick Vallaeys: And then you said ideally you want to have perfect conversion data, but I think in this day and age of privacy, it’s near impossible to have that perfection, right? So then you have a couple of options. So you have GA4, or you have OCR, or you have the Google tag.

What has been your experience? Like if you had to choose one of these to be your primary conversion action and the one that feeds the AI, which one would you typically prefer, or does it depend, and if so, what does it depend on?

Bia Camargo: Look, with the latest updates regarding Google Tag Manager, we are having a lot of issues across a few accounts. So my optimal is to use the Google tag sitewide, and then you connect your native—if you’re using Shopify or if you’re using your website—you connect your native one directly to Google Ads, and this should be your first one. And then you can have the others just for comparison or to get more data.

Frederick Vallaeys: Yeah, that makes sense. And another question. So when you connect your primary system to Google, do you like to give it the full cart value, or do you manipulate that value to be the profit? How do you think about things like not returning on ad spend, but profit on ad spend and profit maximization?

Bia Camargo: Yeah. So when we have this data available, I always try to drive to the profit side, even in the campaign structures like most profit products, less profit products. So we have like third parties like Profitpick or something.

But if it’s not available, I let the machine follow up with what the website is sending. I don’t like to add fixed values if I’m not 100% sure. And it’s incredible that the business actually cannot sometimes provide us with that particular number. So I leave a little bit open, of course with close monitoring, with close eyes, to see how it goes. But it depends on the business as well.

Frederick Vallaeys: Okay, makes sense. Great. So that’s the conversion actions. Then the second component you talked about was audiences.

Bia Camargo: Yeah.

Frederick Vallaeys: Now a lot of people think about audiences in terms of PMax campaigns, and there is some debate, and there is some research that says that overly restricting the AI actually has counterproductive results because sometimes the assumptions we make about an audience may not be in line with what actual people do.

So what is your strategy that you have found works well when it comes to the audience component?

Bia Camargo: So the first part is actually to make sure that we have enough good data. So what I’m finding—we have these accounts with 300 audience lists, but they are not up to date. They are not useful, or the size is not valuable for that particular campaign.

So one of the first things I do is a big cleanup of the audiences just to make sure we have the data that is relevant.

And another thing that is easily mistaken is people have this audience, but they don’t apply to the campaigns. They just leave the audiences there. And then moving—once we have all the first part, like if you have a CRM connected, you have a live audience, and you have everything organized—the second thing is creating the audience based on your campaigns.

But for now, what we are doing is adapting your audiences by intent. So in the past we have website visitors, we have like everything, the same setup for everything, like 80-day website visitors, all visitors from GA4. But now with these new signals, we are able to tailor that more. So I don’t think the audiences are limiting PMax. I’m using them as a guide for PMax.

So what I do is a mix of affinity, in-market, more intent related. So people who browse that website that I want, that people—so if it is a competitor website or an interest website that can—the audience needs to answer a question.

So my audiences are built about what they can answer, like what people will be looking to find that particular PMax campaign. And then I use that combined with the search terms to guide the machine to go after those people.

Frederick Vallaeys: That makes sense. That makes sense. And I want to go a little deeper on that. So you have your audience based on the intent and what is the question they want to answer, and then you steer it a little bit with some search themes or search keywords.

And then how do you connect the ad that is shown to the question that was being answered? How do you make sure that there is a match between how the AI shows your ad, is it actually answering the question then? And I think my question then becomes a little bit about, so what is the asset that you deploy in these scenarios?

Bia Camargo: Yeah, it depends. The most thing I’m seeing now is that Google needs a solid foundation, as I mentioned before. And a lot of campaigns are based on, “Oh, new collections, new products,” and then we have problems with stock. We have problems with seasonality. So what I’m moving my clients towards is that you need your PMax campaigns to be your foundation.

So focus on your core products, your core categories, but not actually on product specifics except if you have like a huge budget for that. And then you can be more granular. But if you don’t have that budget for PMax to actually work, you need to be more category-focused.

So that’s why I said follow your website structure. So if you have an e-commerce that is selling beauty brands and you have your categories like shampoo, conditioner, create your asset groups with dedicated creatives and ad copy for that category.

And then my audience will be people looking for shampoos, people looking for solutions for dry hair. And then we try to match as much as we can that particular asset to the audience. So we can be more assertive once the people see the ad and click on the ad. So that’s why it’s very—try to change for a search intent-driven era and not keyword era.

Frederick Vallaeys: Right. And because these are feed-driven campaigns, usually PMax with feed, Google will decide which products to show if it’s kind of a shopping intent query. But do you also allow these campaigns to—so you add assets to them like videos and headlines and descriptions so that the ads can also show in different scenarios on different platforms or different surfaces?

Bia Camargo: Yes. So I always push all assets to be excellent. People say that I’m a little bit crazy about that. Like they call me Marie Kondo of Google Ads because I’m like a cleaning freak to organize everything. But actually I can see proven results every time we push asset groups to excellent, our ads to excellent, because it’s actually the easiest way to do what Google is asking from us without damaging our campaigns.

Because we know some recommendations are not good to apply, let’s say that. But this one I get very about—have all the assets, all the videos, all the creatives, and continually look at these numbers and understand what is happening, like what this creative is doing for the campaign.

And regarding the setup, if my client has budget for a search brand campaign, my PMax campaigns will exclude brand terms because I don’t want to—things, because normally brand terms cannibalize PMax the other terms.

But if they don’t, we can have just one PMax with brand and non-brand. But what I’m telling the team weekly is that it’s not one thing set and done. You need to check PMax search terms at least on a weekly basis because this is our work now. It’s not even like putting the keywords and making them do the work.

It’s giving the machine liberty to do whatever the machine thinks is the best, but then going in the backend and working on placement exclusions, working on negative search terms, negative keywords, negative whatever we need to do to make sure the machine is going to the right places.

So even if you are adding brand terms on your PMax, you need to check what they are doing in there. If they are getting all the attention, if they’re not getting all the attention, if you need to move to a separate campaign. And this is our work now.

Frederick Vallaeys: Yeah, that makes sense. And so that kind of then goes into the structure, right? Brand versus non-brand, steering the traffic to the asset groups or the campaigns that you wanted by doing the negative keyword management. But you also then talked about simplifying the structure. You kind of mentioned, okay, SCAGs, we shouldn’t be doing SCAGs, and I agree with that. But sort of what is that sweet spot where you have enough structure to steer the ship in the right direction without overcomplicating it? Because I think there is a fine balance there, right? So where do you find that balance?

Bia Camargo: Yeah. So the first thing is every single business I have worked with, I try to build the full funnel. So I use Demand Gen and YouTube campaigns for top of the funnel. Middle of the funnel will be our search non-brand campaign. Bottom of the funnel will be PMax shopping and brand campaigns. So that’s how I build this structure.

Sometimes the client doesn’t have the budget for the top of the funnel yet or doesn’t want the middle of the funnel. But this is the structure I try to follow because if the client wants to scale, the structure is there. And from there, what I’m trying—this is the most difficult mindset I need to say—is just that if you have your ad groups going to the same landing page, have the same themes, they need to be in the same place, in the same campaign, like the same ad group. You don’t need 300 ad groups anymore.

And this has been very difficult for people to understand. Say, “Oh, but I—” Google gives an example like before you have your ad group, leggings, black leggings, blue leggings, green leggings. You don’t need to have that anymore. You can have one ad group for leggings, and you tailor your keywords to include everything else. So this is the structure that we try to move away from. And one thing of my framework is that we don’t reinvent the wheel.

We don’t create new campaigns as much as possible. So we use all the existing data, all the existing campaigns, and we just update them to match this new structure that is easier for AI to understand.

Frederick Vallaeys: Makes sense. Okay. So yeah, great structure there. Is there anything deeper? So we’ve covered the structure, we’ve covered the conversion actions, we’ve covered the audiences. What other components do you think people should not be missing out on?

Bia Camargo: So the first part is the measure of fundamentals, is the foundation of the house. We have the wiring, and then when we finish, we go to the next level that is keyword expansion if needed to match these new terms coming in, and then creative quality.

So we review all the creatives. We push everything to be excellent. But besides that, we actually look at the creative’s performance, what is working, what is not working, what is relevant.

So I see a lot of accounts going to, “Oh, we need to refresh all ad copy. Oh, we need to refresh all the PMax asset groups.” But a few people are actually just replacing the low-performance ones or getting the best-performing ones and applying to other places that are relevant.

And this is something different that I’m trying to get the attention of the people. You say you’re not going to have 100% like 15 headlines excellent. But you can learn from them. You can learn from what is working and what is not working.

And from there, I go to plumbing because plumbing needs to be working from foundation until you finish your house. And my plumbing is GMC—so Google Merchant Center, if you are e-commerce, feed health, and everything that you need to keep running while you’re working on your foundation, while you’re building your house.

You always need to keep working on your feed. So that is like the last step, but actually happens at the same time we are working through those.

Frederick Vallaeys: Great. And so when it comes to the feed, do you use AI to rewrite descriptions? Do you use AI to generate image assets? Talk a bit about how AI fits into all of this plumbing.

Bia Camargo: Yeah. So the first thing I was so blessed with was that I learned a lot about Google Merchant Center with one of the best people, that is Casey Jill. She’s also one of the top PPC in the world, and I worked with her for two years. So like this is mind-shifting about organizing your feed, having the right information, having the right properties, and what can you do.

So the first thing is feed health. We organize everything that is not working. So if you have issues on your feed or if you have information that is not getting right, we will work to fix all of that. From there, according to the business, we will do optimizations. So we use AI to identify these optimizations.

So for example, the brand name is appearing first, but for that particular business, we need to display the product, the product itself first, then the brand.

So these kinds of things we use AI, and we use third-party platforms like DataFeedWatch or Rich Intelligent to make these AI optimizations for us. And then we go to again the importance of actually having a look at the results. What are the products being pushed? Where is the health of the feed? And this is what Optmyzr helps us as well.

So I’m not going to shy out, but yes. So it was the first thing when I moved to Ecom Nation, I said, “Okay, we need Optmyzr.” So I always run the feed quality score, the feed audit, and then from there I go with the clients, and we can work on all the issues because for e-commerce, this is non-negotiable.

Frederick Vallaeys: Well, thanks for the kind shout-out to Optmyzr. We’re happy to have you as a customer as well. Cool. Okay. So no, this is a great framework, right? I mean I think when you look at a lot of PPC campaigns, there’s a factor of laziness. There’s a factor of just not knowing what needs to be done. And so you’ll see a lot of campaigns that don’t do these things and haven’t really thought through it.

But let’s shift focus a little bit. I’d love to hear from you, like where have you really pushed the envelope, or where have you used AI to do something dramatically new, or where are you just waiting to try something new that you’ve always wanted to but maybe you haven’t gotten to? Like where are we pushing the envelope?

Bia Camargo: So I had this framework, and I said it’s great for medium business or small business because I can apply very quickly. We get 20 quick wins from the day we start. So since I started at Ecom Nation, we already did five or six account restructures, and from the day we did, we saw the uplift. I never had a case that the account got worse after applying those.

But then I had an issue. How can I scale that for the biggest business? Because I need to have visibility, and I had one particular account that I’m working on that is like a huge account, and I need to understand the intent behind every campaign to actually build my audience, build my keywords, and how can I consolidate this account following intent.

Then I was at an event, and I spoke with Mike. I said, “Mike, I need help. Can you help me? I need to scale this. I need to understand the intent behind this.” So we started working on his 8020 Brain. So we connect Google Ads GPT, and then I put all my framework in there. And from that, all of these that I mentioned for you are automated. So I just run one prompt, and I have all this framework implemented for your account.

From there, I told like, “But I need—”

Frederick Vallaeys: So that’s really cool. So explain that a little bit more. So a lot of people may also not know what the 8020 framework from Mike is. So give us a bit more detail there.

Bia Camargo: Yeah. So imagine ChatGPT. You go to ChatGPT, and Mike built a connector that you can connect this ChatGPT to your Google Ads account. So it has visibility of everything happening in the account. And he already built some prompts in there that tell like, “Can you give me the performance for the last 30 days?” And then it gives key insights because he already trained this chat to answer these questions.

Then I went there and said, “Okay, we are now doing Bia way,” and I put all this information and everything that I do. It was more than 300 questions and answers. So I get the results in the way I want. So I have now a Google ChatGPT that does what I want to do, basically.

Frederick Vallaeys: Okay. So you have like, yeah, very simple terms. Okay. So you have a mega prompt, and the mega prompt asks the AI to give 300 or answer 300 questions while being connected to the Google Ads account. So it’s basically do the Bia audit. Like that.

Bia Camargo: Yes. So I actually just need to type fetch framework for this account, and I put just the Google ID. And then it will give me all the analysis of conversion actions, all the analysis for the audience, account structure, account performance, performance the last 12 months, what is right, where is wrong, what accounts I need to kill, what accounts I need to keep, consolidate. And from there, GMC, and then we are trying to put Optmyzr inside as well.

So I say like, “Go to Optmyzr website and run that, read the information in there for me on that page.” And from there, I have this document that is just the starter document, of course, because we need a human to guide everything. But it’s done with one prompt.

Frederick Vallaeys: Okay, great. And by the way, we have an MCP. I don’t know if it’s kind of a secret, but if you want the Optmyzr MCP, we can give it to you. And anyone listening, we can get it to you as well.

Bia Camargo: Count me in.

Frederick Vallaeys: Okay, cool. We’ll get it your way. Yeah, that’s really amazing. Okay, so that’s kind of pushing the envelope of what’s possible with AI.

Bia Camargo: It gets better.

Frederick Vallaeys: Okay, let’s hear it.

Bia Camargo: Yeah, it gets better because the thing I asked Mike, I told him, “I need something so I can map the intent behind the campaigns.” And he built it for me. So I have intent mapper. So imagine if you can export your Google Ads account into a map tree view.

And once you click on your campaign—for example, if it’s a search campaign—we will open the ad group. You click in the ad group, we will open the keywords, we will open the search terms, we will open all the information related to that particular campaign.

So I can see how it works, and we are in the final stages to build the AI will identify the intent for that particular campaign. So this is what it’s coming. So we are pushing to have a tool that you can identify your intent by campaign so you can plan and adapt your account accordingly.

Frederick Vallaeys: Very cool. Okay. And if people want to stay in touch with you, remind them right now, where are you most active?

Bia Camargo: Yeah. So you can follow me on LinkedIn, Bia Camargo. That’s where I’m most—I pass my days now. So I’m always sharing good news and everything that is happening in the Google Ads world. And yeah, it’s the best place to find me.

Frederick Vallaeys: Great. Well, thank you for sharing what you’re working on. It sounds very exciting. And then you did mention you share all the happy news from Google Ads, the happy and the unhappy. But what are some recent things that you think people should pay more attention to when it comes to the new Google capabilities?

Bia Camargo: I think we need to think outside the box. So one of my latest articles, it’s how BookTok is actually driving intent for Google Shopping. And this just happened from November last year, from Black Friday in December. And I explained the demand is there. It’s just not where you are looking at. So this difference between keywords and intent.

I think the most shift is that how can I understand the intent behind the campaigns and how can I think about intent, where my people are, what they will search for me to show up? And that is the biggest difference. So this is I think is the most important thing right now is how to understand intent and how can you adapt for that.

Frederick Vallaeys: Okay. Yeah. The intent and then that goes back to the audience who has the intent and the thing that you’re building. And talk a bit more about—so BookTok is driving shopping behavior. Like explain that a bit more.

Bia Camargo: Yeah. So what happened is BookTok has a section that is called—TikTok has a BookTok community, and this BookTok community is huge. And what was noticed in the past months is that it gets out of BookTok and actually turns into a shopping behavior. So even brands that have non-book-related products were adapting their products to be book-related.

We call it bookish. So we have companies like Popsocket. They’re in their third collection about book-inspired phone cases. We have Black Milk doing book-inspired clothing collections because they are not waiting for the keyword to pick up.

They are looking for the intent where the intent is, and they have very successful numbers around those. So this is just one thing to look. People sometimes say, “Oh, I need Google,” “But your people is on Microsoft Ads.” So you need to look beyond.

Frederick Vallaeys: So that sounds like a very innovative thing to do, right? And how do you personally stay up with kind of these new consumer trends and what seems to be working in marketing? And how do you stay on top of it?

Bia Camargo: One of the biggest tools I use now is actually Google Trends. So Google has a website with Google Trends, and you can search for whatever the terms you are looking for or even your brand name or whatever you’re looking.

And then even gives you how it’s performing according to—if you want Australia, worldwide, you can see the behavior happening. And also gives you search terms that are rising, like queries that are rising. And that is the ones I want in my PMax next campaigns. So they are related queries for the intent for the campaigns I want to build.

So these—what people are looking in Google to find that particular product or that particular brand. And this is what I want in my account. And most important, not just on my account, but I need on my account, I need on my website, I need on my SEO. There is not one Google does that, SEO does this and website. No, we all need to have the same communication, the same language so AI can understand all of this and have just one concise information about my business.

Frederick Vallaeys: Yes, that makes a lot of sense. You need to communicate across your agency, not always the easiest thing, you know, when you have different teams in charge of different things. But then also I think AI comes into this, right? So if you need to address your findings in 10 different places from landing page to keywords to social campaigns, it is nice because AI can actually fill a lot of those gaps and do a lot of that work for you.

And maybe go to the teams and say, “Okay, here’s something that’s already 70% done,” and they can take it across the finish line.

But how do you think about in an agency world or in a team-based world, how do you align people behind the marketing objectives more than maybe those silos that have come to exist between PPC and SEO and social?

Bia Camargo: Yeah. So I know it’s something very new, and it depends on the culture. So in my agency, we are lucky enough to have a paced agency relationship. So for example, with our clients, we have directly Slack channels with their teams.

With that, we can also invite their other teams. So we have, for example, one client, Beia client, and we have their SEO team, they have our CRM team, they have Google team, Meta team. And then we can align where is important so it’s one message across all channels and all the business.

Not easy to do. It’s something we are starting to get used to, especially on how we work with those teams. And most important, what can we benefit from working with them. So one thing that is this new culture of what is working on SEO, “Oh, I can adapt that for Google.”

What is working on Meta, especially when, for example, my client doesn’t have a budget to do top of the funnel, but they are doing Meta. Great. Let’s use Meta as our Google Ads creative lab. So whatever is working on Meta, “Okay, it’s working really well, and we have the foundation. Okay, let’s bring that to Google.”

So it’s on an AI era. We cannot look just one place anymore. And for all of the other things that you used to waste our time on, like manual tasks, revising PMax search terms and kind of thing, we have AI, we have automations for that. We don’t need to waste our time on that. And with that, we have more room to plan and strategize. And from there, create these new collaborations with other teams.

Frederick Vallaeys: That’s great. And so speaking of AI, which ones do you like best? Which ones do you use the most?

Bia Camargo: Okay. Of course, because of Mike, I use Claude Pro for all the coding and every the backend we do. But I use personally ChatGPT for everything that is my mom life related. So like from my grocery shopping to my kids’ appointments and everything, it’s like ChatGPT is my best friend, my personal assistant. But we also have different tools like Optmyzr for Google Ads. We have Coworker for everything internal. So it’s not one tool for everything. We have like I’m 100% AI person. So I have tons of them.

Frederick Vallaeys: You try them all. Very nice. Okay. Well, this has been amazing. Thank you so much for sharing a great framework that I think a lot of people can just action today. And if people wanted to get some of these resources or the way that you think about things, where can they go and find them right now?

Bia Camargo: Yeah. So I’m working on these materials, but actually I will be presenting the full framework and also case studies related at the AdWorld experience in Italy. So I will be speaking on the two days, and I want everyone to watch that. So it will be the first place that I will show the full framework in real time and show every single piece of bits. But yeah.

Frederick Vallaeys: Very cool. And I believe AdWorld experience, is that the one that’s in October?

Bia Camargo: Yes, it is.

Frederick Vallaeys: Okay. So for the people who are eager to get—

Bia Camargo: Yeah, I will slowly release some things like actual checklist that you can do right away, especially in my coaching sessions. I do one coaching for business owners and also train Google Ads there. We will be releasing a little bit of helpful information as we go, but the full framework will be just for the event.

Frederick Vallaeys: Okay. But you can get bits and pieces by connecting with Bia on social media, on LinkedIn and the learning platforms. I will put those in the show notes so that people can connect with you. So yeah, this has been amazing.

Thank you so much, Bia. And thank you, everyone, for watching. I hope you’ve enjoyed this episode. And if you want to know when the next one comes out, please subscribe, please like this one, put comments about what you want to know more about. And we’ll all be monitoring that. So thank you, Bia. Thank you, everyone, and we’ll see you for the next one.

Bia Camargo: Thank you for having me.