How Brooke Osmundson Manages Multiple Google Ads Accounts Without a Large Team

Episode Description

Fred Vallaeys, CEO & Co-Founder of Optmyzr and author of The AI Amplified Marketer, sits down with Brooke Osmundson, Director of Growth Marketing at Smith Micro Software and Search Engine Journal contributor, to talk about what AI actually looks like in practice for a solo marketer managing multiple Google Ads accounts in a regulated industry.

Brooke operates in a space most AI conversations don’t account for: family safety products marketed through wireless carrier partners who have strict brand guidelines, their own AI policies still being written, and restrictions on what image and video AI tools can even generate.

What she’s figured out is that different AI tools serve different purposes, and knowing which one to reach for matters more than picking a favorite.

Here’s what was discussed in this chat:

• How Brooke uses ChatGPT, Claude, and Google’s AI tools differently
• Why AI is helping marketers do more with less
• Practical AI workflows for teams of one
• How AI supports writing, brainstorming, and brand storytelling
• Brooke’s experience with Google Ads Advisor and Analytics Advisor
• The challenges of using AI in regulated industries
• Why marketers still need to verify AI-generated research
• The ongoing trust and hallucination challenges with AI
• What AI means for lead generation marketers
• Why offline conversions and CRM data remain critical in an AI-first world

Episode Takeaways

Brooke Osmundson is a one-person team managing multiple Google Ads accounts for Smith Micro Software, a company whose products sit at the intersection of family safety and mobile technology. She also writes about PPC for Search Engine Journal on the side.

Between the day job, the writing, and keeping up with an industry that announces something new every week, she doesn’t have the luxury of experimenting with AI tools for months before deciding if they’re worth it. She needs things that work, and she needs to know which one to reach for depending on what she’s trying to do.

That last part is what makes her perspective useful. Most AI conversations in marketing treat the tools as interchangeable. Brooke has actually developed a clear sense of which tool earns its place at which stage of the work, and why. She’s also working in a context that most AI workflows don’t account for: regulated industries, carrier partner brand guidelines, and AI image restrictions that stop certain creative directions entirely before they start.

How Brooke uses ChatGPT, Claude, and Google’s AI tools differently

The distinction Brooke draws between her tools isn’t about which AI is “best.” It’s about what each one has proven it can do reliably in her specific workflow.

ChatGPT is where she starts when she needs to get her thoughts out. Over time, she’s trained it to understand her voice, her audience, and what she wants to sound like.

“I’ve trained it enough to really understand what I want to sound like, who I’m talking to,” Brooke explained.

It’s become a reliable first stop for writing workflows, less because of anything unique to ChatGPT and more because the accumulated training has made it responsive to her particular way of working.

Claude handles a different problem. When Brooke is building a new product story or helping a brand find its voice for a new audience, Claude does that work better.

“When we’re building new product stories, brand stories, it helps me formulate — I’ve got so many things that I want to get out, what do I want to say, what do I want to sound like for new audiences,” Brooke said.

That’s moved her skill set beyond pure campaign management into genuine brand building work she might not have taken on before.

Google’s own AI tools are still a work in progress for her. The Ads Advisor and Analytics Advisor tools give her a way to step back from task mode and ask whether something she hasn’t thought of might be affecting performance. She’s had better results with Analytics Advisor than Ads Advisor so far, but she’s approaching both with patience rather than frustration, treating them as early-stage tools worth continuing to test rather than writing them off.

Why AI is helping marketers do more with less

The pressure to deliver more with fewer resources isn’t new. What’s changed is the tooling available to respond to it.

Brooke is candid that her increased AI adoption over the past year has been driven primarily by time pressure rather than enthusiasm for the technology itself.

“Our company, as well as other companies, were always asked to do a lot more with less time,” Brooke explained. “And so I’ve really been leaning into the different tools.”

For someone managing multiple accounts solo, that time pressure is constant. Every hour spent on a task that AI can handle reasonably well is an hour that could go toward the things only a human can do: the judgment calls, the client context, the creative direction. The shift isn’t about AI taking over. It’s about redistributing where the human attention actually goes.

That redistribution is also what makes the “do more with less” framing more accurate than “save time.” Brooke isn’t finishing work early and going home. She’s taking on scope that previously would have required more headcount, because the AI is absorbing enough of the execution layer to make that expansion possible.

Practical AI workflows for teams of one

The practical reality of being a marketing team of one is that you can’t afford to have a broken workflow. If the AI tool you’re using isn’t returning something usable, you’re not passing it to someone else to fix. You’re fixing it yourself or starting over.

What Brooke has built is a workflow where each tool has a defined role and she’s not asking any single tool to do everything. Brain dumps and writing go to ChatGPT. Brand narrative work goes to Claude. Research stays mostly manual because the hallucination risk is too high to trust the output without verification. Google’s AI advisor tools get checked during performance review rather than used as a primary guide.

She’s also built in personal AI use for the organizational tasks that bleed into work. Some of what came out of Google IO excited her most for personal productivity: agents that could help her organize her weekend so she stops sitting in what she called “paralysis mode” when facing a list of 17 things with no clear priority order. Getting better organized personally makes her sharper professionally, and she’s not pretending those two things are separate.

How AI supports writing, brainstorming, and brand storytelling

Writing is where Brooke has found the most immediate and reliable value from AI. Not because the AI writes things she couldn’t write herself, but because it helps her get past the blank page faster and work through the structural thinking before she commits to a direction.

For Search Engine Journal articles, she’ll use AI to help build the foundation of a longer piece — structuring an argument about privacy compliance or AI adoption across industries — and then do her own research to verify the factual claims and push the thinking further. The AI gets her to a starting point. The human work that follows is where the piece actually becomes worth publishing.

On the brand storytelling side, Claude has opened up work she describes as genuinely expanding her skill set.

“I’ve found Claude helps from a brand building perspective a lot better. So when we’re building new product stories, brand stories, it helps me formulate — I’ve got so many things that I want to get out, what do I want to say, what do I want to sound like for new audiences — and that’s helped me really start to bring brands to life,” said Brooke.

Brooke’s experience with Google Ads Advisor and Analytics Advisor

Brooke is testing Google’s AI advisor tools with measured expectations. Both are designed to surface recommendations and flag performance changes, but she’s found them genuinely useful at different things.

Analytics Advisor has been more practically helpful for her. When multiple things are changing at once and she’s trying to understand what’s actually driving performance volatility, it gives her a structured way to look at the situation and sometimes surfaces angles she hadn’t considered. Ads Advisor has been less useful so far, though she’s treating that as an early-stage product issue rather than a fundamental flaw.

What she values about both tools is the permission they give her to step back. As someone in constant task mode, having a tool that encourages a diagnostic stance rather than an immediate action stance is useful in itself, even when the specific recommendations need to be verified before acting on them.

The challenges of using AI in regulated industries

This is where Brooke’s situation diverges most sharply from the standard AI in marketing conversation. The family safety tools sold through wireless carrier partners. Those carriers have strict brand safety guidelines, and their AI policies are actively being written right now. That means the rules are changing while she’s trying to test what’s possible.

On the creative side, Google’s image and video AI tools can’t generate content featuring children. For a company whose products are designed around family use cases and whose natural creative instinct is to show families using phones together, that restriction closes off a significant category of creative direction.

“If I wanted to have children playing with a phone, I can never get anywhere because Google cannot create images or videos of children,” Brooke said.

The response to that constraint isn’t frustration. It’s problem-solving: what can the tools create that still communicates what the brand needs to communicate while staying within both Google’s restrictions and the carrier partners’ guidelines? That’s a harder creative brief than most marketers are working from, and it’s one that requires human judgment at the center rather than AI automation.

Why marketers still need to verify AI-generated research

Research is the area where Brooke has the least trust in AI, and she’s explicit about why. She’s seen hallucinations. She’s watched AI respond to corrections with enthusiastic agreement — “You’re absolutely right, you’re absolutely right” — without actually changing the underlying answer. She’s tested it deliberately by saying something false to see if it would push back.

“If I say in all capital letters, ‘do not be biased,’ then it starts to understand, but then two or three conversations later, it goes back to the bias, maybe some hallucination,” Brooke mentioned.

For a Search Engine Journal writer whose work goes to a real audience, publishing something factually wrong because an AI said it confidently isn’t an acceptable outcome. So research stays manual. AI can help structure an article and identify the questions worth answering, but the verification work is hers. She frames this not as a limitation she’s frustrated by but as a realistic assessment of where the tools are right now and where her responsibility as a publisher sits.

What AI means for lead generation marketers

Brooke works in lead generation and app marketing, not e-commerce. Most of the AI-driven Google Ads updates and announcements are built around e-commerce use cases and purchase conversion signals. That gap is real and she’s aware of it, but she’s also watching Google start to close it.

The biggest issue for lead generation isn’t campaign automation. It’s lead quality. Volume is achievable. But quality leads that eventually become closed deals, sometimes after sales cycles stretching eighteen months, are much harder to optimize toward.

“The biggest thing that I hear from a lot of people is lead quality. And so what I’m hoping for is, are there any advancements or guardrails that are going to come in terms of lead quality from lead gen campaigns? I’m not sure what that would look like. That might be a tall ask, but there’s always hope,” Brooke remarked.

Whether AI can eventually help with that, and what guardrails that would require, is something she’s watching for.

She’s also thinking about what Microsoft’s LinkedIn integration gives them as an advantage in B2B lead gen that Google hasn’t matched, and whether the tools Google is building will eventually close that gap.

Why offline conversions and CRM data remain critical in an AI-first world

Fred made a point that Brooke agreed with directly: getting lead quality right in Google Ads requires proper CRM setup, offline conversion tracking, and the discipline to feed that signal back to Google consistently.

“To get good lead quality conversions at some level, you have to properly set up your CRM. You have to properly communicate that back into Google. Your analytics has to be correctly set up. And that stuff remains very complicated, and honestly I think most PPC people — that is not what we spend the majority of our time on. That should be the one-time thing you do at the beginning of managing a campaign,” Fred explained.

For agencies in particular, this is a structural problem. CRM setup and data hygiene isn’t their specialty. They’re relying on clients who have their own systems and their own ways of measuring success, and aligning those systems with what Google actually needs to optimize toward requires conversations that go well beyond campaign settings.

“Garbage in, garbage out” has been the principle since the beginning, and AI doesn’t change it. If anything, it makes it more important, because the algorithms are more dependent on signal quality now than they were when human bidding could compensate for data gaps. Brooke sees AI potentially helping here too, not by solving the CRM problem but by shifting where she spends her attention toward the foundational work that actually determines whether the campaigns perform.

Episode Transcript

Frederick Vallaeys: Hey, I’m Fred Vallaeys, and I’m talking to AI amplified digital marketers to find out how they’re using AI to make themselves better. Today we’ve got Brooke Osmundson in the studio from Smith Micro Software. Brooke, good to see you again.

Brooke Osmundson: Thanks for having me.

Frederick Vallaeys: Yeah, thanks for coming in. And you’re here for GML, Google Marketing Live, in town this week.

Brooke Osmundson: Yes.

Frederick Vallaeys: But AI is going to be obviously a big theme at that conference, but I wanted to hear from you as a practitioner. How has AI changed you as a marketer?

Brooke Osmundson: I will say I’ve been adopting AI a lot more in the last year than I have in the last couple years. And it’s more so from a time-saving perspective. So our company, as well as other companies, were always asked to do a lot more with less time. And so I’ve really been leaning into the different tools, understanding which ones are going to work best for whatever workflows that I’m working on.

So I’ve got one tool that really helps me with my brain dumps and helping get all my thoughts out.

Frederick Vallaeys: So what tool is that?

Brooke Osmundson: I use ChatGPT from a writing perspective. I have tested other tools. Claude kind of gets me maybe 60% of the way there. Tested out Gemini for things like that. But what I found is ChatGPT — I’ve trained it enough to really understand what I want to sound like, who I’m talking to — that really helps me get my thoughts in order, which is great.

I’ve found Claude helps from a brand building perspective a lot better. So when we’re building new product stories, brand stories, it helps me formulate — I’ve got so many things that I want to get out, what do I want to say, what do I want to sound like for new audiences — and that’s helped me really start to bring brands to life. So it’s helped me expand, I think, my skill set out of just even Google Ads from that perspective. So just testing the different tools, learning to make mistakes with it, and learning to be a little bit more flexible, get out of my comfort zone.

Frederick Vallaeys: Great. And what is it you’re hoping that Google will announce in terms of new AI capabilities, or something they’ve maybe already announced at Google IO that you’re excited about trying?

Brooke Osmundson: I would say from a creative standpoint, I know that they’ve brought out a lot of AI tools with, I believe it’s Llama Nano Pro. So that’s something I really haven’t been able to take advantage of. Right now, just with the carrier partners, everything is brand safety guidelines. Now that they’ve started to bring a lot of that out, I’m excited to try that and I want to see what other advancements they’re bringing to that.

From a Google IO perspective, I actually like a lot of the agents and what they’re helping me do. I thought more I could do a lot of that in my personal life to get me more organized, so then I can better compartmentalize — here’s what I have to do for work, and not have to worry about all of these other things. So I’m excited from both aspects.

Frederick Vallaeys: Yeah, I’m excited to try those agents as well.

Brooke Osmundson: I know. Anything to help me organize my weekends better, because otherwise I sit in paralysis mode of I’ve got 17 things to do and which one do I need to prioritize?

Frederick Vallaeys: And now it does sound like you’re using Claude and GPT. Do you use the AI inside of Google itself? Like the AI Studio, Creative Studio?

Brooke Osmundson: I’m starting to test that out a little bit more, and I always feel like I’m in a unique position because with the companies that I work for and the products that we make, it’s around family safety products and there are regulations on what Google can and can’t create. So for example, if I wanted to have children playing with a phone, I can never get anywhere because Google cannot create images or videos of children. So that’s something I’m trying to get a little bit more creative about — what could I have it create from that perspective but still stay within brand guidelines when working with the different carriers.

Other aspects of the AI components within Google Ads or analytics — I have been testing out their Ads Advisor, Analytics Advisor, just to help me better understand if a bunch of things are changing performance volatility, what are things that I have not thought of that might be happening. I will say I’ve had better success with Analytics Advisor than Ads Advisor from a performance standpoint. But I’m learning to be patient and learn that a lot of these things are in their infancy or early stages. But it gives me time to just take a step back instead of being in task mode all the time, to really start understanding how do I work with these tools to make my life a little bit easier.

Frederick Vallaeys: Yeah, that makes sense. And talk to me about the client’s expectation. They obviously hear a lot of these things about AI, and I think sometimes they have exaggerated expectations. How do you keep that in check?

Brooke Osmundson: That is a concern I have actually, especially when I work with carrier partners in the wireless industry. They have very strict AI guidelines, and especially from a creative standpoint when anything comes to images or video. They’re in the midst of creating their own AI policies of what is allowed and what isn’t allowed. So it does maybe hinder us from using some of those tools to the full capacity. But if I can use AI in other aspects, even just from a text writing perspective, and help it stay on brand, it’s going to save me time because I’m a team of one managing a lot of these Google Ads accounts.

Everything we still have to go through brand approvals and whatnot, but I want to be able to test what I can while they’re still adopting their greater AI policies.

Frederick Vallaeys: Makes sense. Now, you’re also the SEJ PPC person, or one of two, I believe.

Brooke Osmundson: Yeah, I’m one of the main PPC writers, so it helps me in my free time, whatever that is. That’s where I like to stay up to date on what’s happening, especially because I work in such a unique industry of mainly app marketing. I get to make sure I understand what’s going on in e-commerce, what’s going on in lead gen, small businesses, enterprise. Being a writer in my spare time helps me, even though I’m so niche and specialized, to understand the greater impact of AI and all these other industries.

Frederick Vallaeys: And we’ve talked a little bit about how AI makes you better at writing or more efficient at writing, but what about the research phase? Obviously there’s Gemini deep research, there’s deep research from all of the other AIs. Do you find it still hallucinates, or how do you get your news and how do you know what’s actually true?

Brooke Osmundson: Yeah, the research part is hard for me, and I will say that is something I still do a lot manually. And I think that you make a good point — that’s something that I should challenge myself on because there is a lot of trust issues. I have seen hallucinations. Anytime I try to give it feedback, it’s always, “You’re absolutely right, you’re absolutely right.” I’m like, why don’t you challenge me when I say something? And so sometimes I just test it by saying something actually false.

Frederick Vallaeys: And does it also agree with you if you say something false?

Brooke Osmundson: If I say in all capital letters, “Do not be biased,” then it starts to understand, but then two or three conversations later it goes back to the bias, maybe some hallucination. So I will say transparently, I need to do a little bit more with the deep research. And it can help me, I think, from a fundamental standpoint, help structure something where I am writing a larger article on the state of whether it’s privacy compliance or AI in these capacities. And then I’m doing research on, okay, is this correct, is this correct, where can I go a little bit deeper.

So it helps with the foundation, but at the end of the day it just shifts, I think, where I spend my time and making sure things are factual. Then it starts to get me thinking, okay, is there a different angle on this, how do I want to frame it? So it has its pros and cons, but I would never trust it — I should not put a date on “right now” — I don’t fully trust it, whatever AI tool I’m using.

Frederick Vallaeys: Yeah. Yeah. There’s still a lot of stuff for the human to do.

Brooke Osmundson: Oh yeah. It’s really just going to shift what we do and how we spend our time. And I see that as a manager, even outside of Google Ads. I truly don’t think my job is going away. It takes a lot of the tasks, but how many years have we been saying garbage in, garbage out? And it’s the same with any new AI piece that they bring in.

Frederick Vallaeys: Yeah. And Brooke, you do work mostly with what amounts to lead gen and app marketing. When it comes to Google, their updates are always so focused on e-commerce. But what do you think are some of the big things in lead gen that you either hope for or that you’re excited by and that you think others could use?

Brooke Osmundson: Yeah, and I know that Google gets a bad rap of everything is e-commerce. But at the end of the day, we’re all consumers, whether we’re B2C or B2B. And I think that they’re starting to really get that feedback that lead gen is really important. And I think if they want to keep up with some of the other platforms where Microsoft does a really great job and has that integration with LinkedIn, they don’t have that advantage. So they have to bring in some tools to help with lead gen.

The biggest thing that I hear from a lot of people is lead quality. And so what I’m hoping for is, are there any advancements or guardrails that are going to come in terms of lead quality from lead gen campaigns? I’m not sure what that would look like. That might be a tall ask, but there’s always hope.

Frederick Vallaeys: And listen, I think that’s my hope too. To get good lead quality conversions at some level, you have to properly set up your CRM. You have to properly communicate that back into Google. Your analytics has to be correctly set up. And that stuff remains very complicated, and honestly I think most PPC people — that is not what we spend the majority of our time on. That should be the one-time thing you do at the beginning of managing a campaign.

And so it’s hard for us to keep up with how does that work, is this the best practice, and even if we know the best practice, where did that button go this month? And especially for agencies too, because that’s really not their specialty. You’re relying on the clients to make sure that they have a good CRM system set up.

And to your point, they do it once. Listening to the Ads Decoded podcast, they are spending a lot of time really hounding the basics of making sure, especially for lead gen, having offline conversions set up and really understanding the business goals. What does this mean to you? What is a form fill versus then what happens afterwards?

Brooke Osmundson: So I think it’s not just particular to Google Ads. It’s any platform. They really need more information about that end lead, whether it’s a $10,000 closed deal — sometimes it’s millions of dollars, takes 18 months — but you expect immediate success in Google Ads. And so I think that’s always just going to be a pain point for a lot of people. Nothing’s ever going to be perfect, especially if you deal with long sales cycles or lead cycles.

But I think that’s something that again AI might start to shift — what you’re spending your time on instead of all of these busy tasks. Okay, I might get a little bit smarter looking at some of these different CRM perspectives. I think more well-rounded, but who knows what’s going to come out.

Frederick Vallaeys: And even if it’s not instant perfection that it delivers, at least can it make it easier to get one step closer. Any positive feedback that it gets is one step forward.

Brooke Osmundson: I agree. Yeah.

Frederick Vallaeys: Hey, Brooke, it’s been great having you in the studio. Thanks for sharing some of your insights, and we hope to have you back on another episode soon.

Brooke Osmundson: Awesome. Thanks again.