How Should PPC Teams Split Budget Beyond Google Ads?

Episode Description

Fred Vallaeys, CEO & Co-Founder of Optmyzr, is joined by Aaron Levy for a rare double-hosted episode, sitting down with Joe Martinez of Paid Media Pros for his latest return to the show.

Joe has spent years deep in Google Ads, but these days that’s one of the smaller pieces of his week. Most of his time now goes into managing and testing Microsoft, LinkedIn, Reddit, Meta, OpenAI, TikTok, and Quora, often for the same lead-gen clients at the same time.

The conversation moves through how Joe decides where the next advertising dollar goes once Google stops being the obvious answer, what he’s actually seeing from OpenAI ads so far, why native platform reporting often understates what a channel is doing, and what separates agencies that are thriving right now from the ones quietly losing clients to AI-generated everything.

Here’s what was discussed in this chat:

• Why Joe now spends less time in Google Ads than in almost any other channel he manages
• What Joe has seen so far testing OpenAI ads, and where the reporting still falls short
• Why LinkedIn remains the default next step for B2B lead gen, with Reddit close behind
• Why native platform metrics can make a channel look like it’s failing when a media mix model says otherwise
• What it actually costs to get into high-production channels like Disney ads
• Why Demand Gen has started outperforming standalone video campaigns for conversions
• What separates agencies that are winning right now from ones quietly losing clients
• Joe’s platform-by-platform take on where the next ad dollar should go

Why Keep Running to Google?

Joe opens with something that would have sounded strange from him a few years ago.

“I’d honestly say Google is probably one of the least amount of channels in terms of time that I work when working anymore, which is weird to say,” Joe said. “I never thought it would be like that getting into this.”

That’s not a complaint about Google’s performance exactly. Costs have gone up while conversion rates have stayed flat, which means the same budget buys less than it used to.

For a lead-gen specialist working with clients whose budgets don’t grow every year, that math forces a search for cheaper alternatives elsewhere. Aaron pushed on why that search feels necessary now, floating the idea that Google has simply gotten crowded with competent operators, leaving less room for an edge.

Joe agreed that’s part of it, but he was just as honest about the other part.

“I’ll fully admit part of it is out of sheer boredom,” Joe said. “It’s more fun to work in a dozen-plus different ad platforms than just Google all day.”

Google still absorbs a large share of the actual ad spend Joe manages. It just runs quietly in the background at this point, largely on autopilot through the automation his team has already tuned. The interesting work, and the interesting conversation, has moved elsewhere.

What Joe Has Learned Running Ads on OpenAI

OpenAI’s advertising product is still in beta, and Joe treats it accordingly: promising, actively evolving, and missing some fairly basic reporting.

His biggest frustration is one that will sound familiar to anyone who managed search before query-level reporting existed. A client had a strong month, then a flat one, with no changed signals and no changed ad copy on Joe’s end.

“We’re just pretty much fluctuating bids and budgets,” Joe said. “It’s that lack of clarity of what’s being shown, to try to optimize our signals better.”

Fred connected this to something he’s hearing directly from contacts at Google: the entire concept of a “search query” is starting to break down. What used to be a two- or three-word phrase is now an open-ended conversation, and that shift changes what a reporting tool even needs to capture.

Joe’s wish list for what OpenAI or Google could give advertisers wasn’t a demand for raw prompts. He compared it to something closer to Optmyzr’s own negative keyword tooling: phrases or topic patterns that correlate with poor outcomes, without needing the exact query itself.

One detail Joe stumbled into by accident turned out to matter more than expected. Editing his own live ads, he noticed a yellow warning flag on every single one, indicating the ad was eligible to serve but would get truncated well before the full character count anyone had written.

“They gave you more characters than actually show up,” Joe remarked. That’s a meaningfully different failure mode than a channel simply underperforming. The ad looked fine in the editor and read badly in the wild, for reasons the interface hadn’t made obvious.

Fred offered a comparison that reframed the whole conversation about OpenAI’s behavior so far.

“I want to buy a new pair of shorts, but I want to try a new brand, what can you recommend? And ChatGPT comes back and it’s like, you bought this brand and that brand. I recognize two of the brands, but the third one, I’m like, is it hallucinating? I go through my closet and I look around and I’m like, it actually knows that I have this brand of shorts. It was right, and I’d totally forgotten,” Fred said.

Where organic research on AI tools tends to send people back to Google to convert at higher rates on lower volume, what Joe described sounded closer to the reverse: OpenAI pushing people out to do more research elsewhere rather than closing the loop itself.

Joe’s read on why lined up with his own personal usage. He treats ChatGPT as a discovery tool, the kind of thing you’d ask about concert-week traffic in a city you’re visiting, not a purchase engine.

For his B2B clients selling six-figure, multi-year commitments, that distinction matters. Nobody expects a direct sale to come out of a chat conversation.

“We know people aren’t going to become customers from ChatGPT,” Joe said. “But it’s providing the resources, trying to make those signals available, that they discover us with those answers somehow.”

Fred tied back to a classic B2B targeting problem: a keyword like “printers” can’t distinguish a consumer buying one $99 unit from a corporate buyer outfitting ten offices, but a conversational AI that already knows who it’s talking to might not need to guess. It’s an early, unproven advantage, and both agreed the actual attribution data to confirm it isn’t there yet.

Where the Next Dollar Goes: Microsoft, LinkedIn, and Reddit

Once Google’s headroom runs out, Joe’s list doesn’t start with anything exotic.

Microsoft comes next, largely because it still delivers lower CPCs and lower costs per conversion than Google, even with noticeably less volume. LinkedIn follows immediately after that for his B2B lead-gen clients, described in blunt terms.

“LinkedIn is the no-brainer next step,” Joe said. “It is just too precise.”

That precision holds even as Joe has watched the platform feel somewhat less professional over time, and even as LinkedIn’s cost per lead hasn’t moved much yet in response. When LinkedIn underperforms in specific accounts, Reddit has increasingly been where that budget goes instead, including tests of Reddit’s own automated campaign product, which Joe refers to informally as “RMax.” Reddit’s actual name for it is Max campaigns, though the mechanics Joe describes, hand it some assets and let it run with minimal targeting, match what he’s testing either way.

The results in at least one account surprised him. A service-based client started at roughly $7,000 a month and has since doubled that spend based on performance, a number that would have looked far worse without the right measurement layered on top of it.

Why Native Platform Metrics Don’t Tell the Full Story

Joe’s clients running Reddit Max campaigns looked weak by native, in-platform metrics alone. Once his team connected a conversion API and layered in media mix modeling, the picture changed entirely.

“This has been the problem for visual medium forever, for all of us performance marketers. Oh, display doesn’t work, or billboards don’t work, because you’re not going to go rip a billboard off the side of the road and take it in and say, here’s my coupon. The world ain’t a vending machine. We’re not going to put a quarter in and get a treat, but you got to feed that top of the funnel and then catch stuff at the bottom,” Joe said.

Joe also added that if we didn’t have a MM, we wouldn’t be sure of the results. Once the MMM data entered the picture, feeding budget in or pulling it away produced results that tracked far more accurately than what the platform’s own dashboard suggested.

Fred drew the comparison out further: in-app metrics on a channel like Reddit can look weak enough to justify cutting spend, while a media mix model can show that same spend contributing somewhere further along in the funnel, invisible to the platform doing the reporting.

Joe extended the point to video and display advertising generally, channels that have absorbed this same “doesn’t work” verdict for years, simply because nobody can point to the exact billboard-to-purchase moment. Aaron used the moment to ask Joe directly what advice he’d give to teams who want to test unproven channels without an MMM already in place to justify the spend.

Joe’s answer traced back to his own history at a branding agency, discovering that a made-up word invented purely for a funny video campaign started generating real search volume once the campaign ran, proof that awareness spend was doing something even without a clean attribution path. His broader takeaway from years of that pattern was less about tooling and more about incentives.

“A lot of search folks, present company excluded, we think that we caused everything because we were there at the bottom to catch,” Joe said.

The Cost of Scaling Into Video

Joe’s genuine interest in Disney’s advertising platform ran into a wall almost immediately: production standards, not budget.

The minimum ad spend itself is a modest $500. Getting creative approved is a different problem entirely, since Disney rejected footage Joe considered broadcast-quality for falling short of its sound and video standards.

“That’s not going to cut it,” Joe said, of trying to lean on generative AI video tools to fill that gap for something like Disney. Aaron’s blunter framing: a $500 media budget paired with what can amount to a six-figure production and approval process.

Joe’s own relationship with YouTube and Demand Gen carries a similar tension. He’s been a public advocate for video advertising for years, and he’s still had to pull pure video campaigns entirely due to junk placements and low-quality traffic that placement exclusions couldn’t fully clean up, while Demand Gen has quietly outperformed standalone video for the conversion-focused work his clients actually need.

What Makes an Agency Successful Now

Fred handed this question to Aaron directly, framing it around AI’s growing role in ad execution and asking what’s actually separating agencies that are winning from ones that are struggling.

Aaron’s answer split agencies into two camps without much room in between.

“The agencies that are losing are the ones who are just doing what they did in the last era but harder,” Joe said, or ones leaning on a “proprietary AI solution” that every competitor now claims to have too.

The agencies winning, in his view, are winning on strategy, data, and efficiency, and on a specific skill that’s harder to fake: making complicated problems legible to a client without oversimplifying them.

“You guys make the easy seem hard, and the hard seem easy,” Aaron told Joe, contrasting that with what generative AI tends to produce on its own: overcomplicated answers to genuinely hard problems, without the judgment to know which parts actually matter for a given client.

Joe’s own read on the client side backed this up with something more concrete than strategy talk. Plenty of clients can write technically correct, keyword-friish ad copy on their own now.

What they can’t always judge is whether that copy actually differentiates them, or whether it reads like every competitor’s ad and gets scrolled past. That gap, knowing what to do with technically correct inputs, is where Joe still sees the clearest need for a human in the process.

Rapid-Fire: Ranking the Platforms

Fred closed the episode with a quick pass down Joe’s full platform list, asking him to rank what comes next once Google’s ceiling is reached.

Microsoft holds the second spot, largely for lead-gen clients chasing search impressions Google isn’t catching, plus company-level targeting that overlaps with LinkedIn’s. LinkedIn itself remains the obvious third step for B2B work, followed closely by Reddit, where Joe’s most successful use case has been community and subreddit targeting rather than the automated Max product alone.

TikTok earns a place on the list too, though less for direct response and more for cheap, high-volume email capture through lead forms, exported into other channels afterward. Quora, once a bigger part of Joe’s mix, has slipped down the ranking as ad quality and audience volume have both declined, something he attributes partly to a rise in AI-generated spam questions crowding the platform.

“It’s scary how well it’s worked, for how cheap it is too,” Joe had said earlier about a very different channel test, and that line ends up capturing the spirit of the whole episode. The platforms worth testing aren’t always the newest or the most talked about. They’re the ones where someone actually checked the real numbers before deciding.

Episode Transcript

Frederick Vallaeys: Welcome to another episode of PPC Town Hall. My name is Fred Vallaeys. I’m your host. I’m also CEO and co-founder at Optmyzr, a PPC management suite.

So, for today’s episode, we have a longtime guest who’s coming back to us. He’s been on the show a couple of times, and his name is Joe Martinez of Paid Media Pros. Joe has been on a tear giving a lot of presentations on a similar topic, and that topic is why do we keep running to Google, and that piqued our interest.

Google is obviously the big fish, the 800-pound gorilla in the room, even in this day when generative AI and people are maybe going to ChatGPT a little bit more. But Google’s still that big player. So why is Joe making the case, or asking the question, why we continue going back to Google?

I look forward to this episode going into a lot of details on what there might be besides Google, how to make Google perform better, and learning from one of the OGs in the industry who’s been doing this for a long time.

So, with that, let’s get rolling with this episode of PPC Town Hall. All right, Joe, welcome back to the show. Good to see you.

Joe Martinez: Yeah, it’s good to be back. Good to see you guys again. I know it’s been a while.

Frederick Vallaeys: Yeah. We also have Aaron on here. So, Aaron Levy, also at Optmyzr these days, and I thought we’d try something a little different with these podcasts, and we’ll see if Aaron can help me co-host. And, you know, Aaron is generally considered the funnier between the two of us, so I thought that might add a nice dimension to the show.

Aaron Levy: That depends on who you’re asking. But I think Fred and I have both been so jealous of the two-guys-in-a-comedy-podcast thing over the years, so we figured we’d give it a shot.

Frederick Vallaeys: Yeah, exactly. One thing I do miss already is Joe’s mohawk. Where did that go?

Joe Martinez: Age.

Frederick Vallaeys: What happened with age?

Joe Martinez: Well, I did promise myself once I turned 40 I’d get rid of it. I miss it though. But hair is falling out anyway, so can’t do anything about it.

Frederick Vallaeys: This episode, you may be hearing it before or after Hero Conf, but really, one of the topics, Joe, you’re going to give that presentation, but why do we keep running to Google? Tell people in 90 seconds, what are we talking about, why this question?

Joe Martinez: Yeah, I mean, for me it’s just personally what I deal with with clients. It’s something where we’ve seen a decline in performance from Google. In some cases, it’s not necessarily a decline in performance, it’s just costs have gotten higher, but when conversion rates maintain the same, but costs are higher, we’re just getting less in return.

So, not necessarily, not all the time is Google not working, it’s we’re just finding cheaper solutions, because not every year budgets increase, or sometimes budgets decrease. So we have to find alternative solutions to try to maintain the same amount of leads.

Since I am primarily lead gen, I don’t do shopping or e-commerce. Trying to maintain the volume with similar or possibly lower budgets, we have to find cheaper solutions.

Aaron Levy: Joe, do you think it’s possible? I know that you work with some smaller-budget clients. Something I’ve observed over the years is that there’s a lot more people who are B2B plus students at Google, since the industry has been around a lot more, there are a lot of people who are okay at running Google ads.

So I get the sense that you, and I’m hearing some of what you’re saying, that you’re looking for that area where you can be an expert and find really cheap stuff before anyone else can, and Google’s just a little bit too saturated to do that at this point.

Joe Martinez: Yeah, yeah, that’s definitely the mix. And I’ll fully admit part of it is out of sheer boredom. It’s more fun to work in a dozen-plus different ad platforms than just Google all day, especially when there’s less things to do in Google than you used to.

I’d like to learn and fill in those gaps, whether it’s for content or my own learning. I’ve definitely been playing around with OpenAI ads more and more, and clients are asking us about them, especially since they opened up the location targeting to more than just the country.

So that’s something where, to me, it’s fun to play around. And as much as I worked in Google all the time, I’d honestly say Google is probably one of the least amount of channels in terms of time that I work when working anymore, which is weird to say. I never thought it would be like that getting into this.

Frederick Vallaeys: Still a lot of the spend goes through it, but it just manages itself, given the systems that you put in place, by and large.

Joe Martinez: Yeah, yeah, I would say so. And I don’t know, we’ve just found that a lot of the automation does work, and then we learned fairly quickly if stuff works or not, being much more fluid to shift budgets around if need be.

Luckily, I don’t have any clients that are like, no, we have to have a different invoice for every ad channel. We’ve worked with those before and that’s not fun. All of my clients are fluid, like, no, this is our budget for the month, move it wherever, whenever. So it comes and goes.

But when we find stuff that doesn’t work in a channel like Google, like, hey, we tried AI Max for a little bit with extra budget, it didn’t work, boom, let’s pause it, we haven’t tested OpenAI yet, let’s move it around. We find where we can find wasted budget, and then try something new, and keep what’s working, and shift the other wasted budget around to something else.

Frederick Vallaeys: And so, I don’t know if it’s really true when you said a dozen platforms, that sounds like a lot, and I’m super curious, because there’s like the big ones that we all know about. But let’s jump in maybe one at a time here.

You mentioned OpenAI a couple of times, and that being one of the newer entrants, obviously, like you said, they’re adding a lot of capabilities, like more specific geo targeting, they’re also rolling it out to most of Europe at this point. So what has been your initial experience? How do you compare it to Google, and what do people need to know before they jump into that with OpenAI?

Joe Martinez: Yeah, it’ll get there, I believe it’s still under beta, correct me if I’m wrong, I thought it’s still under beta. And at least the, we’re not supposed to talk about it, I think, are we?

Frederick Vallaeys: (laughs) I don’t have a video about it yet, so I learned it from everybody else, blame who I read it from. It’s like the cat’s out of the box on this one, right? Even if it’s in secret beta, people are doing it, and Optmyzr supports it, you’re running ads for it.

Aaron Levy: Yeah, I believe it’s in Supermetrics already too, reporting on it already.

Joe Martinez: But yeah, they’re adding features like crazy, so they’re catching up to the same type of structure and features that we’re used to in Google and paid social accounts. So I think they’re working hard to get up to the functionality that we’re used to.

Not there yet, but the biggest complaint is, we don’t get any feedback of what was the user searching that led to our ads being shown. So we had one client, just had a meeting with this morning, July, they got a bunch of conversions this month, they didn’t, why, I don’t know, we didn’t change the signals, we didn’t change the ads, so we’re just pretty much fluctuating bids and budgets.

So it’s that lack of clarity of what’s being shown, to try to optimize our signals better, to see, are we putting in the right signals. The only other thing we could do is just create another ad group with different signals and see if that helps at all. So that’s probably my biggest gripe for it. But I truly believe they’ll start giving us some of that information.

Frederick Vallaeys: Joe, we’re prompting, we’re not searching.

Joe Martinez: It is.

Frederick Vallaeys: Well, it is interesting. Not to jump ahead, but I’m very curious what’s going to happen with what queries look like, because they’re not queries anymore, they’re a conversation. We’re having some chats with friends at Google who are hinting at challenges with all these search query reports, because they’re so long, because they’re all unique, they all have very specific information. What would you want from people like ChatGPT or Google to help manage things better, aside from anything?

Joe Martinez: Yeah, I kind of look at it like Optmyzr’s negative keyword tool. You find the phrases within the whole grand scheme of things, it might not be the exact actual search terms or keywords, but you’re finding little phrases that, whenever someone mentions this, it doesn’t convert, or something like that. That type of information, I think, would be helpful itself.

Even just overall topics or something would be helpful, or certain prompts. They’re collecting persona information too, so maybe some persona information of the type of person who’s searching for certain things, what performance do we get, maybe that could help.

Frederick Vallaeys: It sounds like the query mix change would be very important. But what’s curious in what you said is it went from being a good month for this one client to all of a sudden it’s not, the CPA got too expensive, or the ROAS was bad, but it sounds like the conversions just went away. So it’s like, what changed in the targeting? Did they just start showing ads for completely different prompts and search themes, or was there something else they changed in the back end?

And what do you do in that case? Do you just shift budget to another platform, and do you still go back to OpenAI at some point, or are you like, well, we’ve tried it, it worked for a while, let’s move on?

Joe Martinez: Yeah, I have to say, I’m lucky enough where I have clients who believe in longer testing. It’s not like we tried it for two weeks, it doesn’t work, turn it off. We’re pretty much on our first round of ad assets, sometimes the second round, so it’s more of, alright, we need to learn certain things here.

Also discovering, a little late, I’ll admit my fault, that it’s one of those things too, where they give you like 50 characters for a headline, but really only half of it shows anyway, and the rest gets truncated. You could just have bad-looking ads, people might not even read exactly what we’re saying, same thing with the description or the text. They gave you more characters than actually show up. So we’re still learning.

Frederick Vallaeys: So, at the same time, I’m curious on that one, how did you discover that? Because it’s really not that easy to see the ads that you want to see, your own ads, to preview them, right?

Joe Martinez: Honestly, it was going back in there and editing. I at least wanted to report on engagement initially, and going back and seeing, okay, what was the headline, what was the most visible part of the ad, and then that’s pretty much what we’re naming the ads is the headline.

And then I was going back in to kind of capture the best one, and I realized like all of them, when I was editing them just to look at stuff, had a yellow warning, like it’s eligible to serve, but it’s going to get truncated at a certain point. So I went to all of them, and they were all like that.

So we’re going under a recent change. And this is one where, due to legal reasons, the client needs to really help with the ad copy, because we can’t make false claims on anything, it’s in the pharma space, so they’re the ones that write the copy. So I try to help them out as best as possible, I said, here’s a new template, help me out, I’ll get to make it go.

Frederick Vallaeys: Amazed you got a pharma company to approve it.

Aaron Levy: You know, it’s interesting, I’ve had a philosophy based on what you’re saying, because I’m guessing when you put those ads in before, they didn’t have that yellow flag, and then OpenAI realized this is a problem, builds a new feature, so you’re like, oh, wait a minute, the rules of the game changed, the playing field changed, okay, we got to discount everything I learned last week, now we’re going to learn new stuff this week.

Are you able to test it all, given how fast things are moving, or is it just that, again, you’re playing a different game every week, so the stuff you learned last week doesn’t matter?

Joe Martinez: Yeah, it’s definitely new stuff. Like, I have not even tested out audience-type of stuff in there yet. Second client I’m using it for, they’re a service provider, so we’re lucky enough where it’s easy enough for me, where their target audience is every single person in this geo location, we could service everybody.

So when you just let it go like that, now, with a client like that, what we’ve noticed is the relevancy and time on site that we see within analytics, as well as their internal tools, because not everyone trusts Google Analytics anymore. But a lot of the stuff that we’re seeing is much more engagement.

It’s not just, oh, we visited the site for 10 seconds and we bounced, we’re seeing much more research. At least to be fair, we’re not sending them to a sign-up-right-away page, it’s more of an educational type of stuff, here are some guides, here are some resources. They’re doing a lot more of those actions from ChatGPT than what we’re seeing with some more awareness search and social campaigns.

Frederick Vallaeys: That’s interesting, because thinking about our benchmark report that we pulled, we’ve seen over the course of this year that it seems people are doing a ton of research on generative AI and are then going to Google and searching, so they’re converting at a higher rate and clicking at a higher rate, but less impression volume.

What you’re describing from OpenAI almost sounds like the opposite, which, as I think about it, makes a little bit of sense, because it’s not like a long conversation you’re having with OpenAI where you get your answer and then go solve your answer, it’s a little bit more like OpenAI is sort of kicking you somewhere else to do your research, versus what we’re seeing on search when people channel-switch. So it’s interesting, I wonder if, when we get better tracking, maybe the customers will be higher quality, because they self-vetted once they got there. Who knows?

Joe Martinez: Yeah, I am one person out of how many that use the platform. I use ChatGPT more for discovery research, like, I know the last thing I used it for was, hey, I’m going to Chicago tomorrow for a concert, are there any other events going on that’s going to clog up traffic or anything, those types of discovery. I’m not looking to buy something, I’m looking to learn, get an answer to something.

So that’s why we weren’t pushing necessarily the contact-us, and even for the service provider one, and for both of my clients on it, they’re not selling $15 sunglasses or t-shirts, you sign up for this thing, it’s hundreds of thousands of dollars for one, and a year-long commitment for another service provider, so that’s another almost thousand dollars.

So we know people aren’t going to become customers from ChatGPT, but it’s providing the resources, trying to make those signals available, that they discover us with those answers somehow, right?

Frederick Vallaeys: But that also speaks to the whole frustration that’s so long existed in B2B marketing and PPC, the classic example being printers. If you type in the keyword printers as a consumer, sure, you’re going to buy a $99 HP, but HP doesn’t want to spend a lot of money on that. Whereas if you’re a corporate buyer installing printers across your whole, like, 10 offices, that’s what they want, right? But you can’t disambiguate on just the keyword printers.

But that’s where ChatGPT, even if you just typed in printers, it’s already like, well, okay, Fred runs a company, he’s got this many offices, he might actually be looking for printers.

It’s been amazingly good. At one point I was like, I want to buy a new pair of shorts, but I want to try a new brand, what can you recommend? And ChatGPT comes back and it’s like, you bought this brand and that brand and this other brand, and I’m like, I recognize two of the brands, but the third one, I’m like, is it hallucinating? And I go through my closet and I look around and I’m like, it actually knows that I have this brand of shorts, it was right, and I totally forgotten.

And so that’s sort of that promise, is that memory aspect of all of these LLMs, that they end up knowing you at some level better than you yourself do, because it never forgets in some way. And so as a B2B, I’d be super stoked about, okay, I’m not going to get that many clicks, but when they do come in, they’re pre-qualified.

And even you talked about audience targeting, but to what degree is that necessary when its incentive is to show you the right ad at the right time, and part of that audience component is just baked into how you prompt. So are you seeing any of that with the advertisers that you’re running OpenAI for, in terms of quality?

Joe Martinez: Yeah, quality, I mean, I think so, just in terms of, one, we have gotten conversions from it in the past for the one that’s in the pharma industry, so we can clearly see where it’s coming from.

Frederick Vallaeys: And how long to conversion, how quickly did you see conversions when you started these OpenAI ads?

Joe Martinez: Probably a month.

Frederick Vallaeys: Okay.

Joe Martinez: Yeah, so, again, I don’t know what they converted on or for, but we have seen it tie in, for sure. And then on the other one, where it’s more like resources, and they target everyone in a location, that one is still fairly new, so that’s why we’re still looking at more of the engagement part of it, because zip code targeting is very, very new within the platform, so we haven’t been doing it too long.

Frederick Vallaeys: I’ll tell you another example of what we see, and kind of curious if you’ve seen this, but with Optmyzr, we do get a lot of referral traffic from the chatbots, but it sometimes overpromises, it thinks we can do certain things that, okay, we’re not the best at, so it’s not really the core value proposition of Optmyzr. And so we do get these trials, we do get quote-unquote conversions, but they’re trial conversions, they’re not paid conversions necessarily.

So when you’re measuring this with your clients, since these are like multi-month, big deals, are you tracking it all the way through, what is the lead quality, on top of just the conversion volume that’s there?

Joe Martinez: Overall, yes. Their main use, typically with anything paid social that is higher-level awareness-focused, they’re really looking at MQL quantity, and then after we get feedback, we’ll try to adjust if we feel like overall quantity is not where we want it to be. But for the most part, for them, quantity is the goal.

When something like OpenAI is a little bit different, compared to the bulk of what my clients do, they’re heavily LinkedIn-focused, we can really hone in on quality from the get-go, we get the information right away, like the company’s report in LinkedIn is absolutely phenomenal, and we use it all the time. I wish we had something like that from OpenAI, or like a Reddit tool, which we use Reddit a lot as well, to hone in on who is actually seeing our ads, even from a professional standpoint, are they in the right industry.

So I would say, from what we could tell in terms of engagement, it’s going good for quality, but it’s still too early to tell, because they have such longer cycles.

Frederick Vallaeys: And so, let’s make that comparison now. We’ve talked about OpenAI, which I think we’re all excited about because it’s the new kid on the block. You did mention how you measure a little bit of social, then you talked about LinkedIn and Reddit, so let’s maybe talk about LinkedIn and Reddit first.

And also, when we talk about LinkedIn, I do want to bring it back to Microsoft, some of that LinkedIn audience capability is available in Microsoft Ads. So before we fully jump away from your traditional search engines, how do you feel about Microsoft Ads?

Joe Martinez: Oh, we still use it heavily, I would say pretty much almost all my clients that are on Google are on Microsoft, especially in the lead-gen space. I really haven’t had a change of opinion on Microsoft over the years, for me it’s always been lower CPCs, much lower cost per conversions for the most part, there’s always the outlying ones every here and there, but just way less impressions.

Aaron Levy: Joe, we’ve talked about, I think you joked and said 12 platforms before, but I think you’ve named all of them. With limited budgets and all the obvious interest in this cool new kid on the block who just moved in, OpenAI, do you run into any resistance in testing other platforms? And how do you make those decisions about where the next dollar goes first?

Joe Martinez: Yeah, yeah, everyone wants search first. (laughter) We rarely get someone coming to us being like, hey, can you help us run some TikTok campaigns, (laughter) which I’ll admit, we’ve done it before, but it’s been a while since I’ve been on TikTok professionally. I’m on it a lot personally, which is bad.

Aaron Levy: (laughter) I don’t know if that’s a good qualifier, Joe.

Joe Martinez: Well, yeah, if, I said I’m going to a concert in Chicago tomorrow, I’m not going to Google to find out where I’m eating, I’m going to TikTok so I can actually see where the good food is in the south part of Chicago. Now I lost my train of thought, because I went on a tangent.

Frederick Vallaeys: How do you make the budgeting decision?

Aaron Levy: (laughter) Budgeting decisions for food and for work.

Joe Martinez: For me, it’s targeting, that’s somewhere, if a client says, we want to reach this person, I was like, okay, cool, we’re going to capitalize on search as much as possible, but here’s the volume that we can expect on average, about this much volume, you said you want to spend this much, there’s no way we can spend this on search, let’s expand. And then it’s just pulling audience options depending on where they want to go.

Typically for our clients, LinkedIn is the no-brainer next step, it is just too precise. Even as I think the LinkedIn platform has gotten less professional, we’re still finding it to be pretty good, and we haven’t really seen cost per lead impacted that much yet, we’ll see as that platform slowly changes.

And sometimes because of that, in certain accounts where LinkedIn hasn’t been working as well in the past, we’ve shifted some stuff to Reddit. Reddit is now becoming one of my favorite platforms, surprisingly enough. In a few of our clients, we’ve tested their version of Performance Max, so we just call it RMax, whether that’s the official name or not. And in certain cases, it’s scary how well it’s worked, for how cheap it is too, and that’s, no targeting, just, here it is, here’s some assets, let it run.

Frederick Vallaeys: Give people some perspective, you say a month, what kind of budget for a month before you start seeing results?

Joe Martinez: It took about a month or so, and some of them, it was the same client again, that service one, here are the zip codes, target everyone in that area. It’s scary how well, they were running, we started off with probably around 7K a month, now they’ve doubled it, so now we’re up to 15K a month. And this is the one that, if we didn’t have an MMM tool, we would think it sucked.

Aaron Levy: Mhm.

Joe Martinez: So, in-channel analytics doesn’t look great. When we hooked up the conversion API, when we attached it with the MMM, and gave it some decent volume of how it’s impacting overall performance, it’s a totally different story. And this is a company where we really rely on MMM data, we find it much more accurate than internal numbers.

When we look at the MMM data, and we make changes based off of that, we see more accurate results, in terms of when we feed it more budget, when we pull budget away, it accurately depicts, better than what we see with the on-channel platforms.

Frederick Vallaeys: In Google Ads, Reddit, the in-app metrics look a little bit weak, and maybe don’t justify continuing that spend, but the media mix model tells you that it is actually somewhere in the funnel, contributing towards those conversions, wherever they end up happening.

Joe Martinez: Yes, there you go. This has been the problem for visual medium forever, for all of us performance marketers, like, oh, display doesn’t work, or billboards don’t work, because you’re not going to go and rip a billboard off the side of the road and take it in, be like, here’s my coupon.

Aaron Levy: Yeah, exactly. And pay to watch Joe do that though. I’m gonna, in the spirit of being a jerk to Joe, it wouldn’t happen. (laughter) One of my favorite things is to be a jerk to Joe.

So I’m gonna ask Joe an impossible question. All these mediums that we know are sexy, new, shiny, every CMO wants to test them, you know that they will show literally nothing without an MMM. What advice do you have for people that want to test these things but don’t have an MMM?

Joe Martinez: Oh yeah, yep, that’s definitely part of the… (laughter) I worked, probably the second agency I ever worked for was a branding agency, and that’s where I discovered video advertising, this was my first taste into YouTube ads, or video ads, when they were called lightbox ads.

Frederick Vallaeys: Was that right? That was the term, lightbox ads, within Google.

Joe Martinez: All right, so I was at a branding agency at the time, and at first I was like, it’s not all about the brand, why are we wasting all this money on commercials and all that. And you saw it reflect, this is when I first discovered Google Trends and all that type of stuff, where you saw the impact of it, and we continue to see that more in terms of traffic and everything.

I will give some of the platforms credit, that we do get some better attribution reports, I still do look at that as best as possible within GA4, even though it will never tell the whole story, but we still get more information than we have in the past.

But I always tied a lot of stuff into brand traffic, even with search. So Michelle and I, when we try to launch clients who want to do a big branding one, we always try to heavily encourage running some brand campaigns alongside it. And a lot of times what we see is, once we do that, whether it’s Microsoft, Google, whatever, we see a lot of those campaigns be limited by budget, as we launch big awareness campaigns, because we are seeing more.

This is an example I gave a long time ago, and it’s funny, this is when I was working at that brand, there was a video initiative, it was a funny ad, they had a made-up word that they invented, they’re trying to be funny. So we ran a search campaign, and we put the brand name plus that made-up word with it, and all of a sudden it started to get search traffic. So that was our way to prove to them, people are seeing it now, now it’s showing up in the search term or the keyword planner tool as it’s got volume now.

So I know from my experience, I know you’re asking, how do you explain it to a client, it’s just, out of my dozen years of experience, it’s just what we see. The world ain’t a vending machine, right, we’re not going to put a quarter in and get a treat, well, we’re not going to put a quarter in anyway because you got to use the little swipy thing, but you got to feed that top of the funnel and then catch stuff at the bottom.

And I think that over the years, a lot of search folks, present company excluded, we think that we caused everything because we were there at the bottom to catch. But having that conversation about how you feed the top of the funnel, and use all these great tools that we have, and all these audience metrics, and all this language information that we have, telling that story is going to be a huge part of people’s futures, if they’re not good at it already.

Frederick Vallaeys: So you’re a step ahead somehow.

Joe Martinez: Yeah man, I’m trying to get on other stuff too, like, I’d love to get on Disney ads to play with that, because their targeting is so good. We’ve done it before, but it’s been a while, I want to do more of the Disney advertising now that everything is combined, once Hulu goes away and everything’s rolled into Disney, or Disney Plus.

I’m very curious, because they have a lot of cool business targeting options there, like all the first-party data that Disney collects, it’s crazy. And we have clients asking about TV stuff all the time, and I was like, it’s a $500 self-served minimum buy-in, do it, for a lot of my clients, that’s nothing.

Frederick Vallaeys: That’s nothing in terms of money, right, but I think the hold-up is usually like, I don’t have videos, I know how to write words and do text ads. So, obviously you’re a big video guy, but for someone who doesn’t have your skills, where do they start? Is generative AI good enough to generate some of this video content, or where do you go?

Joe Martinez: I think for Disney ads, no. I’ve made a couple videos about Disney ads for our channel, and I used client videos, because our videos got rejected, they weren’t up to the quality standards that Disney wants. I had videos from clients that I thought were pretty professional, TV-worthy, some of those got rejected because the video quality still wasn’t good enough, the sound quality didn’t meet their standard.

So from that one, they want high-quality TV-style ads, where if you try to go into Google’s AI tools now, in their planners and everything, that’s not going to cut it.

Aaron Levy: Yeah, use that stuff for YouTube and Demand Gen, fine, if it’s okay, but so we’re not talking about a $500 budget to try these video ads, right? It’s your $500 ad buy, and your million-dollar production budget, and then maybe if you’re lucky they’ll approve it, or send you back for another $100,000 worth of sound editing. I’m probably exaggerating the numbers here, but, right, it’s, so, you want to react to that, I see you want to.

Joe Martinez: Well, yeah, they had something, because one of my clients did ask about it, and they’re like, they’re probably going to do it, but what about the part where you can get a QR code on it, there’s some extra feature or reporting that they wanted, and that required like a minimum, 100K ad spend, to get these additional reporting features. It reminds me of Spotify and Netflix, same thing, they’re like, $50 minimum if you want your ad to show one time on one crappy podcast.

Aaron Levy: Not a good one like ours.

Joe Martinez: (laughs) It is interesting though, because there was so much success with UGC and things like Demand Gen, well, Meta first, Demand Gen now, and so everyone got conditioned to do this UGC or creator-led stuff, and Disney, of course, is like, well, we don’t want any infomercials, draw a real ad.

Frederick Vallaeys: But so that almost determines your media mix in a way, whoever has production budget can do the Disney stuff, whoever doesn’t is going to have to get really good at YouTube.

Joe Martinez: Yeah, and that’s, we have messed with Demand Gen more, I know I wasn’t happy with it initially, it’s been two years now though, when video action got shifted to Demand Gen, I wasn’t fully happy with it, but again, we’re seeing better performance now.

And that hurts, as someone who’s talked about YouTube for the longest time, we’ve had some where we just turn it off. No matter how many placement exclusions we add to a video, a pure video campaign, especially one that’s created without goals guidance, the amount of junk placements and low-quality traffic we’re seeing from video campaigns now, where we switch it to Demand Gen, which is a more clickable, actionable format, that’s the way to go.

Does it work every time? No, but if it’s conversion-focused, especially from the lead-gen perspective, we’re seeing Demand Gen just blow standard video campaigns out of the water. And it’s, in my presentation, they moved standalone video campaigns, it’s in the awareness campaign objective category now, so they’re not really pushing video as a conversion-type campaign objective anymore.

Frederick Vallaeys: I believe that, that makes sense. And speaking of video, you have a channel right on YouTube, it’s been a little…

Joe Martinez: Yeah, it’s a little light on content.

Frederick Vallaeys: What’s going on, you don’t believe in video anymore?

Joe Martinez: No, we do, see, here we go, this will tie back into a previous conversation, it’s like, why do you have it, why do you leave it up there, I was like, the amount of business that we’ve gotten from it is crazy, no one, that’s not going to be tied into our analytics, we’re not going to see that in our CRM, and these are not even ads, this is organic video, it’s content.

But the amount of people that reach out to us on our website, or book consulting time, did you hear about us, oh, we saw the YouTube video, that’s stuff that we don’t always record all the time, but that’s the proof of getting awareness out there leading to business.

Frederick Vallaeys: There you go, I’ll have to remember that one.

Joe Martinez: There you go, that answers your question. But yeah, it’s been busy for Michelle and I, it’s been a very busy year on the client side, so zero complaints about that one. And since we are focused on Hero Conf, this will be our third year doing the Google Ads workshop, a lot has changed over the past year, so we’ve had to update way more slides this year than we had to do last year, and we’re still working on it, with less than a month to go. (laughter)

But once Hero Conf is done, we’re going to be going back to a standard video cadence again, we’ve had two videos in like the past five months, which is not good, and we’ve had many people reach out to us being like, are you doing video again, yeah, we’re trying.

Frederick Vallaeys: But you’ve kept the money coming though, that’s the truth. Still just you and Michelle?

Joe Martinez: Yeah.

Frederick Vallaeys: Great story though, that you’re so busy, and obviously what you’re doing, just not putting all your eggs in Google’s basket, seems to get client success, more clients are coming to you. Now, I don’t know that that’s the story for every agency out there, and maybe Aaron, you have some thoughts on that, but what’s the playbook? Someone who’s maybe seeing a lot more pressure from AI managing ads, from AI-generated assets taking over the work that sometimes was done, where do you guys see yourselves, as the two humans in the agency, what is making you successful?

Aaron Levy: I will, Joe, I’ll start first, because mine’s going to be more top-level, and then you can go boots-on-the-ground, because you actually deal with this. I will say the fun part about my job now, working with Fred instead of working in an agency, is now agencies tell me everything, and they didn’t used to tell me anything before.

(clears throat) I want to tie this back to what I think Joe and Michelle do very well, but I’ll start with the negative first. The agencies that are losing are the ones who are just doing what they did in the last era but harder, or if they are relying on, we have a proprietary AI solution, so does everybody, and you can’t sell it.

But the people who are winning are winning on strategy, they’re winning on data, they’re winning on efficiency, and they’re winning on making their clients look good. And the way you make clients look good is not just in performance numbers. We spent a half hour or so talking about ChatGPT ads, that the eventual conclusion was, they don’t work very good yet, keep trying them.

Joe Martinez: Yeah.

Aaron Levy: But that’s the way you make your client look good, is that you have those answers, you have that information, you don’t just go into ChatGPT and ask it your client’s email and answer it back, that’s how those agencies are getting fired. The ones that are keeping their thought work, and are doing things really well, and are explaining things, which, going back to you guys, the rare compliment I’ll give to Joe and Michelle, we like each other very much, we’re like siblings, I just like messing with them.

Joe Martinez: I’m gonna make him repeat it too, once I get a compliment out of Aaron, I need to make sure I heard it right.

Aaron Levy: So, you’re going to have to say, wow, I was looking for the… I was looking for the Joe’s butt card for this.

Joe Martinez: (laughter) Joe knows what that means.

Aaron Levy: You guys make the easy seem hard, and the hard seem easy. What I mean by that is you will give very detailed answers to very complex questions, in a way that’s easy to understand, where I find that, in its current iteration, most generative AI gives you too complicated of a solution for a hard problem.

You guys tell people how to do it, you also explain the basics, but explain how you can make them more advanced. That’s making the easy seem hard. So it not only shows that you guys can solve really complex problems, and you’re willing to help other people do it themselves, but you’re also not just going to flip on a smart campaign because it was a best practice, you’re going to get a little bit clever as to how you do it.

That’s the thing that you guys are doing that makes you really approachable, makes it very salable. That’s also what I’m hearing from winning agencies, is that they have either honed in on a niche where they know everything about it and they’re the best at it, or they have gotten really strategic in a certain area, or they put out lots of data reports that they can share and make clients look better, and they put out POVs and thought work.

Being an agency that’s a really good executor right now, those are the ones hurting. The ones that are really good thought workers are the ones who are winning. Was that accurate, Joe?

Joe Martinez: Thank you, thank you, that’s heard, people are just listening.

Frederick Vallaeys: (laughs) Joe’s shedding some tears over there. Oh wait, this is the audio-only version, isn’t it?

Joe Martinez: (laughs) Guess I could have kept the mohawk then, for the people listening in the car. So, what do we need to describe here?

Yeah, no, we still get a variety of company sizes, we don’t work with the one-stop local mom-and-pop shop, that’s still a little bit too small for us. Our clients’ budgets probably range from 5K to up in the million-dollar range, ad budget a month type of thing.

In between that range, no matter the high or low spectrum, we get clients who know the channel, they know what it can do, but they’ve all been overall marketing people, they’ve never been in the channels themselves. So taking what they need to do and transferring it to an ad platform, I think that part is still very valuable.

We’ve seen clients just set it and forget it on all the automated stuff in Google, and Reddit has the… what is it, not Advantage, the automated one, where you can just—

Frederick Vallaeys: RMax, you called it before.

Joe Martinez: Yeah, yeah, RMax. And Quora has a fast setup, I think Pinterest even has a fast setup option, just, here, give us these three things, we’ll let it run.

We have worked with several brands that you’d think they’d have the internal team to know not to do it, and want to pull all the levers, they still don’t do it, or not even know where all the levers are in a platform to do it, those hidden features, all the boxes checked that are under a few different dropdowns that you got to turn them off every time, but they’re always hidden.

It’s those little things, and that’s very technical, but I’ve always been more interested in the marketing part. I have clients who have to write the ad copy, that’s great, but I still need to review it, you’re saying everything right, you’re saying all the stuff that’s keyword-friendly, and everything you’re saying is accurate, but they haven’t done the work all the time to really say, is that going to make an impact, or are you just saying every single thing that all the other ads are saying, so you just blend in and everyone’s just going to overlook it.

Are you really looking at the image assets in our paid social, to really see, is it going to, if someone watches this on mute while they’re slowly scrolling by, are they going to stop and pay attention, or on the Meta side, are you actually putting all the physical words you need in your assets, so it has a better chance of understanding what you’re doing, if you are doing more of the accelerated stuff.

So there’s so many what-ifs, and ins and outs, that one just comes with time and experience, but it’s still not something that you can just flip a switch on, import something from another channel, and just let it run, and it’s going to run immediately and perform well. I still love to do that, I like the challenges, I don’t mind the changes, I don’t complain about the changes that the platforms make, even though I don’t like them all, but I kind of like that challenge stuff, and for me, I get bored easily, so if everything stayed the same year after year in this industry, I would have been out a long time ago.

Frederick Vallaeys: Nice. So it’s great to have someone like you who just wants to jump in and try it all. But realistically, you mentioned 12 platforms, most people, they’re doing Google, that’s a given. Is it better to try one of these automated systems from one of the platforms you want to jump into, to see if there’s something there, before you engage with learning it more deeply, or going to your agency, or is that the wrong approach? Do you have to go all in right away to see the result that’s going to make you want to stick around?

Joe Martinez: I don’t think so, I don’t think you always have to jump in immediately from the get-go, and that’s not even, I’m discounting what all the reps from all these channels recommend, I still see good and bad reps that, on their recommendations of, oh, just here, turn everything on, type of thing.

I think sometimes part of the problem is, it’s one thing to turn it on, it’s another thing to turn it on with the right conversion tracking setup, with the right expectation of how many clicks, how much budget. How many clicks do you need to get a conversion, based on the average conversion rate in your industry, and what is the average CPC associated to that, on the platform you’re going into? If you don’t know that answer, then you don’t know.

Do I need $1,000 to even see if this works, if I have a 95% probability, or do I need to spend $10,000 to get to 95% probability of knowing if this is the right setup? And then even if you’ve done that calculation correctly, but now you have set up your conversion tracking incorrectly, so, okay, you’ve spent the $10,000, and you didn’t see the needle move on the business, but again, the measurement wasn’t there, so what do you know that now you can do something with?

I think that’s the risk, that people jump in too quickly into these newfangled methods, and the newfangled methods, they’re good at automating the bidding, the targeting, and the creative, but knowing how that impacts your business, that is a black box, and if that black box isn’t shed light on, they’re never going to be able to do a good job, and it really is just money down the hole.

Frederick Vallaeys: Yeah.

Joe Martinez: One thing I hated, probably within the past couple years, was Meta taking away all these targeting options and then trying to put all this Advantage+ stuff into it, to make it really automated, I initially hated it. Eka DeFazio, she talked a lot to me and just educated me when I was trying to do more Meta stuff, and really helped me better understand how the automation can really be a positive functionality within the platform.

At first I was like, no, I’m going to create something in the asset studio, one ad, it’s going to have one for each type, very structured. And then she educated me, no, give it multiple assets, this is how your assets should be created, and changed my mind, that the automation here works really great when you feed it the right stuff.

That being said, not only Meta, other stuff from more purely lead-gen, having all the offline stuff imported back into these channels, is a huge difference. There are certain accounts, depending on what they are in the industry, if they have such high-ticket items, if it actually closes, we’re going to get very few conversions, I’m to the point of, if you don’t have a way for me to import some of this information back in, I don’t want to work with you, because it’s going to be extremely tough, and then you’re going to complain, and we’re going to have to try to prove if it’s working or not.

It’s using all the conversion API stuff in Meta, Reddit, and having offline data and CRM stuff imported back in, I see a night-and-day type of result between clients when they implement all that stuff.

My friend and former co-worker, Michelle Merkelin at Tenuity, we were in a secret meeting with Google, and her feedback to them, which I’ve stolen a lot, and I should probably pay her royalties, but, get rid of the words “simple” and “just,” as in, Google’s notorious for just, turn this on, it’s that simple. It’s really complicated to make something really simple work. So that sort of language coming from our platform partners devalues our work as an agency, and we’re stuck explaining, Joe, exactly what you just described, this automation seems like a nightmare, oh my god, I have no control.

Frederick Vallaeys: No, you have tons of control, you just have tons of control in a different place that they aren’t necessarily telling you about, that’s where the expertise still comes in.

Joe Martinez: Yeah.

Frederick Vallaeys: All right, sounds complicated but also promising. Joe, let’s do a little bit of a lightning round here, we had a list of 12, I think we’ve covered maybe, what, five or six?

Aaron Levy: Mhm.

Frederick Vallaeys: Just go down the list, you get what you want from Google, you think you can’t get much more from that without blowing your ROAS or CPA, where do you go next?

Joe Martinez: Microsoft, if I’m speaking for my lead-gen clients, at Microsoft I try to capitalize on that little bit of search impressions we’re missing out on, and we have done some audience ads, like you mentioned before, the LinkedIn targeting, the company ones, they don’t have many options to target, you can find almost any company on LinkedIn, but not every company available to target on LinkedIn is available to target on Microsoft, so sometimes you need a little bit of the bigger companies there. So Microsoft would be second, for sure.

Frederick Vallaeys: Go right down the line.

Joe Martinez: LinkedIn, without a doubt.

Frederick Vallaeys: LinkedIn. And again, this is B2B, so lead gen, that makes a ton of sense. For everyone listening or watching, we’re not going to talk about shopping, so shopping, maybe you’d look at Instagram feed-based ads. But, lead gen, who’s next after LinkedIn, is it Reddit?

Joe Martinez: I would say Reddit, one, more than just the Max-type of stuff, we can find communities and subreddits to target that are really spot-on. I haven’t used it yet, but a couple months ago, Reddit did do a beta announcement for lead-gen ads within Reddit too, so that’s something I need to test, haven’t done it yet, but that’s available now.

Frederick Vallaeys: Any hidden nuggets that people might not even be thinking about?

Joe Martinez: I surprisingly enough still do like TikTok, they have the instant forms and lead forms. In my industry I use it mostly for a type of retargeting, and then in a few of my clients, I’ve used it more for, they wanted to build their newsletter database, so it’s something where, sign up, we’re only asking for email, let’s get a lot of cheap, loaded-up sign-ups, and then use those customer lists in other platforms.

We could take that export, saying all of these come from TikTok, upload it into other channels, so we know anyone who converted from this campaign or ad group, from this customer list, initially came from TikTok.

Frederick Vallaeys: Cool. Yeah, if I had to guess what’s next on the list, it’s been a while since I’ve used it too, but I did like Quora. I knew it was coming, it had to come at some point, Joe was the king of Quora for a number of years.

Joe Martinez: Yeah, but it has slowly gone down the list, as I think the audience quality has gone down, a lot of it just seems like weird AI-generated spam questions. However, if some topics are still highly frequented in terms of traffic, it’s just the audience, the impressions and volume is just so low, I’m seeing it lower than it was years ago, in terms of who I’m trying to reach. So, yeah, that is something I’d look at, but it’s still on the lower end. So I would probably put Meta back up front over that one.

Frederick Vallaeys: And when you say Meta, you mean Facebook, Instagram?

Joe Martinez: Yeah, I would lump them all together.

Aaron Levy: Yeah, we leave Threads on too.

Joe Martinez: Yeah, all that together, I’d say just Meta as a whole.

Frederick Vallaeys: Well, this has been great. People who want to dive into this a little bit more, you are doing this presentation in a number of cities, there is also the workshop that you’re doing, so we’ll put all of that in the show notes. If people are going to be in San Diego, Joe said he might be on the morning run, so for those of you who can make it up by like 6:00 a.m., get a bit of free consulting in while doing a 5K.

Joe Martinez: Got an option on that, I could talk and run at the same time, no problem. I’ll be leading the walkers, if anybody wants a dainty jaunt early in the morning, that’s better, for free consulting I need a donut or a burrito at the end of it. (laughter)

Frederick Vallaeys: That sounds good.

Joe Martinez: I’m talking about that crack tacos, I want the burrito with the fries inside and everything, this is like my one cheat trip out of the year, or I negate all my run completely afterwards. (laughter)

Frederick Vallaeys: The way it should be. Well, very good, gentlemen, thank you for joining me, everyone, thank you for watching. I hope you’ve enjoyed the episode and learned a thing or two. And if you did want to find out more about Joe and Paid Media Pros, please check out his YouTube channel, his website, we’ll put all that in the show notes, he’s also on LinkedIn.

And if you want to know more about Optmyzr, obviously reach out to myself or Aaron, we’re happy to show you what we’re doing there, and some of the cool new AI and MCP capabilities we’ve just launched in the Anthropic directory as well. So with that, thanks for watching, please subscribe, and we’ll see you for the next one.