21 Paid Digital Advertising for the Present and the Future - Paul Rakovich May 19 2022

Erik Martinez: [00:00:00] Welcome to today's episode of the Digital Velocity Podcast. I'm Erik Martinez from Blue Tangerine,

Tim Curtis: and I'm Tim Curtis from CohereOne.

Erik Martinez: Today we have Paul Rakovich, who is the CEO of Clicks and Clients to talk to us today about paid media in 2022 and beyond. Paul, thanks for joining us today on Digital Velocity.

Paul Rakovich: Hi guys. Thanks for having me today. Pleasure to be with you.

Tim Curtis: So, I'm curious, before the show, we were chatting and kind of laughing about some of the things that, we've all experienced in our careers, and you know, Paul, you go back, we were kind of joking, you know, BG, before Google, and one [00:01:00] of the things I'd love to hear just from your perspective is maybe a little of the lessons you've learned growing your company. That's a pretty broad time span, but you've certainly seen a lot in that time.

Paul Rakovich: It was fun to reminisce about the old stuff. The old bid jamming and overture, and so someone would bid a buck. Then you'd jam at 99 cents. Then they jam you at 98. Then you log in an hour later and you'd bid a buck 20, and that was the game with misspellings and just basically pay to play. If I would bid more on the term, I would get it. It didn't matter how good or bad your ad was, and then that changed, and then Google came in. Then social came in.

The ones that always stand out to me are cracking the code of these difficult problems. Those are some of the funnest ones. I feel like right now we're in the mix of it with an account again. There's some cool stories. Like, getting this account to go from like 50,000 bucks a month to 300,000 bucks a month to a million bucks a month, but like along the way, I thought the client was going to fire us. A client came up with these really great ideas. We came with great ideas. [00:02:00] They totally stunk. They completely bombed, and it's like, iterating on that and trying to crack the code and gain insight to move accounts forward. So, I think some of the bigger lessons are like nothing lasts forever.

Overshare was great until it wasn't great, and Google's still pretty great, but it's changed. Facebook, we were talking about, was gravy train awesome until it wasn't. So, it's like the fundamentals of marketing are still there.

I think the other one is the most successful clients and the coolest things we've seen, like, they just play the game differently. They're focused a lot more on lifetime value. They're focused more on like tweaking and refining things along the way, versus if I change this headline from black to red it triples sales. Those kinds of like silly things that aren't the reality of getting in and doing the work.

It's hard work, and most stuff doesn't work and it's frustrating and difficult, and every once in a while, you borrow something from old times that comes in and saves the day, but I feel like we're still faced with [00:03:00] those, this should work, it didn't work, now what the heck do we do kind of moments? We're right in the thick of one. It feels exactly the same way. Those are some of my memories. Copywriting is still great. Super important.

Tim Curtis: When you're talking about digital, you know, there's a lot of times the conversation tends to drift to the tactical cause there's so much here, there, and everywhere. You're focusing on bid score, bid quality. There's all of those kinds of things that are obviously meshed in, but I think taking a step back and looking at that overall strategy for your digital, in particular, your pay-per-click, both within the social and the Google-type ecosystems.

One of the things we were mentioning before the show was talking about Apple's intelligent tracking prevention and the shutdown of the third-party cookies. Of course, that's already gone through in Apple and iOS and we're seeing just massive impacts from that. Obviously, Google's set to do that next year with the Android side. How are you seeing the impact of this for your clients, and then how are you sort of then pivoting, based off this pool of knowledge you have from over the years, how are you then pivoting to recommend to them to, [00:04:00] not necessarily get around the issue, but get their hands on it and continue to be successful?

Paul Rakovich: Learn to live with it and learn to adapt. There's tactical pieces. Like on Facebook, you know, have the API connected, right? So, you have the API connected. Now you're getting a little bit more data through and like just making that change to be more successful. There's execution pieces with things. Certainly, that can be helpful. Demographic targeting, geographic targeting, but it still comes back to like the stuff we're working on right now.

So, there's the tactical side of it on those big spenders or not even big spenders, but hey, we have to have the Facebook API connected. If you want this to perform well like you just got to do it and suck it up and figure it out. So, there's stuff like that, but we're still going back to creative. We're still going back to copy. We're just really having to sharpen the sword on things we're looking at.

We have a client where desktop leads are quite a bit more valuable than mobile leads. Most people have kind of forgot about desktop and in Facebook, it's not a whole lot, but in Google, like [00:05:00] there's still a decent amount of desktop flow. So, we're going back to accounts and saying, step one is a lead, step two is a booked appointment, step three is a shown appointment. Can we break this down by device? Maybe we have been a little scared to have more expensive leads quote-unquote, but maybe these desktop leads are like a lot better.

So, that's where we're kind of borrowing from other clients to really sharpen the sword on things. So, I think there's both tactical pieces, but then there's also just back to copy, back to creative. How can we say this better? How could it be more precise? I think we're just really having to sharpen the sword. So, it's get Facebook back to where you can get it, as close to where you can get it, and then you just got to sharpen the sword on every angle at this point.

Tim Curtis: You know, you've had some clients that have really seen this huge drop in Facebook performance that we were talking about.

Paul Rakovich: Yup. It was so easy. You just put a pixel there and said, give me people that buy this and set the audience broad. Don't even target. Just give me [00:06:00] people that buy this and Facebook's like, here you go, you know, silver platter, and now it's like, whoops. So, that broad doesn't quite work, but also going like ultra segmented doesn't quite work.

Originally, there were these tools that would separate out. Well, how does the audience that likes this versus this perform, and then you'd run them and then try and slice and dice the data if you could. It's like that doesn't work anymore. So, it's almost this happy medium from a targeting standpoint to try, and I think probably layer in as many different data points and to try and get a little closer. So, you can't be super broad. You can't be hyper-focused. You got to really find the happy medium, and even then that doesn't always work.

Tim Curtis: That's the frustration.

Paul Rakovich: Yeah. You might have the happy medium, but do you do a bid cap? Do you do a cost cap campaign or lowest cost campaign? Do you put in all the states, or do you put in your top-performing states? Does a lowest-cost campaign work up to a thousand dollars a day, but when you raise the budget to $5,000 a day, it completely falls apart? It still feels [00:07:00] like there's a lot of massaging and it feels like even that massaging is almost on an account-by-account basis.

Erik Martinez: And so that implies that there's a lot of testing.

Paul Rakovich: A lot of testing.

Erik Martinez: I was just having this conversation with the folks at iHeart Media yesterday, talking to them about a particular client, and I kind of feel like some of the digital tools that we have today are just exactly what you said. You can hypertarget, but then you hypertarget yourself so narrow that there's no audience to reach, right?

Paul Rakovich: Yup, yup.

Erik Martinez: Or you're getting to that audience, but you're already getting to them anyways, somewhere else, right? So, you have to have this layering mentality of multiple types of media, both digital and traditional, to help move that needle. I think the question is for then to know that anything is working, you have to have a game plan, and so the question is what is that game plan and how long do you need to measure [00:08:00] it to say this is successful or not before we move on to the next thing?

My paid search team at one point in time, I remember them telling me, I can micro-test something for two or three days and then get results of move and that's not true anymore. We can't seem to make that work. So, what is the right set of intentionality that you're going after, and how long before you know it's successful or not?

Paul Rakovich: Yeah, I totally agree. It almost seems like with Google too, you could just change, it's not TCP anymore. You could like change a bid and tomorrow will be different, and now it almost takes a week to do it. So, what I look for, the clients have to have an appetite for spending. I think to one of your points, Tim, it's feeling more and more like you have to have a lot of dough to throw at this to get stuff to work. You're not going to do it on a $ 50-a-day campaign.

So, in these big accounts with a thousand, $2,000 a day as like a test campaign, we're still trying to get out of learning to feel like we found [00:09:00] something out or like have an answer, but in the meantime, we're looking at things. I'm trying to make a judgment call of is this a game we can win?

So, we had this idea the other day around setting up a campaign a certain way. It's a targeting method by geographies and everyone's like, this is cool. Let's do it, and we do it, and everyone's like 1500 bucks a day. We're targeting a lot of different areas. It's going to take some time and we're on day three and the CPMs is like over a hundred dollars. So, the cost per link click is like 10 bucks. So, we're looking at it right now, three days in, going like, is this a game we can win, and then we're trying to pull from other accounts of like, when we did this, did it start off at a hundred, go down to 50, end up at 40? Not that CPM is like the end all be all.

The thing I tell the team is like, are we playing a game we can win? So, if we're at a hundred dollars CPM, our cost per leads coming in 4X what it's supposed to be. Is this the game we can win in the next two weeks, and if it is, or we think it is, let's [00:10:00] place the bet, but let's be pretty confident about it.

So, I'm feeling like it takes a couple of weeks with a decent amount of budget. Got to get out of learning on Facebook. Gotta be patient, but I do look at some leading indicators and get a little antsy. On this project, particularly. Everyone's cool with the testing plan and they're cool with losing some money until they're not, right? Even if they're like, yeah, that's a hundred bucks a day. No problem. No problem, and then we're going to hang for two weeks. Everyone in agreement? Yeah, no problem, and then we're four days in, the cost per lead's a hundred bucks. Everyone's like biting their nails.

So, I try and stay a bit ahead of where I think the client's going to go with that, but yeah, it takes time. You gotta be able to break some eggs, and I just feel like it's taking more and more budget to test stuff out. Like at 50, 100 bucks a day, there better be a decent amount of conversions behind it to get some data, or else it's going to struggle, in my opinion.

Tim Curtis: The runway to success is shorter. You have less time now to prove that success because the aptitude there for spending like that, it's pretty diminished. You know, people are being very judicious about their dollars [00:11:00] because you've got these businesses that have built, you know, themselves on Facebook. They use Facebook as their primary advertising vehicle or their primary lead generation vehicle. You know, and all of a sudden somebody moved their cheese in a big way with these privacy changes and you know, it's just not as successful. So, that's a rude awakening if you will, and now, to your point, people aren't just hanging out and letting that credit card run for a week without looking at it. That's just not happening.

Paul Rakovich: Yeah, but they have to. They don't want to, but they kind of have to. There has to be an appetite to crack some eggs. There just has to. Even if it's a loss, there better be a pretty decent learning lesson out of it, and that might save it as well. Like, wow, this completely fell on its face, but there was a couple of glimmers of hope. So, the next iteration of it is, instead of targeting these 60 geographic areas, we're going to try the top five that actually performed. That might be the saving grace of this campaign. That's looking a little rough, but it's only two days in. It's not out of learning and everyone agreed this is going to take two [00:12:00] weeks, so.

Erik Martinez: You said that before companies focus on Facebook ads or any kind of paid ad, that some things need to be fixed first. What are those things, in your mind, that need to be fixed before they start pouring money down the Facebook and Google path?

Paul Rakovich: I'm in this online ski group and it's like the coolest thing I've ever done, and it's like the most amazing service ever. They were talking this morning and the guy who runs it was like, hey, can I get the feedback from the group? What do you think is better for a headline for the website, learn to ski online? You know, it was all about them, all about the product, not benefit-driven at all, right?

I think going back through, there's some suggestions about the story brand piece. What's the benefit? What do we want people to do? If you have these five different ideas, what's the measurement that you're going to go off of that this is successful? Is it time on page? Is it clicks deeper into the site? Is it people signing up for the course? How would you measure quote-unquote success?

So, [00:13:00] I think being clear on what the goals are, at least some measureables. Even if it's not like you gotta have something right off the bat. Even if it's like an email opt-in. Just something to measure off of, and then you could start to throw some traffic at it, knowing I'm measuring something. I think that's really the first step. So, getting kind of the house in order around benefit-driven, clear call to action. Where am I? What can I do here? Why should I do it? Then a basic measurement before people start to turn traffic live. I think that's really important.

The other piece is just high intent keywords. Like, is there volume there for people looking for this, if we're going into Google specifically, and if there is and there's other people advertising on it, then I'm usually like, all right, cool. This is going to work. I know I can play. At least play, if not beat the folks that are in here. At least play. So, that's the other piece, I think, is really important to look at. Is there high intent keywords for Google specifically?

The example that I was giving someone yesterday. My [00:14:00] brother and sister-in-law have a charter business in the Virgin Islands and there's keywords exactly for that. Excursion, XYZ island, day trips, private charters, So, it's like, I know we can go in and nail that. Can we go in and get people that are searching for like, things to do on ABC island and then decide like, oh yeah, maybe I want to go on a boat? Yeah. Okay. I'm going to sign up. Probably not. Is there a certain glamour on that specific thing? If not, you're going to have to generate demand or interest or get someone to turn the corner. I think that's a much more difficult game to play.

Erik Martinez: That makes sense. So, are you guys, are any of your clients using AI to manage their budgets or spends? Are you using those kinds of tools? Are you seeing them be effective? I mean, and I've always been on this position. Our agency works with midsize accounts. So, not necessarily the budgets that you guys are working with on a daily basis, and one thing that's always seen true, when AI is involved, you [00:15:00] still need a lot of data.

Paul Rakovich: Yeah. I think back to one of my early clients. He was a social security disability attorney in San Diego and we were running ads for him, and one day at the end of this phone call, it was like, hey who's the best client, and he's like, people that have been turned down for SSDI, and I was like, okay, well, why is that? He's like, I have a 94% success ratio at overturning those verdicts, and I was like, do you think I could say that in an ad, and he's like, absolutely. So, the ad copy was, "Denied SSDI question mark, 94% overturn rate. That ad like just killed it and I don't feel like AI is at the point where it's figuring that stuff out. You know, like it's not thinking that deep.

We've done a couple of programs, not around bid management necessarily, but around copy and creative. So, we used a copy one that takes Facebook ads, changes the headline, [00:16:00] and it's a really cool presentation. It makes intuitive sense. We have mapped these keywords to specific emotions and based on the account performance, your audience responds to the emotions of fear and blah, blah, blah, and then it like outputs these headlines and it like kind of, sort of works, like kind of, sort of, a little bit.

Then we used another one that was supposed to give a predictive score on Facebook creative, but the pitch was, you're going to have to throw this in the market, put budget behind it, find out if it's a winner or loser, and instead, you can upload to the API and the API will give a score. Like, a red light, green light, yellow light, and it'll tell you if it's going to be a winner. Therefore you're like only putting winners or like high confidence winners in, and that thing never ever picked a winner, ever. Like ever, and all kinds of cool ways to slice and dice the data, right?

Like, this ad has two people versus one person. You know, the perfect recipe is one [00:17:00] person, plus this headline, plus this color palette, plus this. It never worked. I think part of it is there's just too much behind the scenes. Even in a Google search. Even if an algorithm can look at something like you got demo, age, time of day, device, internet speed. There's just all these other variables, and so that I don't even think the AI can see all that to make a prediction.

Erik Martinez: There's too many permutations.

Paul Rakovich: I think so. Yeah.

Erik Martinez: And not enough consistency in what people are doing. Isn't that, the promise of AI? I mean, really that's way, way back early in my career. My first job out of college was working on an AI project. This is exactly the type of stuff that they wanted to do. You know, this was, I hate to say it, mid-nineties, and the technology was completely new. Let's be honest, our marketing systems, including our digital tracking systems, are less than perfect [00:18:00] and often spit out a lot of noise and garbage.

So, you're expecting the machine to take it in the context and create all these little tiny nuances and they can handle more permutations than us. You know, it kind of comes back to that micro audience, too broad of an audience thing. You gotta be somewhere in the middle type thing.

Paul Rakovich: Yep.

Erik Martinez: Right?

Paul Rakovich: Yeah. Is it actionable, right? If it's like, Instagram is a really great placement for you with this creative. So, you pull that creative out, you run it just Instagram only placement and it performs like crap because you're not running all placements. I see you and I agree. However, like to make that actionable, there's often a disconnect there in my experience.

Tim Curtis: You know, you think about AI and AI has been the benefit of some great marketing. AI is positioned out there. People don't want to question AI. We've got all sorts of examples, both, of course, with clients, but also there's plenty of stories about how AI has some deep flaws. AI is [00:19:00] only as good as, number one, the information that we put into it, and as readable as the context is. Those are things that a great CMO or a great VP of Ecommerce or whatever your title, needs to approach and needs to have the confidence to approach AI with some skepticism. Is turning over all of my bids to Google AI and allowing them to run with it, is that really in my best interest?

Paul Rakovich: The bidding piece in Google is getting better because Google has all the info. If it's like a third-party platform trying to layer on top of it, I think it's tough, but even then you still got to test. The best place I found is like for cool copy ideas. So, there's a couple of AI copywriting tools out there. It's like, that's kind of a cool angle. That's kind of a cool headline that we're going to mix in with this thing, but we still got to tell the thing, we still know the most important client is someone who's been turned down for disability insurance. Therefore, how can we attract that person? I want to speak to them. You know, but it might have like a cool call to action or something like, okay, that's kind of interesting.

Erik Martinez: On this topic [00:20:00] of AI. This is actually a true story this week. Google has been working harder and harder at automating their internal processes. So, all the review, the ads, and all that stuff is running through this crazy automation system. Almost two weeks ago now, we had a client whose shopping feed, you know, we work in a lot of Ecommerce clients and their shopping feed went down. It went in review.

The response from our Google rep has been, hey, it's in review. It could take up to seven days. I'm like, hey, this is having a serious impact on revenue. The shopping program is a big piece of the marketing budget. Day eight, still, no answer. Day nine, still no answer. So, we try to escalate and they actually suggested put in another ticket because you know, the way our systems work and it's, we have no visibility to it. So, we put in another ticket.

We got another response yesterday and the response was yeah, it's still in review. Our review teams are way behind, and so it may extend the [00:21:00] process a couple more days. Okay. This is Google who has automated all their systems to the point that nobody can actually peek in underneath it, unless you have a Ph.D. in engineering, that can't help this one client out who, quite frankly, now we're looking at going this may go on forever.

We're going to find a way to reclaim some of those sales and guess what? I'm not putting them back into the Google bucket. So, AI is a kind of a double-edged sword in my opinion, and so that was the main reason for that question.

So, changing tracks a little bit, Paul. As we start looking at different spend levels. Your company, you know, you got clients that are spending $10,000 a month, a hundred thousand dollars a month, a million dollars a month in Facebook ads. What are the things to consider as you're moving between levels?

Paul Rakovich: That's a really good question. Generally speaking, the smaller budget ones [00:22:00] are tracking an action. It's like a cost per conversion. Generally, in my experience, like they're tracking a conversion. It's a lead. We think the leads are good. Maybe there's a measure on it past that, but usually not. It's like a lead or, it's a smaller campaign where it's like, I spend $10,000 on Google. I make $30,000 in sales. My cost of product is this, therefore like I'm comfortable spending that. So, that's the smaller budgets.

The next level up, I think, those clients are looking at cost per acquisition for the most part. So, we're allowed to get a little bit away from the hyper-focus on cost per lead. Cheaper leads are better. With this one it's leads, booked meetings, showed meetings, sales. The booked meetings, it's a little more leading indicator. The percentage of leads to book meetings as an indicator of quality. So, we're looking at that, but like the sales cycles still a little longer. So, at some point, we're looking at a cost per sale, but we can't tie the data all the way [00:23:00] back to say, desktop cost per sale is a thousand bucks, mobile it's 2000 bucks. We should dedicate more towards the mobile, for example.

The big spenders, in my experience, are looking almost exclusively at lifetime value. There's been a couple of clients where we've kind of built a name in certain industries. Like, oh, you guys handle this account. Like, wow. That's pretty cool. We've talked to relate to people, and when we start to talk about like cost per lead and stuff, they're just like, what? How in the world can you guys spend that much for a lead. Holy cow. That doesn't make any sense.

I would say, on our side, we're looking at lifetime value, ultimate cost per sale, several steps further down the funnel. Those large spenders, those businesses are like impeccably well-run. I mean, the sales process is world-class, the follow-ups, world-class, the ability to test and try stuff. Two of those AI things, the ad copy one was from this client and we probably spent half a million dollars in ad spend, testing this thing out.

The whole [00:24:00] reason he kept doing it is that it beat their email control. The AI software did. So, he was like, look, it's already paid for itself. If we can find another nugget in the next year, like totally cool, but their willingness, and their CMOs specifically, his willingness to like test, try, find stuff out, is so much greater at that larger spend level, and there's such a bigger tolerance for shoot that didn't work.

In fact, we would get calls sometimes and he'd say like, why don't you guys turn that off? You only spent $10,000 on that. That's not nearly enough. We're going to test this. So, their tolerance to test is by far the highest. They are really well-run businesses too.

So, we look pretty good dumping a million dollars a month of prospects in the top, but their systems process structure is freaking awesome. It's just unbeatable. So, their follow-ups are going to crank out more leads. The sales process is going to crank out more leads, and when we talked to these lesser sophisticated folks that were kind of in the same industry and they're like, this doesn't make any sense that you're generating a lead for 50 bucks. This [00:25:00] does not compute kind of thing. Like, there just weren't as good as operating the business, in my opinion.

Erik Martinez: I think you bring up a good point. I mean, even for our Ecommerce retailers, being well-organized, and having good operation from the time the order comes in the door to the time it goes through all the processes, credit card processing all the way out the back end and into your home. Those companies that execute that well do better with their marketing spends than the companies that just kind of like constantly have stuff on backorder or never shows up and orders mysteriously get canceled.

Paul Rakovich: Yeah.

Erik Martinez: So, that makes a ton of sense.

Paul Rakovich: The Ecomm folks that don't keep their shopping feed up, and you look at the shopping ROAS and it kind of sucks, and then you poke into it and it's like, oh, you only have one size available in this product. You're completely sold out of these 10. I just find the organization of the business has an impact. So, like shopping feeds tight, descriptions are well-written, everything's up to [00:26:00] date. That's going to make you or I more successful, no question about it.

It's a lot easier just to say, can you guys get the ROAS a lot higher? Can you guys pull a magic rabbit out of the hat? So yeah, those and those will do okay. We've done like some clothing boutiques and they're okay, but they just never like get to that next notch, from what I've seen. They almost have to be reinvented or the leadership has to change, or they need to level up the business in some form or fashion.

Tim Curtis: Kind of pivoting a little bit. We've sort of been through the first two rounds, at least. I think we're on what? 2.1 or 2.2 of Apple's intelligent tracking prevention. We've got Google. Google has not made the third-party cookie change completely yet. So, that's coming, but understanding, and maybe this is a sort of a bit of a two-part I guess you could consider.

Number one, what would you say are the three most significant changes that Google has made in the last year that, you know, the listeners should really care about and pay attention to, and then we'll later we'll get to the future? [00:27:00]

Paul Rakovich: Yeah. I think this performance max focus when you run those performance campaigns, they just like completely take over the budget. Completely, and those scare me. I'm sure it will get solved. There was a point where if you advertised on Bing's audience network, we were running a campaign and all these leads came in that were just complete garbage, but what was interesting is whoever was filling them out or whatever automated thing was filling them out was actually grabbing an IP address from the appropriate city.

So, it would grab an IP address, click through the site, bot would fill out the form, but they always did it like a Yahoo email address. I was like, oh, you dummies. Step your frog game up, and so on these other campaigns where Google is like completely taking over, we see a lot of what feels reminiscent to that. Like, the forms are getting filled out, but their garbage. The Geo's garbage. The client can't contact them, and so I feel like you really gotta be careful in looking at a step or two deeper in the quality when you let Google take the reins.[00:28:00] That's the one that I'm kind of most scared of.

The coolest one, or where we're seeing accounts do a lot better, is max conversion setting. So, let max conversions go. Let Google figure it out. Set, what used to be a TCPA bid on it. Those are generally performing better like pretty much across the board. The caveat to that is like, now you really don't have a whole lot of leverage to control. So, you're running a display campaign, discovery campaign with max conversions.

You can pretty much manage the daily bid in the creative and that's like, probably all that literally all you can do. There's no other levers to pull, right? It's like, let's redo the creative, let's raise the budget, lower the budget, and we can maybe like, try and think, hey, if I raise the budget, they're going to be less efficient with it. If I lower the budget, it'll be more efficient, but that's kind of a guessing game, honestly. So, that's my second one in terms of what I'm seeing at the moment.

Third one. Biggest change. Biggest change. I don't know if it's a problem or not, but like the close variant piece, like exact [00:29:00] match, not being exact match anymore. I'm kind of 50/50 on it. Part of me is, in cases where we've even been able to slice and dice the data like a bajillion ways, is there a difference between a click for personal injury attorney and personal injury lawyer, like three steps down? You know, like, I don't know that you can really get there.

So, if Google mixes them up, probably not the end of the world, but that one just annoys me cause I, you know, I always had my exact match campaigns, phrase match, broad. Exact would get ten bucks, phrase would get five, broad would get two-fifty in terms of a bid. I just feel like that's like been taken away from me. I want it back.

Erik Martinez: Our search team has felt the exact same thing. I would say on the flip side of that coin, I think Google has gotten better at contextualizing the searches for those in determining intent. The results are better. There's a handful of searches I can still find where Google is confused and you can just tell by looking at [00:30:00] your results. It's just all over the place, but it's not as prevalent. So, maybe the need for those things isn't as necessary anymore because they've gotten better on determining intent.

Paul Rakovich: Times my might be a change in on that one, for sure. I just kinda miss it.

Tim Curtis: You know, the truth is flexibility. A lot of these things require flexibility on a personal level because you know, we have run things the same way for quite a while and you're used to that organization and when that's taken away, you know, your cheese has been moved. That's a little hard. That's Google, right? Much more touted, I guess you could say, or much more in the news would be Facebook, Instagram.

Paul Rakovich: Sure.

Tim Curtis: And you know, we've talked about this before with previous guests, but I believe it was late February, The Wall Street Journal released an article. It was really a good article that really conceptualized what the challenges were, and it was Inside Facebook's $10 Billion Breakup with Advertisers, and it was going into some of the detail about the loss [00:31:00] of that transactional information, a lot of that privacy that has been turned off through the iOS channels, and the Android side as well. There've been other changes, but let's go back and identify, just like with Google three, what are the three biggest for you or your clients?

Paul Rakovich: Phew, that's a good one. Biggest change. We're revisiting a lot of stuff that didn't technically work in the past. So, specifically around bid strategy. There are several accounts for like lowest cost just never worked and now we're revisiting it, and what we're finding is that has to be a piece of the puzzle to have some scale. Running an account solely on a cost cap probably isn't going to work, and oddly, we're finding when you have both that the accounts are performing better.

I mean, in the accounts we're running, we can't just throw everything into a cost cap. It just won't spend. We can't throw everything into a lowest cost. It kind of hits this point and it just doesn't work. It just go sideways or it's awesome for like two weeks and all of the sudden, it's like, oh no, this is not working anymore.[00:32:00] Same thing with Google, like there's only so many levers you can pull on a campaign like that. It's creative copy. Copy, I don't really pay as much attention to you, but creative for sure. Okay. That can be a difference, but daily budget that's about it. You either raise it or lower it. Hope Facebook's more efficient from it. So, revisiting a lot of stuff.

I'm still pressing clients and potential clients on creative. I still think outside of the API. There's still nothing better that you can do than work on creative, test creative, find a better creative. If you were to like tie my hands behind the back. I could only do one thing, I would spend every penny on creative still. No matter what.

Third thing. Third thing. I think like we were talking about earlier, it's this, like, can't go too broad. Can't go too small. On this campaign, that's even a little shaky. We're targeting a pretty broad audience. It's a 5% lookalike campaign to get a decent size audience there, and I was talking to the CMO, this big spender, who we just talked shop a lot. [00:33:00] Our CPMs are a hundred bucks, super hot, and our cost per link clicks 10 bucks, which is not viable long-term. His take was 1% is going to have a better click-through rate. I bet you'll see a lower CPM, lower link click if you go smaller. So, we're just testing.

You know, we're just trying to figure out like is one right, and then that only spends 500 bucks a day. Is five right? I know it's specific. There's just a lot of freaking massaging. The expectation is you really got a massage a heck of a lot more than you used to, and then all of a sudden like it pops, right? So, it's one to 3%, and if you go 21 to 65, the account does great. If you go 24 to 55, the campaign sucks and doesn't spend. The older and younger leads aren't the best, but you kinda gotta have them for the overall to perform well and so I feel like it's a lot of massaging.

Erik Martinez: A lot of massaging. So, can we go back to [00:34:00] creative for a second?

Paul Rakovich: Sure.

Erik Martinez: When we were talking pre-show you are saying, hey, I've got two people doing creative all the time. One of the things I'm seeing, just kind of across the digital landscape right now is value being placed on originality because we have been a copy culture for the last four, five, six, seven years, right? Lots of posts and reposts of other people's content, and I'm starting to see, in Google in particular, on the organic side, but I'm also starting to see this a little bit on Facebook. It seems like originality of content is becoming more and more valuable because we're in a sea of copyright now.

Paul Rakovich: Sure.

Erik Martinez: So, as your team works on creative, what's the mix, not that there's a million things you can do on Facebook in terms of creative, at least on static ads. How much do you put into those static ads versus, you know, like developing motion [00:35:00] ads or video ads? Where do you guys sit on that?

Paul Rakovich: Yeah. The big spenders, at least 50% or more is going to video being like UGC video ads or motion graphic, and then static still has a place, but it's definitely not. Even that's account specific. I looked at an account the other day and it's only dynamic ads, no other ads, and the counts killing it, and I looked at a golf account for a client. We do his Google work and he's running only a carousel ad. I look at both of those and I'm like, carousels usually never work that great. If you can get them to work. Great, but like we've never had carousels for us work super great. So, to see an account just running carousels is like flabbergasting, but for us, video motion graphics is a big piece of it. Doesn't have to be anything fancy.

We still do UGC, and it's kind of a battle of there's a couple of phrases in a couple of different industries where like, we just cannot get away from this thing works. [00:36:00] We know it works. Man, we'd like to say something else cause we're so tired of this, but like, if we don't have it, it literally doesn't work. We really vacillate between, you got to put this in there or it's not going to work, and like, let's find something that works better. So, it's a balance.

When we first started doing UGC, one of our employees went down to the parking garage and did a video in their car. We tried to beat that control for like six months until it dawned on me like, well, hey, why don't we just send someone else to the garage? Like, give them a couple of hundred bucks. Like maybe, you know, she's not the only person in the parking lot that will work. That's where we started UGC. People rip off our ads in that vertical all the time.

One of my early clients was this kid who was a network marketer, and it was the typical story of like, lives in his mom's basement and makes million dollars a year. It was like, so, so ridiculous, and every time we like said how much more money he makes the ads did better. People would rip off the ads I wrote back then in like Yahoo and what was always [00:37:00] funny to me was I would see an ad that I knew, like tested really bad and I'd see people rip it off and I'd be like, ha, ha, ha. You don't know that ad sucks yet. Go ahead. Copy me.

So yeah, it's like a balance between other people are doing it. This schtick works. Let's keep doing this schtick, and like, we got to reinvent the wheel. I'd say the reinventions usually come by like some accident or something weird What's the saying? Like, good ideas from work never come when you're at work.

Tim Curtis: So, the UGC content. I mean, I guess the question is really what's UGC content? If it's employees doing it, is that UGC content or is that juicing?

Paul Rakovich: Yeah. They're an actor at that point, and I think the use of these other actors are actors and that's what it is, blanketing it as UGC. I would agree, isn't necessarily a user. So, there's actors, and then we do have clients that will take their clients, but that style of ad just still kills it. Even the TikTok reps were telling us the other day, like, don't make the ads nice. Really, [00:38:00] don't make them nice. If you can make them a little shaky, that's even better, a little grainy.

Erik Martinez: Because they feel more authentic.

Tim Curtis: They feel more authentic. Back in the day, you know, I had some huge, huge, huge global brands and we very heavily leveraged Olapic for a lot of the UGC, and you know, you're talking about, at that point in time, sort of the evolution of really the creative and how those ads come across, because authenticity is that grainy texture, right? It's something that's shot spontaneously. The stage is set to come across as authentic. That seems to be kind of the direction that it's going. The downside of that is it's harder to be novelty.

Paul Rakovich: You got to do something silly. Like, one of our clients wanted to do, he saw the TikTok that I'm sure it was fake. This woman like drives her Tesla to the gas station. It's like, oh my gosh, how does this work? Can we do something really stupid like that? You just got to keep raising the bar of stupidity.

There's a lot of cool stories about influencers that I think are authentically into something and authentically [00:39:00] recommending a brand and they're like selling out. Better than any advertising could ever possibly do.

Tim Curtis: Yeah. Some of those micro-influencers that dabble in that UGC-type content, it's amazing the product that can move. You know, you think of micro-influencer, their reach is limited, but you wouldn't know that by the amount of product that they're able to move through and it's well done. I've seen those guys begin to master, not only social, but now they're mastering these Amazon live feeds. I don't know, it's like a QVC marries Amazon and has a baby kind of thing.

Paul Rakovich: Yeah. Right.

Erik Martinez: That's awesome. Circling back to a topic we were talking about. You mentioned several times during the course of our conversation about lifetime value and how to measure it. You know this is a topic that Tim and I talk about a lot. We actually grew up in the direct marketing industry and lifetime value is a huge thing. That's how you acquired customers and that's how you knew what to spend. Google has put in Google Analytics a little lifetime value calculator, right, but it doesn't have enough data to really put a whole [00:40:00] picture. So, when your clients are looking at lifetime value, what are they looking at? What are the inputs into it and how did they set it up so that they know they're getting a reasonable approximation of reality?

Paul Rakovich: Yep. So, they either literally have all of the data. Like literally have it. So, this client we do, there's 1, 2, 3. There's 4 steps in the sales process. The fourth step is the biggest step in the sales process, and they can literally go back over years of data and say total leads in divided by total step four. Lifetime value is $2,200. and they can do it at specific stages.

So, actually, it's not 2200 of an initial lead, it's 2200 of the third step, and so what they would measure us for is they would say, I know it's a little hard being cryptic in terms of what it is, but they would say once someone reaches the [00:41:00] third step, we'll spend 700 to a thousand dollars to make 2850. It was like literally the number. They knew it and so they knew going from step three to step four, did the math, they did it.

Erik Martinez: But the data is being captured inside a CRM, right?

Paul Rakovich: Inside Fusion Soft for like 10 years and so crazy ability to slice and dice, but at the end of the day, they would just make that call. To the point of being a more sophisticated advertiser, they just had such a tolerance for like these last five campaigns, they didn't do that good, but like, hey, sometimes stuff just doesn't do good, and there wasn't this pressure to like, you've got to figure out what happened in Chicago and do a deep dive analysis. Maybe it was just a bad couple of days.

So, they either have the data stacked, ready to rock, or they're at the point where they're willing to make some approximation of it that I think is grounded in some reality, and I've seen it work and [00:42:00] I've seen it not work. The one I've seen work is around my in-laws' business. We made an approximation around how many people will come back and charter again, but it's incredibly conservative. So, it's a $ 5,000-a-day charter. We made an approximation of when people will come back and they're cool with it.

There's not a big loss leader on the front end, but they're using it more for like, hey, I can press a little harder and spend a little more budget. The cost per lead is going to go up, but you're going to get more deals. Initially, it's not going to be quite as profitable, but like, let's think long term here cause they're thinking years out.

And then I've seen it not work where we did some work for a company that had a ton of funding. The additional marketing director said lifetime value is, let's say like a hundred bucks, right? Just acquire stuff for a hundred bucks, and the spend was, it was like hot and heavy. You know, like, woo, let's go crazy. I remember looking at the campaign on day one of the items in our shopping feed was the S'well water bottles right when they started to come out and they were really cool, and it [00:43:00] was costing us like $85 to sell $30 for a water bottle.

I remember thinking to myself, like this train is going to come to a screeching halt at some point, Could these people potentially be coming back and buying other stuff from the store? Like, it just felt like maybe high level when you divide one number by two, you get this a hundred dollar number, but like, it just didn't seem like reality and sure enough, they ended up running out of money. So, I think there's some nuance to it. At least with that client where there's a bunch of different skews and a bunch of different products. It was yeah, we can be aggressive, but we can't spend $85.

Erik Martinez: You still have to have common sense, right? Everything we've talked about today in digital advertising, we had an earlier conversation on a different topic, but it really still comes down to common sense, right? You gotta make sound decisions with the information you have, but you've gotta remember that information comes from tools and tools are less than perfect, and so you need to bring all that information [00:44:00] together, create a story, but still like, does it pass the test of reasonableness when you just go, hmm. Does that make sense? Should I spend a hundred dollars to get a $30 water bottle sold? Never, ever, ever, ever, ever.

Paul Rakovich: Yeah.

Tim Curtis: Kind of pivoting a little bit towards, you know, future and paid media right now, paid digital advertising, we've had one shoe drop with Apple. We have Google and their privacy and third-party age out happening here before too terribly long. You've got Facebook in a bit of disarray at some of these privacy changes. So, what's your prediction for where this is going as the wild west sort of comes under martial law?

Paul Rakovich: I think Facebook will figure it out. I'll be a little contrarian. I think there's so much money behind it and there's so much information that I think they'll figure it out. I think Apple will do something. I question that the moves are entirely like, we're just protecting your [00:45:00] customers. I don't know that it's entirely altruistic, but I think Facebook will figure it out. Hoping Apple comes up with something interesting that does balance privacy, but also information.

I don't think people want to see like crappy ads. I don't want to see ads for Medicare and stuff. So, people say they don't want it, but I don't want to see ads for incontinence or, you know, catheter replacement.

Tim Curtis: Some of its preference, obviously, you know, there's things you don't want to get, but when you think about the advertiser does not want to be paying to show you an app. You know, that's the greatest tragedy of it all, is that someone somewhere is paying money to show the ad to the wrong person.

Paul Rakovich: When my kids are on YouTube and I watch them like click-through something. I like cringe when they click on the ad because I'm like, oh, this person doesn't know that it's like a 10-year-old kid that fat-fingered this, but how many times is that happening? We were having a conversation the other day about this display creative. It's in a discovery campaign. Our hands are kinda tied, and I was telling the team, you know, we interviewed a creative director a long time ago and he said like, if I threw an ad down on the table and you [00:46:00] walk by it and you glance over at it, do you get what it is? I'm like, that was my direction to the team still and that probably will continue to be a direction for a long time.

It's a golf client. It's a beautiful picture of a golf course, but like, no one's reading the ad. People are barely looking at this. And so is it someone selling a golf course, selling a vacation? Like, it could be a hundred things, and so my challenge to the team was think harder about someone walking by this ad on a table and knowing instantly what it is. Maybe it's as simple as like a text-based image ad that asks a really good question. So, I still feel like that's not going anywhere for a while.

Erik Martinez: Well, Paul, thank you so much for your time today.

Paul Rakovich: Cool guys. Thank you so much.

If anybody wanted to reach out to you, what's the best way?

Sure. Clients@clicksandclients.com. When I first started, it was just me and I thought if I put clients, it would seem bigger. So, it's still my email.

Erik Martinez: Yeah. We've all done [00:47:00] that. Thanks again for coming on the show. We really appreciate it. I'm Erik Martinez from Blue Tangerine

Tim Curtis: and I'm Tim Curtis from Cohere One.