Using PPC Automation to Become More Profitable Than Ever

Episode Description

Leads and conversions are nice, but PPC experts are here for one thing: profit.

How did the smallest team in a media company like Star Tribune manage to become its most profitable division? How did Solutions 8 help hundreds of client teams grow and scale? They both figured out how to leverage automation while still maintaining a personal touch.

Episode Takeaways

Leveraging PPC Automation for Profitability:

• Efficiency through automation: Automation tools are crucial for managing large volumes of accounts efficiently, especially in traditional agencies transitioning to digital.
• Scaling campaigns: Automation facilitates scaling by handling repetitive tasks, allowing teams to manage more accounts with fewer resources.
• Balancing automation and oversight: While automation can enhance productivity, careful monitoring is necessary to ensure campaign quality and relevance.

Strategies for Effective PPC Campaign Management:

• Segmentation and targeting: Differentiating campaigns based on product margins and target demographics can optimize return on ad spend (ROAS).
• Use of smart bidding: Smart bidding strategies like Maximize Clicks and Maximize Conversions help in optimizing campaigns based on the available data and desired outcomes.
• Cross-pollination awareness: Ensuring that campaigns are not competing against each other in the same geographical areas prevents self-cannibalization of clicks and budget.

The Role of Data and Continuous Learning in PPC:

• Data-driven decisions: Utilizing data effectively, such as through the Lin-Rodnitzky Ratio, can guide the decision-making process for bidding and budget allocation.
• Continuous adaptation and testing: Regular testing and adaptation to new tools and features released by platforms like Google are crucial for staying competitive.
• Importance of contextual and historical data: Understanding the historical performance of similar campaigns and industry benchmarks can guide strategy and expectations.

Episode Transcript

Frederick Vallaeys: Hello and welcome to PPC Town Hall. My name is Fred Vallaeys. I’m your host for the show today. I’m also the co founder and CEO at Optmyzr. So today we have two completely new panelists and I thought they were fascinating. So one of the panelists reached out and said he wanted to come on the show and tell a little bit about his story.

And I thought his story was really interesting because here’s a guy and you get to meet him in a second, but he works for what many would consider a traditional Agency sort of like old school media and he told me that he was able to really turn the company around and turn part of his business into the most successful it’s ever been, even though we have the pandemic, even though there’s been a lot of social unrest, and I thought that was an amazing story to tell.

And I wanted to have everyone here. How he’s done it. And hopefully people can learn from that and apply some of those learnings to your own agencies and businesses. And then the other panelist is someone I got to talk to on his own podcast. And I’ll tell you more about that in just a second. The super high energy guy.

Really cool person to talk to has got lots of opinions on everything to do with PPC. So I thought, let’s bring these two people together. Let’s have a great hour of just talking about everything. That’s the latest news in PPC, different opinions and learn from each other. So that’s our PPC town hall for today.

So let’s get started.

All right. Using PPC automation to be the most profitable ever. So you guys, you think that’s a good topic? So my guests, Ken Chang and Qasim Aslam, welcome to the show. Thanks.

Kasim Aslam: Yeah. Thanks for having us, Fred. Appreciate you.

Frederick Vallaeys: Yeah. So so I’ll go in the order that I sort of teased you in the pre roll here.

But Ken, you’re the man the traditional media company. Tell us a little bit about yourself and what you do there and what department you work in.

Ken Chang: Ken Chang, I work over at Star Tribune. I’m the senior paid media strategist there. I work in a very small team. We were three team members and we’re just hiring our fourth team member, onboarding them.

Basically, I believe next week and we manage around 300 Google ads accounts and a couple thousand display accounts as you’d imagine. We, we work with every advertiser there is being being a media company and our challenges were basically. Combining automation to, to scale. Basically when you’re looking at a traditional company they might have like, or traditional agency, they might have like 30 or 40 accounts and they might have like, like an equal number of, of people assigned to each account, so they might have like one or two,

Frederick Vallaeys: one, two

Ken Chang: strategies.

Yeah. Yeah. So, so we’re, we’re basically looking at automating everything and then, and then being able to view the accounts daily so that we could, we could pace things for our clients.

Frederick Vallaeys: Okay. So small, but super powerful within the company. That’s awesome. And I can’t wait to hear some of these details that you’re let’s see, you’re eager to get into them.

Hey, but tell us Star Tribune, where’s that based?

Ken Chang: Sure. It’s based in Minneapolis, Minnesota. Basically, we’ve been here for a couple hundred years, so, so I guess over, over 153 years.

Frederick Vallaeys: But you’re a little bit younger than that, right?

Ken Chang: Yeah. Yeah, I’m a, I’m a junior, junior member of the team. Yeah.

Frederick Vallaeys: Right, but on the rise.

Awesome. Qasim let’s bring you in. So you’re the second person I teased up. High energy, you run an agency, recently been promoted, I heard, to some some cool thing that you’re doing. You want to tell us about that?

Kasim Aslam: Yeah, there’s a podcast out in the the ether of the interwebs, Perpetual Traffic.

It’s the main host is a guy named Ralph Burns. Ralph runs Tier 11. Some people will have heard of that. I think they’re probably the world’s foremost authority on Facebook ads. Really high end Facebook agency. He’s been doing that for about five years, and I was just made co host. Virtually yesterday.

It’s only been a couple of episodes, but those drop every Tuesday. If I don’t annoy the daylights out of you and you’re listening to this and you want to hear more of me, you can go check us out just Google perpetual traffic, wherever podcasts, wherever you go for your podcast, you’re going to find.

Frederick Vallaeys: Well, that’s amazing. Since we’re best friends now. I’m sure I get to come on at some point too. I was,

Kasim Aslam: I was going to try to con you into doing it, Frederick. So you just did it to yourself. Yeah. You have no one but yourself to blame. Okay.

Frederick Vallaeys: So we have a couple of topics lined up you know, that, that growth of Ken’s agency within the company.

Definitely want to talk about that. I also want to make sure we have some time to talk about some of the things we’ve seen lately. In in google so the optimized targeting I thought was a really interesting thing that they’ve recently come out with I wanted to talk a little bit about the retirement of third party cookies and what’s going on with flock but why don’t we start going back to ken?

By the way, this is a live show. We’re live streaming So anyone who’s got questions pop them right there into the youtube chat. We show them on the screen like you see right here Can just say hello to where you’re calling in from. But that is a great way to put your questions in front of the panelists.

And of course, if you want to have a little side discussion and want to argue about different points of view, go for it. That’s what we’re here for. And if you like these, make sure to subscribe as well. All right. So Ken, talk to us a little bit more about the agency. And so you guys went from three people to four people.

So it’s still relatively small. You’re saying you’re managing thousands of display accounts. I’m kind of hearing here that you’re probably pro automation.

Ken Chang: Yes. Yes. The bulk of it is, is hiring like a bunch of bunch of superstars that are, that are well versed in, in, in PPC. So, so it’s not, it’s not just on me.

It’s, it’s also like, like we hired like a couple other amazing people and then we’ve worked with a couple of other amazing teams too. And then, and then with, we also work with a very amazing sales, sales team too, that, that brings in all these accounts. And, and we’re also good.

Frederick Vallaeys: So do you feel like the sales team sometimes over promises and then you guys have to just like figure out how to actually.

Do all of that work?

Ken Chang: No, no, the I believe the only friction comes from we’re actually the, the white label agency for, for for other ad agencies. So, so during the pandemic, when, when things got difficult for the other agencies they weren’t able to, to pace campaigns or, or give campaigns the amount of love and, and treatment that we could.

So, so we actually took on more business. During the, during the pandemic just because the other agencies weren’t able to, to pace things. So, so like when you’re, when you’re looking at things from an automation standpoint there are things that are time saving, but then basically the structure of the day has to be fairly, fairly structured, you can’t.

Exactly. Dilly dally throughout the day. You have to, you have to just be hyper focused on, on how you spend your time and then place a

Frederick Vallaeys: little bit, if you don’t mind, like who’s your typical clients.

Ken Chang: Sure. So before the pandemic, our typical client were restaurants, bars service companies like plumbers, electricians, roofers.

Things like that after the pandemic, we did a one 80 and, and our clients are now banks, insurance companies, software companies basically cloud services

Frederick Vallaeys: All the sorts of businesses that stayed open during the pandemic that didn’t have to yes And now you you’re based in minnesota, right? So was there any?

You you had the pandemic and then I guess there was the social unrest

Ken Chang: Yeah, yeah the

Frederick Vallaeys: the types of businesses you were working with as well

Ken Chang: Yeah, the basically we had a whole division closed down our city pages division that focused on bars, restaurants and events basically closed down because of the pandemic.

And then it also made you know, you probably, we were basically at the epicenter of all the social unrest and, and rioting, and then also the, the murder of George Floyd and then the Derek Chauvin trial. Our, our, our, our building is actually. Right next door to, to the the Henneman County government center where the, where the trial was held.

So, so basically we’ve, we went through all of that and we stayed open, published a paper you know, like I don’t know if you actually see it, but they, you know, they, they won the Pulitzer for, for for for reporting all that coverage. And then during that same time, we were able to support the journalists by, by pacing ads at that time too.

Frederick Vallaeys: And I’d like to hear from you for a minute as well. So what are your typical clients and tell us a little, you told us about the podcast, right? But you also run an agency. So I think you’re going to have some interesting viewpoints here as well, but who do you guys work with mostly?

Kasim Aslam: I just want to say Frederick, that it’s not fair for you to team me up after Ken says that they want to pull it, sir.

Like I have nothing to put up against everything that you just said. That was so compelling, man. Like what an amazing place for you to be in history. Really appreciated you sharing that story. Thank you. I’m going to bring this all down a notch from a pedigree perspective I’ve got I’ve got a hundred and fifty hundred and sixty some odd clients 80 different industries I have 45 employees So my my employee to client ratio is about what I think Frederick probably makes fun of on a daily basis And maybe y’all can help me with that We run a Google Ads specifically We were something of a me too agency years ago and we try to do all the things and we just found out that Google ads is what we were best at and maybe most importantly, it’s also where the clients that lasted with us the longest were the ones that were successful in Google.

It’s like, I call it the coliseum. You’re taking your client and you’re dropping them into the, you know, the octagon with all of their competitors. And if they come out alive, then they have a viable business model. And so it’s nice because, you know, of course, clients are vetting us, but we’re also vetting them.

Because retention is such an important piece of the agency. So that’s, that’s how we ended up in the Google world. And man, I just love it. I feel like we’re at the tip of the spear for all things marketing. It’s the extreme bottom of the funnel. It’s where people go with intent. So, you know, I think that we, we occupy a pretty important space, all of us.

And I know

Frederick Vallaeys: I love the reference to like Coliseum and Octagon and but it’s like, we’re all fighting our competitors, the other advertisers, yet we’re all here. As friends kind of sharing about how we do it because you go into the Coliseum and then they drop in lions and crocodiles, right? And it’s no longer just your competition.

It’s kind of like the forces above us and that’s Google, right? That’s Google. That’s Microsoft. So they’re dropping these little things on us. And the first goal oftentimes is, well, how do we just. Stay alive as a group should be maybe team up. And I think that’s what we’re doing. Easy town hall. That’s what makes it so interesting with like all this automation being dropped on us, exact match, no longer being exact match.

And so I kind of get the sense that you’re probably on the side of a little bit more. manual, more human. And Ken, I’m guessing that you guys do it with automation, right? So what was that decision point? And Ken, did it transition for you at some point? Because obviously human run PPC became before automation.

So has it always been automation driven or like?

Ken Chang: No, no, it hasn’t. It, it really started when I, when I joined the starch being roughly roughly two years ago. I was I was automating things that at modern climate quite successfully So but modern climate ended up selling their their tv stations and newspapers to scripts for 600 million dollars And they bought that Book a business.

So at the time I was managing 500 accounts at, at modern climate with one other, with one other here. So, so that was, that was a real trial by fire process. And that was when I started using Optmyzr heavily for our accounts. So we would, we would link everything into the Optmyzr and then we would just.

Pace everything. And then we would write a ad hoc script that, that would generate accounts. Like, so we would actually have like a Ruby script that would manipulate Excel files to, to kick out kick out an account for account setup simplicity. And, and basically we use that to scale we, you know, so basically it was just, you know, like running these, these scripts in the Google, Google’s ad API, and then, and then trying to slot things very quickly and very effectively, but we were always.

Frederick Vallaeys: So, sorry. I’m sure people would love to know more about that script because everybody wants to build accounts faster. Was that something you guys developed in house or was that something you found somewhere that others might be able to use?

Ken Chang: That was, that was something we, we wrote in house. We had a, we had another ad agency called Ackman and Dickinson at the time that, that had a staff of programmers.

So we were very lucky at the time to, to have a, have a programming staff on hand that, that could. That could write us a script. So we basically took campaign. So we basically the way it worked was basically you would export all your campaigns out from AdWords editor into a flat file. So it was a flat Excel file.

And then the script would, would mine things from basically it was like another it was basically another kind of AdWords editor that you could select ad groups and then repurpose the ad groups basically.

Frederick Vallaeys: Interesting. And sorry, I’m going to be a little bit all over the place because I’m curious, but you said you just hired another person, right?

What kind of skillset were you hiring for? Because it sounds like you just had that example from the last agency where there was so much value from hiring someone who could do scripting at this point. Are you looking like to build more in house technology, looking for someone more technical, or do you sort of feel like the landscape of automation tools is good enough?

And you just want experts to work with those tools or what?

Ken Chang: No, no, it’s, it’s really, it’s really nonlinear. I would, I wouldn’t say like, like like you hired this kind of major from college or anything like that. Everybody that, that lands in an ad agency role is, is basically has a, has a different type of background.

We basically look for personality first because, because a lot of the work is, is Tedious and, and boring. So it’s like a lot of

Frederick Vallaeys: it. Yeah.

Ken Chang: I mean, I mean, it’s like, it’s like on the worst day, it’s like a data entry job, right? So. So like when you, when you look at it from like, if you’re trying to automate like hundreds of accounts at the same time and you’re staring at six different Excel sheets, it’s, it requires somebody that, that could, that, that could like go like, okay, like that’s, that’s not going to be a big deal for today.

Frederick Vallaeys: Can I ask you what a project management system do you use to coordinate between these six spreadsheets and all the people and hundreds of accounts?

Ken Chang: We use, we use two to two to three different tools. So. So like, like we use a smart sheet. Trello and then TapClicks. So depending on which which team that we’re we’re dealing with because because we still have solid divisions where where we’re at, too.

Frederick Vallaeys: Awesome. What about you guys? Already have 45 people hiring more.

Kasim Aslam: Oh, yeah, we’re I mean, I can’t hire fast enough That’s my biggest rate limiting step specifically in the client management we have a really solid, we split our campaign management up. It’s a trifecta. And I don’t know that every agency does this, but I have found this to be the gold standard.

If you don’t mind me posturing a little bit, the point of contact for the client as a client manager, this person knows strategy, knows how to run a campaign, but doesn’t actually press the buttons. We call them the client advocate internally because they should assume the role of the client when internal conversations are taking place, they connect with the specialist who runs the accounts.

A lot of agencies connect people directly with the specialists, which I don’t. I don’t believe in because I feel like specialists are, they’re more engineers, they’re left brain. They’re actually geared to saying, no, this is why it won’t work. And so they need something of a, you know, right and left brain partnership.

And then sitting on top of that triangle is our, is our strategist. I only have two on staff because they’re, they’re expensive. But they’re the ones that are, you know, overseeing everything and making sure the ship is running properly. We 245 applicants. It takes 245 applicants to produce one specialist.

So we have what I hope, to be the best people in the whole wide world and a really phenomenal system for bringing them in. Specialists, thank goodness, are a little easier for us to find because Google produces them in mass. We actually go and we poach Google’s people and we have ways of doing that too.

It’s the client managers that are hard because you’re, you’re basically trying to hire entrepreneurs and entrepreneurs by nature don’t like to be hired. So I call them intrapreneurs. I love people with side hustles. I think every single one of our client managers has something on the side that they do.

And I think that’s important because it just, it strikes a really solid note of empathy. You know, they really understand our clients and because they’re there, they’re, they’re also trying to, you know, squeeze blood from a turnip in a lot of ways. And they know how to have those conversations.

Frederick Vallaeys: You’re in Arizona, right?

So what do you do? You hit up the bars where the the Google employees go after work. Yeah.

Kasim Aslam: Well, our, our entire, our, our agency is disembodied. So I think we have four or five people in Arizona and all admin basically, like our controllers here, my business partners here, the head of our communication marketing team is here, although she’s up in Flagstaff, you know, three hours away from us and then everybody’s, we’ve got people in Canada, UK, Ukraine, India, the Philippines, all over the States, Hawaii.

I couldn’t run the agency I run if I tried to be in one place. The talent pool isn’t here. Maybe I could, you know, Frederick, if I was in Silicon Valley, I could probably do it there because everybody there is wicked smart, but then I couldn’t afford them. Exactly,

Frederick Vallaeys: you couldn’t afford it. I mean, that’s how we run as well, right?

So we have very few people in the United States, and we definitely benefit from having really smart people in Silicon Valley. Different countries, different locations where we don’t have to pay quite as much.

Kasim Aslam: Yeah.

Frederick Vallaeys: And then that also is how, you know, I, I could still do it, but my tool would become far more expensive. So hopefully it’s a benefit for everyone. Can all of you guys are local?

Ken Chang: Yes. We have, we have one hire that’s, that’s based out of Iowa at the moment, so one, one new hire from Iowa. That’s, that’s moving into Minneapolis.

Frederick Vallaeys: Okay. Hey, so let me tell you this question. So favorite automation to manage your AdWords accounts,

or let me say one you hate the most, because that’s the other part of the question. So favorite and least favorite automations.

Ken Chang: So, so my, my favorite automation would perhaps be, be maximized clicks. Just because, just because the, the, the bidding works when you, when you set like a, like a limited budget it, it, it, you know, basically once you have everything all lined up, it, it works fabulously.

My least favorite is like the auto. Applied ad suggestions because sometimes they’re not, they’re not always relevant. It’s just the they tend to advertise on, on things that, that you’re not working towards.

Frederick Vallaeys: Okay. And great. And then we’ll get to you in a minute, but Yuki is asking about smart bidding.

And so maximize clicks is one of these bidding automations, right? Do you use exclusively maximized click scan or do you use others as well?

Ken Chang: Yeah, maximize maximize conversions is, is a close second favorite. A lot of times we don’t have, a lot of times we’re just spinning up the campaign and we don’t have a lot of data.

So maximize clicks as a way of accelerating. It used to been Google used to have a feature where you’d like accelerate the campaign and then gather a lot of data. In a short period of time maximize clicks and a seasonality adjustment are ways of Of spinning up that additional data to to get like data points very quickly into the campaign but let’s see when you’re when you’re hesitant on adding automation I would suggest like maybe taking a baby step maybe creating a campaign that is automated and then they part that and then and then Have another campaign that is, that is just manual.

And then so maybe, maybe Monday through Wednesdays are days where you have your automated campaign and then Thursday, Fridays are days that you, you punch a manual items in there.

Frederick Vallaeys: Interesting. What about you, Qasim? Do you guys use automated bidding?

Kasim Aslam: Oh, yes, sir. Absolutely. We’re actually very heavy into smart shopping.

I’d say a third of our clients are e commerce based.

Frederick Vallaeys: Okay.

Kasim Aslam: And smart shopping has been one of the highest performing utilities we’ve ever encountered. I’ve got a client They spent 30 day period and made 2, 000, 000. So 15, 000 some odd percent ROAS. Now they’re my single greatest case study. So I like to trump them out, but we have people that produce consistently, you know, like between 400 and 3, 000 percent ROAS consistently, and it’s, it’s been on the back of smart shopping, which, you know, Google’s monitoring who’s.

Purchasing from you and then going out and finding more of those purchasers. So it’s one of the first outbound Google paradigms that I’ve really played with heavily. And it relies entirely on, on smart bidding and, you know, specifically assigning a target ROAS. It does take quite a bit in the way of manual intervention, both in terms of how you run your smart shopping campaign and then also how you do things like feed optimization.

But you know, the, the, the automation is definitely, Critical. We wouldn’t be able to do without it.

Frederick Vallaeys: And what advice do you have for Yuki as far as testing it? And I’ll give you my opinion on that too in a minute.

Kasim Aslam: Yuki, I would need to know what type of campaign you’re running, if you’re running lead generation or if you’re running e com, if you’re running e commerce, you need 90 days.

Google’s literature says that you need 45 days in smart shopping untouched, which is actually really scary to do because when you’re watching a smart shopping campaign, it’s bouncing off the walls. You’re watching a machine learn. We had one client that went two weeks with zero impressions. So don’t expect any performance out of smart shopping for the first 45 days.

And then day 46 is day one in terms of optimization. So then you need another 45 days while you’re making the tweaks that would be necessary in order to improve it. Yuki says lead generation. From a lead gen standpoint, I’d be really careful about applying a ceiling too soon. One of the mistakes that I see early stage clients make is if they’re using smart bidding, they’ll, you know, have a target CPA and all you’re doing is putting a ceiling on top of the campaign before Google can go out and learn.

So you need Google to get really expansive first, which probably means financially inefficient, but then it, It, it starts to kind of narrow down for you. And that 90 day period has been a ubiquitous truth for as long as I’ve been running Google ads. And what’s scary about that is 50 percent of all Google ads campaigns fail within the first 90 days.

So you have to be comfortable with the fact that this might not work. But if and when it works, nothing scales faster. Because we can’t think as quickly as human or as computers can think. So I think it’s absolutely worth the test. I wouldn’t take all hundred grand of your spend and throw it at it. I think the other thing too that I think some agencies maybe get too bullish on is you can start with the With lower spend thresholds.

I actually believe that you have greater intelligence with bottom up spend paradigm versus top down spend paradigm, which Frederick, we can get into if you want to, but I won’t occupy more of the time. How was that for an answer?

Frederick Vallaeys: I think Yuki’s probably going to be pretty happy. My money’s weighing in here as well.

So even though it’s lead gen, start with a TRO as target. And I actually think that’s really good advice, right? So. I think the biggest problem with automation and it not working is that oftentimes we don’t really tell the machine what we really want. We talked about hiring, right? And so we all look for certain qualities in the people that we hire and we assume they’re going to be smart.

And then we tell them, Hey, here’s, Mikasa has a team of you know, the client managers who act on behalf of the clients and their job is to communicate to the people pushing the buttons. What is it? The client truly cares about. Right. And so. I run Optmyzr, like I could go to Google and say, I care about people doing a trial of my software.

I could say that that’s a conversion, but that’s not really a conversion. What I really want is people who become customers who, who pay for it. And that’s the disconnect, right? So if you’ve got lead gen and your leads are this intermediate step, and you’re not informing the machine about what’s happening later on.

Which of these leads actually turn into good customers, then you could be heading down completely the wrong avenue. You could be hitting your TCA, your CPA goals. You could be increasing the number of leads, but they could be lower quality leads because Here’s the thing, right? But, but I often talk about like maximize clicks as a strategy.

Like, I’m actually not a huge fan of it because how do you get the most clicks for a limited amount of money while you get the cheapest clicks? Why is something cheap? It’s usually cheap in a competitive market because. Others didn’t want to buy it. Okay. So then what does that tell you about those types of clicks?

Like maybe I should be willing to pay a little bit more for each click. Those clicks that everybody else wants to other people doing conversion tracking, because those are the clicks that actually lead to something. And I think these are the disconnects that we have to think about. And so really before you even jump into an experiment, whether you use Ken’s methodology of day partying it or customs.

Methodology of giving it a long runway. I think you really need to think about the prerequisites Like have you informed the campaigns correctly? Are you running on last click attribution? I mean i’ve beaten that horse to death on this show and many other places, but please don’t Please

Kasim Aslam: dear god do not

Frederick Vallaeys: Because like I said, there’s so many things that come into it.

Yuki and, you know, everyone watching here, we’re obviously watching this show because we care about PPC. So these are the things we know they take time, unfortunately, and that’s kind of where we have to then communicate to the client. Listen, it’s yes, it sounds to you like it’s an automation, and it sounds to you like we’re going to save a tremendous amount of time.

But we do need runway. We do need a little bit of budget, like Asim was saying, for Some mistakes to be made, but we have really good automations, whether that’s an Optmyzr tool or something else to instantly start monitoring. And like, if there’s some crazy close variant that our ad is now being shown for, or some bids are completely wacko. Let’s, let’s reel it in. Right. We, it’s not saying that we let the machine run for 90 days without doing anything, we still have these negative controls to, to bring it back to where it needs to be.

Kasim Aslam: Can I comment on something that you said, Frederick?

Frederick Vallaeys: Yeah, and then we’ll go to Amit’s point.

Kasim Aslam: Maximize clicks is a really good example because more often than not, I agree with you, but there’s like little, little microcosms, little pockets.

For instance, we have a client that does emergency septic service, and with emergency septic service, all leads are equal. Like if somebody goes emergency septic service now, there’s very little delineation in terms of, you know, like if you get someone who searches for that, they’re going, here’s a 33 percent conversion rate.

It’s insane. All right. And so every now and again, you’ll find use cases for those, you know, otherwise they’re, they’re the lepers of the, the PPC dashboard, but you’ll find one that sort of, this is the only thing that works for this particular implementation.

Frederick Vallaeys: Yeah, absolutely. And there’s never an absolute, there’s always tons of caveats.

And then the other thing is maximize the maximize clicks strategy, just absorbed target CPA strategy. So even though I’m not a huge fan, I now have to use it. Because that’s my way to get to the old school TCPA bit strategy. All right, let’s so meet saying start with a maximize conversions for a fresh campaign. I think I’ll be saying so start with like maximizing clicks and that may be in terms of like build up more data.

Kasim Aslam: Oh, you know who this is? This is Amit Khabra. She’s famous. She’s a, I’m not kidding. This is the PPC girl. She’s out of Canada. I follow her on Instagram. She was just on my show. This is amazing.

This is really cool. Frederick, you post so

Frederick Vallaeys: much.

Kasim Aslam: Yeah, you gotta have Amit on.

Frederick Vallaeys: That’s better.

Kasim Aslam: She’s brilliant. Sorry, I didn’t mean to interrupt you. I just recognized her name. That’s super cool.

Frederick Vallaeys: You know, this is why people make you co host of things.

Kasim Aslam: I just get excited about everything. Yeah. She’s amazing, man.

You really do. I mean, I’m sorry. Oh, I’ll stop.

Frederick Vallaeys: Okay. Well, why don’t you comment on what she’s saying? Because I think. Okay, as you take a swig right there, and I promised you to take him off camera if they were like stuff like drinking. So go ahead and and take that drink right there and I’ll bring it back in three, two, one, and he’s back.

Did you do

Kasim Aslam: it? Yeah, I’m good. I agree with me. In terms of starting, she goes, I always found going max clicks first and then moving to max conversions when we have conversions. I don’t know that I start with max clicks at all, but I don’t like starting with maximized conversions. What I like to do is start with something that, that provides a level of stasis so that Google can figure out.

Here’s the analogy I’ll give you and you guys can challenge me on it. If you give a computer 10, 000 and you say go to the grocery store and get me all the bananas they have. It’s going to come back and say, great, here’s 50 bananas. And I spent 50 bananas. Instead you give a computer send it to the grocery store and say, buy me all the bananas you can get and come back and tell me how much it costs.

And then it’s going to come back and say, I got bananas at 10 cents a banana. And then you can say, awesome. See all the grocery stores in the city. Go get me bananas at 10 cents. So if you can get Google to identify a conversion value before you For you ask for as many conversions as you can, I feel like, and this could be conspiracy theory esque, but I feel like you’ve basically backed Google into now having to match your performance metric as opposed to saying, here’s all my money, please bring me what you can.

Frederick Vallaeys: Great point. Ken, anything to add?

Ken Chang: Sure. I, I guess I guess when I’m, when I’m looking at Amit’s question I would, I would start with a, the only reason why I might start with a max conversion strategy is if, if the industry is, is something that, that has been fairly heavily advertised, like, so for something like automotive dealerships, where, where we know where Google knows, like, the average cost per conversion for a, for a car dealership.

For example, I might, I might just start off with max click or max conversions. If I have that conversion attribution, or if it’s like a, like a dentist, like, like where Google sees like a lot of dental campaigns, then I might, then I might just go ahead and kick it off with max conversions because because Google has, has that data set already.

And there, and we also know like which keywords would pay sin. And convert because of our large data set. So if I have two large data sets, then, then I would just go ahead and just start it off with max conversions. If it’s something more artisan something that’s, that’s more out of the box that needs a little more awareness, then, then I might add in more of the.

Top of the funnel, middle of the funnel keywords. And then I might, I might just start off with max clicks to get that data pool going, then switch to max conversions.

Frederick Vallaeys: Yeah, very interesting. So you’re talking about different attribution models, like maybe use position based, which values the first interaction quite highly for certain industries.

Yep. And then I love this whole point too about big data sets, right? So it’s always, it’s mind blowing to me that Google can in fact now run a Bye. Bye. Smart shopping campaign, which is ROAS driven without seeing a single conversion from your account. And it’s like, wait, but then how do they predict conversions if they’ve never seen one?

And that’s because they have so much data, right? So they can see signals across different advertisers across the Google ads ecosystem and start to make some assumptions. And as they get more data about your account, then they can start to hone in on, okay, it’s not just industry level data, but it’s your data that’s steering the ship at that point.

That’s on the Google side. But Ken, you also made the point that you obviously work with a lot of restaurants, a lot of dentists, a lot of roofing companies. Kassim, you were talking about a plumber, right? I’m assuming you probably have a couple of plumbers in the book of business. How do you learn from these, these people in the same industry and what do you do with that data?

And I think it’s amazing what you should do, obviously, is you run the experiments here, you’ve got a new client there, apply the same learnings, but what, what do you guys do?

Ken Chang: Sure. When I’m running two, two campaigns in a, in a similar market, I need to make sure that they’re not running in the same geos because otherwise I am bidding against myself or our agencies.

Like we’re we have two campaigns and then we would just literally bid against Ourselves on on the campaigns and different accounts for different clients. So So we need to just right off the bat just first look at like we’re not stepping on each other’s toes Then if we see like something that’s working for one client We try to apply it to the to the other client so that everyone’s successful and then if we see things trends that are they’re affecting one account or another we could be proactive about that So so like ad disapprovals or or policy disapprovals we tend to we tend to look at things that that aren’t performing well also if we see like expensive keywords in one account We already know which which keywords to cut out of of the the other account So we try to leverage the the learnings.

It’s not exactly, sometimes it’s not one for one, depending on geography, but, but a lot of cases it is.

Frederick Vallaeys: Yeah. Leverage negative keyword findings. Yeah. And I imagine you’ve a bit like new account builds. So we have this thing called campaign automator. You basically put in structured data and a spreadsheet, and then you can say, okay, crank it out into.

A new plumbing account. And the only variable that maybe I’m changing is the city name. Or if it’s a service company, like what specific types of services do you do? But the more that you can pre fill those negative lists, those targeting ideas that seems to be really beneficial because you’re already.

So you see the thing about machine learning is it could really figure out anything, right? You could say, okay, go and write random ads for me, random landing pages, random keywords, go and figure it out. And sure, a million dollars later, they’d figure it out. But we don’t want to spend a million bucks. And so the thing that we as agencies bring is, how do we limit it?

How do we point it in the right direction so it can do a good job? But that actually leads me to optimized targeting. And I don’t, have you guys heard of this? I’m going to assume, yes. So optimized targeting. For the viewers who haven’t heard of it, Google announced last week that there’s a thing called audience expansion and audience expansion works on display and a couple of other places, but basically it says, okay, if I’m targeting people who are in market for an SUV, expanding that audience would be what is similar to people in market for SUV while it could be in market for a vehicle, maybe they don’t really need an SUV.

They’re going to buy a vehicle, right? So it’s taking your targeting and it’s expanding from that. Now what Google has done, which is really radical is they said. Forget that. We’re going to do it the other way. We’re going to look at what converts and what are the attributes of the person who converts, and we’re going to go look for similar people to that.

So we don’t even care what you’re targeting. We literally just want to see conversions and we’re going to figure out what’s common between the people who you know, have a sewage backup. I don’t know exactly how you’d figure that one out, but that’s basically what Google’s saying. So the people who convert, let’s go find more of them, regardless of how targeting is set up in your account.

Do you guys have a take on that?

Ken Chang: Yeah, I, I guess, like, the initial worry is, like, any time Google rolls out these, these new features is that do your, are your conversions better? Or are they, are they worse? Or are you, are you spending more on the campaign? Are you spending less on the campaign? So anytime they roll out these, These these automations we like to run a Lin Rodsky report to, to look at the quality of our conversions.

Frederick Vallaeys: The lin ratio? Yeah. Tell us what that is.

. So, so basically you look at, you look at all your conversions and then, and then it, it basically generates out a number that that says like, Hey, this is, this is where you’re at in your ad spend. And then like, are you spending too much or too little to, to get a conversion?

Ken Chang: And then. If you could, if you’d loosen things up a little bit to, to get more conversions. So like, like for example, like, like if your account is that like 1. 5, for example that might say like your, your, your You’re spending too conservatively or you could grow your account a little bit more.

Let me give a different example

Frederick Vallaeys: of that. So in the case where you had like a score, I think it was a 1. 0. That literally means that the only conversions you get are from exact match keywords. So it’s great because you specifically said, this is what I want to advertise for, but you’ve given very little flexibility for the system to go and discover new opportunities.

So your, your client might be really happy because you’re coming in at the CPA that they wanted to. But you’re basically limiting the overall growth opportunity. Whereas if your number goes too high, then it’s saying, well, you’re spending a tremendous amount of money on all of these broad matches, close variants.

And it seems like you’re not putting in enough negatives. You’re not putting in enough new exact match keywords to sort of hone in and tell the machine where to go and get more. So that’s what the Linrod Nitsky ratio will tell you. But anyway, sorry, I interrupted you. So you use the Linrod Nitsky

Ken Chang: ratio.

No, you explain it way better. So yeah, yeah, I, I, I like looking at that report because it, it gives me like a, like a velocity of the campaign from like a, like a 30, 000 foot view. So, so like, I would want to run that report and I would just look at like, like, what was my baseline before and then with it after.

And I would just cage that to see if it’s an overspend most cases. I believe like that would, that would loosen up my, my like 1. 5 campaigns into, into somewhere in the two region where, where things would, would grow and, and get better. So, so I don’t view that as necessarily a bad thing. But I would just keep an eye on on conversion quality and and that’s been

Frederick Vallaeys: I said I was just busy typing To lenny’s or lenny’s question. So what’s the difference between audience expansion and optimized targeting? So I wrote a blog post that kind of visualizes the whole thing. Actually maybe I should show it on on screen So let me pull that up while we talk about the next thing because it’ll take me a minute

Kasim Aslam: Can I comment briefly on the?

One, one article that I’d love for people to look at is in, in April of 2015, this is, I think, published by Moz. Google told a woman she was pregnant before she knew, based solely off of her search and communication patterns. Y’all, that was six years ago. They’ve got 70 million demographic and psychographic profiling factors on every human being that engages in the Google ecosystem.

To put that in perspective, Facebook The Google display network reaches 90 percent of all internet users on the planet, 65 percent of whom are reached on a daily basis. Google analytics is on 99 percent of all front facing websites to get really sobering. They know who’s in narcotics, anonymous, who’s about to get divorced, who has HIV, like Google.

If you log into a website, Google has that information on you. So I think that what people don’t necessarily understand or think about is the fact that when Google says, Hey, we’re going to bring you an audience member, that’s more aligned with the people that are already buying from you with enough data, Google can get.

Really granular. And that’s what excites me about things like optimized targeting is it’s not like Facebook’s lookalike audience, or we’re going to go find people that match these five criteria. Google’s learning how people think and act. And it’s bringing more people that are like those people. And also, and this is where it gets dangerous, more people that take the actions that those people take.

And so with optimized targeting, you need to be really careful about the conversion action you’re optimizing for, because if you’re optimizing top of the funnel, you’re going to get a bunch of people that love to download lead magnets, but it doesn’t necessarily mean that they buy. Right. So it, there’s a full funnel discussion and paradigm that needs to be assumed.

And that’s easy for me to say, it’s hard for me to do because what people do at the top of the funnel and what they do at the bottom of the funnel aren’t necessarily always connected. And Google has yet, at least to my mind, and maybe cause you two are clearly much smarter than I am. Maybe y’all have cracked this code, but trying to, to get Google to optimize for a multivariant conversion narrative has proven to be a little difficult.

And that’s what we’re trying. That’s the code we’re trying to crack.

Frederick Vallaeys: Yeah, and thanks for giving me maybe more credit than I deserve. I don’t think anyone You

Kasim Aslam: invented conversion tracking, Frederic. You’re my hero.

Frederick Vallaeys: Okay, that was a long time ago. What have I invented since? I also did invent AdWords Editor while we’re at it.

And I did buy the company that’s now Google Analytics. And I didn’t buy it with my own money, but I was on the small team that did go and buy Urchin out of San Diego.

Kasim Aslam: That’s awesome.

Frederick Vallaeys: Anyway, since we went there. So, but that was a really good explanation, I think, on how optimized targeting is different from audience expansion.

I also put the link to the article that I wrote in the comments, so check that out, that hopefully explains it too. Hey, lots of great engagements, so let’s see if there’s some other questions that we want to Answer here Okay, this one came on screen. So let’s see. Is it better to run smart shopping campaigns per product or for the entire catalog?

Okay Okay, so Smart shopping campaigns, right? So smart shopping campaign is basically the type of shopping campaign where Google does most of this stuff for you Google will say go ahead and just dump your entire catalog into a single smart shopping campaign And let us figure out when that product is the right fit.

I think smart PPC advertisers who have time, right? I mean, we’re all smart, but we often don’t have enough time to do the right thing. But the right thing I think is to actually set up multiple smart shopping campaigns with different targets. And it goes to what Kassim was saying that What we say is the goal to Google is often not really the goal that we have.

So when we go to Google and we say we have a ROAS goal of 300%, that’s usually a made up number. I’ve heard many instances where somebody says, well, I made it 300 percent because last year we were at 250%, so I just wanted to make it better. So I’m going to erase it. Okay. That’s the wrong way to do it.

The right way to do it is to say, I actually care about making profits in my business, or I want to grow my revenue. These, these two things are different by the way, growing revenue, maximizing revenue, maximizing profit. Different things. But say that you actually cared about profits. Well, then you have to figure out for certain products that I sell, these have a different margin than these products, right?

I make more money if I sell a sneaker than if I sell a sock on a percentage basis of the overall sale. And based on what those margins are, I should have multiple smart shopping campaigns with different ROAS targets to help me achieve the maximum possible profitability. Right. So that would be step number one.

Now I wouldn’t over segment it. I certainly wouldn’t put a single product into a single smart shopping campaign. I think that’s just going to become a nightmare in terms of management, but you just want to find things that are similar, group them together, set reasonable targets for those. And then move stuff around, right?

So as you maybe find that I’m going to liquidate my summer furniture right now, because we’re come to the end of the summer, right? I’m going to sell out my my sun umbrellas, the margin changes. So maybe they need to go into a different smart shopping campaign for a little bit so that they actually get the right bits.

So that’s what I would say on that.

Ken Chang: Yeah, I would, I would agree too. And then when you set up that way, you could also try out different ROAS targets too. For your, for your products, if you, if you grew up in that way.

Frederick Vallaeys: Right. And that’s the thing, like what’s the right ROAS target. And it’s hard to know, right?

Because as you change it, your volume of sales will change. So you almost have to like experiment and build a little map that says at this level, I made this many sales, this much profit. And so you start to build your efficiency curve and eventually find that sweet spot where you want to be at.

Kasim Aslam: Can I respectfully disagree on one point? And with the hope that someday I’ll be invited back, Frederick, and you’re not going to. Cut me out of this whole interview. I,

Frederick Vallaeys: We just lost him. We lost him. I’m so sorry about . That looked,

Kasim Aslam: that was awesome. Was that Al because that was amazing. Sorry, that was me.

No, that was you. So we used to segment smart shopping campaigns by margin, low, medium, high. And we still do sometimes what we found though, is, is smart shopping is so heavily reliant on remarketing. It’s very display based. And very often, especially if clients have products that are incestuous in terms of the product buying cycle.

The product that somebody clicks on and the product that they purchase are two different products And what’s interesting about that is the product that they click on is the one that gets the the conversion action credit obviously Which is it makes for some data variance because somebody’s looking at something and saying why I didn’t sell any of these and google’s Telling me I had a conversion and then they go check their shopify dashboard or whatever and it sold whatever it sold Here’s the issue with that is if you separate your products into separate smart shopping campaigns Smart shopping campaigns don’t share remarketing audiences So if they click on product a don’t buy it travel to product b That’s what they’re interested in there Then we’ve we’ve effectively broken google’s and we found this out the hard way We found this out because we made a mistake and we saw campaigns die on the vine that were otherwise functioning and performing So I think separating smart shopping campaigns works really well if your products are very distinct.

You know, we have some clients that have like really broad product catalogs and it’s like tennis shoes and lawnmowers. Well, those will probably never cross pollinate. It’s okay to separate. But if you have products, you know, we have one client who does ophthalmic ophthalmic equipment. Somebody can click on a light bulb for an exam lane.

A light bulb is 4. An exam lane is 30, 000 and end up buying the exam lane. And so you want someone who clicks on the light bulb and is like, gosh, I just need the whole new thing now. And, and, and if that cross pollination narrative is, is possible within that product set you want those products in your smart shopping campaign.

And I would also encourage people not to restrict. It just because product has a low margin doesn’t necessarily mean that you have to rely on that product solely for profitability because sometimes that product is the gateway drug. It’s a loss leader. They come in, they buy that, but then, you know, they buy your candlestick and then they buy 100, 000 worth of wax and you lost money on the candlestick, but then you made money on, you know, the replacement wax for when the candle burns down.

So I do, I don’t disagree entirely. I just think that there, there are guardrails that should be put on that. I just defend anybody.

Frederick Vallaeys: Yeah, you didn’t offend me on what you said made a lot of sense, right? So ultimately it’s like, what is the true business goal? And listen, if you want to have a loss leader, yeah, have a loss leader, but to be able to do that, you need to be able to set those different parameters.

So that makes total sense. The other thing. And then, so you kind of teed up the topic that I’d like to keep discussing. And I don’t think we’re going to have time here today. But so there’s the, what I said was make multiple campaigns, each with a different ROAS, but then you brought in others cross pollination.

So how do you solve that? The way that you solve that is actually by reporting the value of the conversion more correctly. So instead of saying, well, I sold the 5 light bulb and that was 5 or I sold a 30, 000 lane and that was 30, 000 of reported value. How about if you can communicate back to Google, what was the actual profit, right?

So on the 5 bulb. I didn’t make anything on the 30, 000 thing. I made 15, 000. If you use something like conversion value adjustments from Google to feed that data back in, now you can actually run with a single shopping campaign, whether it’s smart or regular, and you can have a single ROAS goal, but you’re manipulating what the system is prioritizing by the way that you report values.

And so you can get very creative with this, right? You can just say. Straight up profitability. That’s my only goal, or you can have like a lot of backend business data that drives these decisions. But that, and that’s the fundamental shift, right? I’m actually writing a new book right now, and it’s about the huge mind shift in PPC management and how it’s going away from us setting bids and choosing keywords and writing every single attacks.

And it’s much more about how do we work with the machine? How do we tell the machine what we truly care about? And how do we push the new buttons that we have to point the machine in the right direction? And it’s, it’s so different from what we’ve used to do. And I think it’s fascinating, but it’s gonna, it is a bit challenging because there may be areas of expertise that a lot of us haven’t, haven’t really built.

Kasim Aslam: Can I tell you why that scares me, though? Because imagine if you’re Google, if we tell Google what our margin is, because that’s what we’re doing, right, like on this 30, I made 10, 000, if you tell Google what your net margin is, Google now has all of its inventory, which it needs to maximize, it has everybody willing to compete for that inventory, and it knows exactly what they make.

The ability for Google to intelligently gerrymander So that you’re making just enough money to continue to advertise because they know exactly what your net profitability is and spending the maximum amount on this trial. We’re basically we’re showing all of our cards to an entity that, you know, God bless them.

They’ve done some phenomenal things, but they’re not like the great magnanimous ruler, right? They’re, they’re the tyrant and we’re arming them with the information necessary to extract. The ultimate amount of value out of us. So I, I, I see exactly what you said, Frederick. And I think it’s brilliant. I can’t wait to read your book.

It’s, it scares me because we’re, we’re giving them everything. We’re giving the keys to the kingdom, everything they would need to make sure that we spend the most amount of money as advertisers, we’re putting affiliates to our own products, man. Like it’s unbelievable. Google is the one that’s profiting from all of this.

Frederick Vallaeys: Right, and that’s sort of the trick then, and I mean, I was the ads evangelist, so I drank the Kool Aid, and it’s the new system.

Kasim Aslam: I’m talking to the enemy now, basically, is what’s happening.

Frederick Vallaeys: But yeah, I mean, ultimately it’s an auction, right? And so I think there are levers that Google has to set thresholds.

Kasim Aslam: Not to interrupt you, but is it an auction when you’re using automated bidding? Automated smart bidding where Google’s assigning your bids based on what it thinks is going to convert from you and doing the same thing for the next three competitors in the four pack. It’s, it’s not an auction if the auctioneer sets the rate.

Frederick Vallaeys: No, it’s an option.

Kasim Aslam: But the auctioneer is telling you what to bid

Frederick Vallaeys: because you’ve asked the auctioneer to do that for you.

Kasim Aslam: But that’s where this gets. Now, how much do we trust this auctioneer?

Ken Chang: So, so if you don’t trust the platform, there’s like, you know, like Microsoft ads and then

Kasim Aslam: you have

Ken Chang: a

Kasim Aslam: mic you can drop.

Cause that was the best. Yeah. So

Ken Chang: you could, there’s always, there’s always like other channels that, that you could advertise on and even newspapers. Yeah.

Kasim Aslam: Yeah

Frederick Vallaeys: I think you’re going to be happy to hear that. I am a subscriber to two physical newspapers I get delivered to my door every morning. I also subscribe to three digital ones.

So i’m a big fan of the industry So Keeping a good journalism alive. Thank you all right, so many why don’t we talk separately? There might be many things to do here to fix your TROAS problem. We’re getting sort of close to time here, so I did want to give both of our amazing panelists a final chance to take one of the questions that they thought was really interesting, maybe answer that or say what they’re up to.

But Ken, why don’t we start with you? Final thoughts?

Ken Chang: Sure. Basically, basically don’t be afraid to try out automation. I, I feel like when, when we, we first tried out automation, we’ve been burned so many times. And it, and it’s just largely related to To I would say for us, it was user error just just because we weren’t we weren’t thinking about providing the right inputs into into how Google thinks because Google’s interface is built by a bunch of computer science geniuses.

And then we have to try to think like a computer science genius. And then through trial and error, we’re able to leverage that automation. So I would say don’t be afraid to To try it out. And then and also check out our Foundry 425 product that we’re launching too we’re launching a branded agency within, within the star Tribune that serves for creative content and, and is, is led by humans.

So, so a little bit, a little bit both paths, so automation and, and humans at the same time.

Frederick Vallaeys: Nice. We’re going to show that up on the screen here in just a minute. Custom, what about your final thoughts?

Kasim Aslam: I know I’m being positioned as the anti automation guy. I’m not anti automation. I don’t think that Tony Stark wouldn’t be Iron Man without the suit.

The suit doesn’t make him Iron Man either, right? So I think that there’s there’s a merger between those two things that Needs to take place and that’s sort of where I sit on the fence and I I just want to say I’m so grateful for being on the show You know chatting up with folks like you that are I mean Ken what you’re doing the numbers that you shared You know four staff members running 300 accounts.

You’re clearly doing something freaking brilliant and that’s really awesome I’d love to learn more and Frederick. You’re like the godfather of you know, google ads and ppc and all those things So i’m flattered to to be on thank you

Frederick Vallaeys: Yeah, it was great having both of you. And thank you to the audience.

This was a amazing engagement, a lot of side conversations happening, but we’re here to learn from each other. If anyone else feels like, Hey, they want to come on and put their face on the screen, reach out to us. Just drop a ticket to support at Optmyzr and we’ll, we’ll have a chat with you and hopefully bring you onto the show and keep these engaging.

If you want to see this again, subscribe to the channel. We are going to have. Two episodes coming up in August. One of those will be with our friends from Google as well as an agency that will bring into sort of counterbalance the Google viewpoint. Maybe Kasim, you should be back to be that counterpoint.

So we’ll see Andrew Lowe is another one we like bringing on. But thanks so much for all the questions. Thanks so much for watching. Have a great week and we’ll see you at the next PPC Town Hall. Thanks.