If implemented properly, Meta Ads can be a cost-effective way for a publisher to drive website traffic and grow their email list.
In this High Value Publishing session, Eric Shanfelt is joined by digital industry legend, Dennis Yu. Dennis is the CTO and co-founder of BlitzMetrics and a former search engine engineer. He’s managed billions of dollars with Meta ad campaigns.
Eric and Dennis show two different, cost-effective, ad campaigns for publishers: one to drive website traffic and another to rapidly build email lists. They show how each campaign is built, what they look like, and how you can do it for your brand.
Full Transcript
Eric Shanfelt:
Hey gang, welcome to this week’s High Value Publishing session. My name’s Eric Shanfelt with Nearview Media, and I am joined today by a good friend of mine. Dennis Yu. Want to say hi, Dennis?
Dennis Yu:
What’s going on, Eric? How long have we known each other?
Eric Shanfelt:
It’s got to be 8 to 10 years now, I would think.
Dennis Yu:
So I’m here because I believe this is super high value for publishers. I’ve seen what Eric is about to show you, and it’s worth your time if you’ve not seen ROI on social media.
Eric Shanfelt:
I appreciate that. I do. Yeah. This is going to be a fun session. We’re going to talk about Meta ad campaigns. We’re going to dive into that here in just a second. I do want to get the housekeeping out of the way though, so hey, you guys go out to highvaluepublishing.com. That will redirect to my website to all these sessions. Sign up for our newsletter there so you never miss one of these sessions. All these past sessions are available for free on the website. Any links we talk about are going to be in there. It’s also available as a podcast on your favorite podcasting platform. Dennis, what’s your favorite podcasting platform, out of curiosity?
Dennis Yu:
I like all of them. We distribute to all of them.
Eric Shanfelt:
All of them. Yeah. I’m personally an Apple dude myself, but we distribute to all the platforms as well, wherever people are. Alright, let’s dive on in here. Dennis and I talked about Meta ad campaigns for publishers, so I wanted to just first introduce you guys and brag a little bit on Dennis. There are very few people out in the industry who I think really know their stuff and are willing to talk about it and share it openly and freely. And Dennis is one of those. If you follow him, which I highly recommend you do on social media, he is constantly sharing ideas, constantly sharing perspectives. He has a passion for teaching people. So he’s the co-founder and CTO of Blitz Metrics. And Dennis, I’ll let you kind of introduce the rest of yourself here and give a little bit of your background and kind of what you’re doing.
Dennis Yu:
I’ve spent a billion dollars on Facebook ads, so I feel like I’ve made more mistakes than almost anyone on the planet when it comes to this. Call it Meta, Facebook, whatever. And prior to that, 20 some years ago, I built the analytics at a search engine. So I’m a search engine engineer, I’m a data-driven person. Sure, I’m sharing thoughts and perspectives and whatever all the other people, but I’d like to think that what I’m sharing is based on actual experience that I have seen firsthand, things that I have done firsthand where I’ve been inside people’s campaigns and we put these things together into checklists and techniques that everyone else can follow, like a recipe. And I think the more people like what you’re going to see today with Eric, have recipes, you should be able to follow the recipe and not worry about who the guru was that came up with it.
Eric Shanfelt:
That’s exactly right. Yeah. Dennis has his big recipe, if you would, as his dollar a day strategy, which was real interesting because when I started to learn about the dollar a day strategy, I was like, oh, this matches almost identically what we’ve been doing for publishers. And so we connected over that and began to share ideas together. Dennis, talk a little bit about what you’re doing with your job creation. I think that’s also really cool about what you’re taking, what you know about SEO and social and applying it to improving people’s lives around the world.
Dennis Yu:
Yeah, so 20 some years ago when I was at Yahoo, which was a tiny company, we needed to train up a lot of people and I was recruiting kids out of Stanford and other Ivys and what-not, teaching them computer programming. They didn’t really have any background, these really bright kids. Fast forward, Facebook opened their platform for ads in 2007. I was one of the first people there and I had to train up a bunch of people and TikTok opened up their thing and we’re training up a bunch of people and we’ve been training people from the Philippines and from Pakistan and all over the world. You have your favorite place for VAs. We have this operations process guide that shows how we do everything. So we’ve been publishing everything that we know how to do for job creation, and I became sort of an unintended teacher, if you will, because that was the only way to be able to help more people at scale because I didn’t want to be the one doing the work, but I needed to make sure the work that was being done by these other young adults, by other agency owners, what-not, was being done competently.
So having these checklists out there gives everyone a clear standard. So if you’re going to hire someone from Upwork, if you’re going to hire an agency, they need to follow a checklist. And we’ve created a quarter million jobs so far, so we’re a quarter of the way there towards a million jobs. I want to see that now. Friends say, Dennis, your number is too low, you should make it like 5 million or 10 million. Who cares what the number is? The point is put everything out there, give away that value like what Eric is doing here. And then people want to hire you in whatever they can pay, but give away your knowledge. The idea of creating jobs is there can be a clear standard that whether you’re a worker getting paid to do digital marketing or you’re hiring somebody, there’s a clear standard. You’ve heard of Carfax, right? There’s a clear standard. You can see if that car has been in an accident and been repaired and had something replaced, so why shouldn’t there be a clear standard in digital marketing? There isn’t until now.
Eric Shanfelt:
I love it, I love it and I get to see a lot of his posts when Dennis travels around the world and he’s changing people’s lives around the world on this and I really appreciate that. So I think it’s incredibly cool what you do, Dennis, and I think we both share a passion for documentation and process, which is something that not a lot of companies actually do very well.
Dennis Yu:
They don’t have the checklist and SOPs kind of don’t trust him. That’s right. Tomorrow morning I’m flying to Austin and I guarantee you that pilot, even if he’s got 20 years of experience, he’s going through a pre-flight checklist. And even if he’s played Microsoft Sim and he’s watched all the YouTube videos and had to fly an airplane, if he hasn’t gone to flight school, I would not get on that plane.
Eric Shanfelt:
So I actually am an instrument rated private pilot, and indeed, every time I take off every single phase of flight, every day, every flight, I do it.
Dennis Yu:
You’ve done it before. Oh, I already know the checklist, I don’t need to go through the checklist.
Eric Shanfelt:
Oh, no way. That’s what the checklist is for is to make sure you don’t forget something. Alright, well hey, let’s dive in here. I want to talk about, this is one of my favorite little simple diagrams here. This is what I call the audience conversion funnel for media companies. Really, it focuses on any company, but I don’t care what market you’re in, whatever your potential market size is, our goal is to move people down this funnel to get potential visitors to become potential market, to become a website visitor, get the website visitors to become email subscribers, get them to become paid or controlled subscribers to your publication and then retain them. So, today we’re going to focus on two particular kinds of Meta ad campaigns. One that will move people from being a potential market, finding out about your product into becoming website visitors. And the second one actually moves people from being a potential market into becoming email subscribers.
Usually it’s very hard to jump stages, but we’ll show you a couple of campaigns today that will move people from potential market to website visitors or from potential market to email subscribers down here. And then I’m looking forward to getting Dennis’s critique and ideas on this as we go through this. So let’s dive on in. Alright, first of all, let’s talk about website traffic campaigns. I wanted to share, this is one of my clients who wanted to launch a website traffic campaign–happened to be a relatively new publication and so they’re not really well established in SEO yet. They’re building it. They don’t have a really big email list. Matter of fact, when we started, they only had a few hundred names in their email list. They’re now up to several thousand names in their email list. So we started the website traffic campaign and we utilized Meta.
And you can see that this is just from June through the end of January. So was it nine months or so? 38,000, almost 39,000 clicks at an average of 6 cents a click. And they spent $2,400 to get these 38,000 people to their site. So just to show you what that is—since from June to January, that’s almost 40% of their traffic to their site. Now over time, the paid social is actually getting less, not because they’re spending less, not because they’re driving less traffic, but because more people are getting aware of their site—they’re building their email newsletter list, right? They’ve now got 18% coming in from email where they had almost 0% before. And not only is this driving traffic, but we’re also tracking email sign-ups that come from this campaign and they’re generating several hundred email signups on this as well. So Dennis, curious to get your thoughts on this kind of a campaign and do you often sometimes do campaigns simply just to drive traffic to a site, not necessarily for other objectives?
Dennis Yu:
Yes and no. So go back one slide. So 6 cents a click, that looks good, right? Well, how do we know those clicks are any good or we get a ton of email signups? How do we know those are any good? What I want to see is that those email subscribers turn into subscribers, turn into people who engaged, actually fit the people that you want to reach. So a media company that has a very defined niche in a particular industry where its content, I hate to say it, but is actually of high quality, then when you run these campaigns to be able to amplify, the signals get passed back to Meta on who is a good visitor, not just that I got a link. You can pass back through the conversion API through different sorts of tools that allow you to pass that. It could be through tag manager, other tools. If you pass back the signal of who you want to reach, then the 6 cent link click is good. But if you just run stuff willy nilly and you say, ta-da, I got a 4 cent link click, yes, but that’s a secondary indicator, that’s a diagnostic metric, not a true metric. So this is good, but as Eric’s going to show you just a little bit, there’s other things that we’re looking for to make sure that this early indicator of people engaging and clicking and what-not are actually helping your publication.
Eric Shanfelt:
Great point. So remember we’re talking about potential market in the website visitors. So hey, they’ve driven almost 39,000 clicks, and by the way, we’re measuring off of website page view visits off the pixel, not just using the link click itself, but that 39,000 is important to them because they’re still trying to establish themselves out there. Now when we look at other metrics, the times spent on site, the page views per visit, their engagement is not as high as if they have someone coming in from an email.
Dennis Yu:
Of course, they’re brand new.
Eric Shanfelt:
Interestingly, it is about as high as organic search.
Dennis Yu:
So that’s what you’re looking for. Plus it’s going to be a high percentage of mobile, probably 80 plus percent mobile.
Eric Shanfelt:
Very high percent mobile.
Dennis Yu:
A typical B2B might be 25, 30%, the LinkedIn traffic will be way higher, desktop versus mobile. So compare apples to apples.
Eric Shanfelt:
You’re absolutely correct. Yeah, so a hundred percent agreed. Interestingly, I didn’t put this data in here, but of these people who all came through, they’ve actually converted multiple people into one. So this is a donation driven site, it’s an open site, no paywall donation driven site. They’ve actually generated probably, I’d say close to a dozen of these people into paid subscribers as well on there. So what I’m looking at it here is not only the growth and the traffic, but how it’s driven, the growth in the email and then even further down the funnel into converting into actual…
Dennis Yu:
That’s the key
Eric Shanfelt:
Subscribers, donors key.
Dennis Yu:
And it’s not just the 39%, it’s the fact that that drives all these other components. And later this year, Google is finally killing off the cookies in Chrome, which people don’t realize is a huge impact. And then what Apple’s been doing with iOS and the privacy data, people don’t realize how important that is because they don’t realize if you don’t pass back the right data, the system can’t optimize for you. So any kind of intelligent ad system like Meta Google or TikTok or what-not, if you don’t pass that data back, you’re just spending money on garbage. Let me give you one example. If you ever go to Costco, it’s my guilty pleasure, and they have these ladies that have the samples on the toothpick. You can try the little, whatever the snack is, right? Right. And what does it cost for them to have the woman there and they get 50 little things of beef jerky or whatever it is, and you try one and if it’s good, what do people do? They’re like, it’s because you feel guilty. This old lady, they’re like, okay, okay, I’ll buy one.
Eric Shanfelt:
I’ll buy one!
Dennis Yu:
Of the beef jerky or whatever. And that’s what we’re doing here with these engagement ads or driving link clicks. You’re the Costco lady handing out those free samples.
Eric Shanfelt:
That’s a great analogy.
Dennis Yu:
And, if the free sample is good, people will do it out of guilt, out of, actually this is good. I like it. They see the other people that are also using the product. They come to your site, they see what’s there, they sign up for the newsletter. If your email nurture sequence is good, then what Eric is showing you with what he’s done to get more emails is good. But if you don’t have a good email nurture sequence, if you’re not super clear on what exactly you do and you don’t feature the other members, I mean you have to have a great publication. I don’t know how else to say this. If these things are good, these other steps of the funnel are good, then the things that we’re talking about here on website visitors and email subscribers will be really good for you. But if these other pieces are broken, no amount of magic through Eric’s techniques are going to help you.
Eric Shanfelt:
I love what you said, actually. There’s two things that I’ve keyed in on what you said. Number one, the content’s got to be good. If you got crap content, you’re just phoning it in, people. This is not going to be very successful.
Dennis Yu:
People say all the time, you’ve got to have good content, but what the heck does that mean, right? Everyone thinks their content’s good content. Who would say all my content? I’m a bad content publisher. No one ever says that.
Eric Shanfelt:
I’ve seen a lot of it out there. But anyway, you got to know for your audience, if it resonates with your audience, what meets the needs of your audience? Either intrigues them personally, intrigues them professionally, helps them do their job. We could go into all what makes really good content, but I think that we’d be diverging into a different topic here, but you’ve got to have …
Dennis Yu:
The only reason I mention this is that if you want to know what Eric is showing with these techniques to drive website visitors and emails and if it’s going to work for you, I see a lot of people fail, then the answer is, are the stages later in the funnel working? Are you getting at least a 40% open rate on your emails? Is your content so good that when people …
Eric Shanfelt:
Post iOS 15
Dennis Yu:
In your Google analytics, are you seeing good engagement and return visitors? And if all these other signs are healthy and you can produce actually valuable content, which is what you’re boosting to get website visitors, not like straight up selling sorts of things as Eric’s going to show you in a minute, then you’ll win. But the reason I’m telling you this is that I have seen this thousands of times and I’ve seen who wins and who doesn’t win. And I’m telling you who will fail attempting this strategy here.
Eric Shanfelt:
I totally agree. The other point that you brought up, not just the good content but the rest of the funnel. It’s one thing to just bring people in and become a website visitor, but you better have a really good strategy in place to convert website visitors and move them further down the funnel to capture that email, right? I’ve always said, look, I’ll take one email address over a thousand page views.
Dennis Yu:
It’s quality over quantity. The more niche you are, the more important because some of those subscribers might be worth a thousand dollars or what-not. Who cares if you had thousands of clicks? It doesn’t mean anything necessarily.
Eric Shanfelt:
Easily. And then do you have the right email nurture sequence to help move them into becoming a paid or controlled subscriber? So you’re absolutely right. While we’re looking at something in isolation, a web traffic campaign, this is only as successful as you are at creating good content that resonates with your audience. And as you are able to convert those website visitors and move them further down the conversion funnel, excellent point. I just want to throw out this little thing on a campaign structure, then this kind of resonates with your dollar a day strategy. But what they’re doing here is they’ve taken, they looked in their Google analytics and said, hey, excluding these campaigns, what are our articles that have had the most engagement that resonate the most with our audience? And so they’ve taken those and they put their five best articles in there and they’re updating this every single month.
They’re adding and removing articles and it’s been interesting. You can see it’s this one right here that’s just done the heavy lifting on all of them. Now that’s not all of them. That’s only 8,400 of the 38,000 clicks. But you can see it’s done the heavy lifting recently of the clicks in there and we’re just letting the algorithm choose and working with people what’s going to generate the best results for the buck on this. And then here’s just an example of when these traffic ads look like they took the ad in here, by the way, because this is kind of a politically-oriented one. We had to go all through the wonderful organizational and personal confirmations with this to get it, which is a pain in the butt. But I mean the engagement, this is the engagement on top of that has been fantastic.
Dennis Yu:
I want to point out one thing about this. A lot of people will say, okay, yeah, it’s true. We load up a lot of creatives because one of them will win and we don’t know in advance. They’ve got a whole group of creatives which is the winner. So we let Facebook pick that is the core behind dollar a day, which is you think you know? Not true, right, and what performs best on the web isn’t necessarily the best on social. So you have to take things that have already worked like Eric said, and then let Facebook or Google or whatever platform kind of determine what works best with their audience. So everyone understands that that’s a one-on-one kind of thing, yet people don’t follow it. But here’s the more advanced insight. Go to the next one where you’re showing that one ad. Notice this, it’s not that we’re getting a lot of link clicks or the engagement rates high. Yes, those are all initial factors, but look at this. Do you see that that thing has more shares than comments? Why is that so important?
Eric Shanfelt:
I’m curious to get your thoughts on that, Dennis.
Dennis Yu:
So we did some research off of several trillion impressions. I think we have a hundred times more data than Cambridge Analytica. I don’t think many people know that–I could have been banned or whatever.
Eric Shanfelt:
Don’t tell anybody. There’ll be another investigation.
Dennis Yu:
I’ve done political campaigns, but nothing dirty, nothing. I mean nothing that if anyone were to share, I’d be like, I can’t believe Dennis worked on so-and-so’s campaign, but here’s this. So we found that a “like”, everything is a point driven kind of system to a degree. So a like is worth one point. A comment is worth six and a share is worth 13. And if you think about it, that makes sense. I had it confirmed with some engineers at Facebook. So a share is basically treated on Facebook as another post. It has its own post ID. It’s as if when they hit share from a technical standpoint, that’s like them making their own post, they’re putting their own personality and reputation behind it when they hit share. A share is a super, super like, if that makes sense.
Eric Shanfelt:
Yeah, I see that.
Dennis Yu:
Okay, if that’s the case, then yes, of course we want more shares, but when the ratio of shares, when there are more shares than comments, what does that tell you? And then when the shares are almost as many as the likes, wholly moly versus I’ve seen people, especially media companies, the ones who seem to go after vanity metrics or they’re just checking the boxes of producing X pieces of content, especially the paid media folks inside the larger media companies, they’re just checking. They’re like, okay, we spent this much–they’re just spending the budget with any sort of consideration of the ROI like an ownership. You’re watching this. You’re probably thinking from an ownership standpoint and your people are not thinking with this kind of hunger that you want to have. So they’re like, yep, we got 10,000 likes. We had this many people to the website, we spent this much money. Yeah, so you got 20,000. So let me ask you, Eric, let’s say this was one example we’re looking at here. Let’s say the other example was it had 10,000 likes on it, but it had only some comments and shares a few versus this one which had only 2,500 likes, but 1300 shares and 1100 comments, which would you prefer?
Guess it depends. Let’s say 20,000 likes end objective. This one has only 2,500 likes, but this other one, option B has 20,000 likes on it. Which one’s better?
Eric Shanfelt:
To be honest. Now you can tell me from an algorithm perspective and how that pans out and I have a feeling what you’re going to say. I’m still looking though at an end result. So what was the objective of this campaign? The objective of this campaign was these guys needed traffic to begin to grow their site and show their investors progress. They need to begin to…
Dennis Yu:
Build their, yes, we’re talking about diagnostics. Yes, you want the end result, but to get to the end result, you have to test through multiple options. And then when you find a winner, you put more money on it and all that and let it run forever like you’ve done. But if you don’t understand why a particular ad was a winner, then you’re just some blind robot dumping stuff into the system and hoping and praying it’s going to work. You need to understand why this one was a winner.
Eric Shanfelt:
No, that’s great. That’s great insight.
Dennis Yu:
Especially if you going to create different variations of it. Once you find a winner, you’re going to keep trying to make a better and better version. You’re just going to sit there and be lazy and say, oh, I’ve already won. I don’t have to do anything. I’ll just sit here and let it run for a year like you’ve done right? You always want to get better, right? What are you going to do to get better?
Unless you see why something’s winning, how do you know what to adjust?
Eric Shanfelt:
Good point. So you obviously place from what your numbers are here, much more value on comments and way more value in share. So if you’re looking at this here, this has pretty good resonance.
Dennis Yu:
And the nature of this is almost clickbait. If you look at this, what liberal elites don’t know about rural America, it’s like, oh, what do they not know? I’m really curious now. Then they see this picture of the dude who’s thinking—so the order is they see the picture, they read the headline and they move up to description. Then after that they think, has the media actually ventured into rural America to ask what the problem solutions really are? Wendell Berry doesn’t think so…which is like, right. Then I click on it, I learn more and then I click on it. So that’s the order. So when people are going through Facebook, they don’t read top to bottom, they scan. That’s what you do when you go through the newsfeed. So when you understand the structure of how people go through the newsfeed, especially on mobile, which is what this is you’re showing here, then you’re designing that experience around creating this kind of interest enough for them and then they go back to the bottom and they click on learn more. So this definitely is this really arouses curiosity and then when they do read the article and they find that it is good content, then they hit share. So did you see how many steps happened here? It wasn’t just like, oh, we got link clicks for 6 cents.
Eric Shanfelt:
Nope. And you hit the absolute key: the time and effort and care it took into us to craft this where it really followed that flow that you’re talking about, Dennis, was very, very intentional and just be very sure that we even are staying within the right character limits so it’s not having to expand and do these kinds of things. Clickbait. Yeah. So clickbait is one extreme. The other extreme I’ve seen a lot of media companies do is they give you everything you need right here and now there’s no reason to even click.
Dennis Yu:
Yeah. And so each component in this flow is designed to do its job, but if you try to make the image do everything, or if you try to make the description carry everything, then it’s just too heavy. If we know that the image has to do just enough to stop them from scrolling, that’s it. Just to get their attention. Then they come down and they read the bold—what liberal elites don’t know about rural America. Now I’ve established what the topic is. Then the description goes a little bit further, like, okay, a little bit more detail about that, enough for me to click on learn more. Now I have to be willing to leave Facebook for a minute. Right? Exactly. Or Instagram or wherever it’s being placed. So that’s the flow. So each part that you mentioned has to do its job, but if you try to make any one of these components do more than its job, it’s too heavy. The weight is too heavy.
Eric Shanfelt:
By the way, this is running on both Instagram and Facebook, but interestingly, we’re getting the bulk of the results off these off Facebook.
Dennis Yu:
Probably because the nature of the audience and then Instagram is less likely to…people who are on Instagram are less likely to click away.
Eric Shanfelt:
Exactly. Alright, well it’d be fun to go dive more into this one, but I want to go onto the email lead gen campaign and again, back up to our market, it’s taking here from potential market straight into email subscriber. So, but by the way, we didn’t get deep into the targeting and choosing the target audiences and how the target audiences were chosen to do these, but very, very carefully chosen. You can see here we’re doing a look-alike audience to a degree. Obviously that’s changing with a lot of stuff that’s happening, but we’re going off of a look like excluding current email subscribers on that because we don’t need to re-sign them up, but let’s go email…
Dennis Yu:
Fact it’s working means there’s no targeting necessary. If the signal’s being passed back and the campaign’s already working, meaning it’s exited the learning phase and it’s starting to scale, you can actually take that look-alike, which is training wheels, copy that, and then run on targeted against just the United States, 25 plus or whatever the demo is. You don’t need a look-alike anymore.
Stolen my thunder, ha. Those are our two other ad sets we’ve got going right now.
OK
That’s good. That was good advice.
No, but everyone knows if you’ve run a lot of Facebook ads, that detailed targeting basically died five years ago. I mean it hasn’t officially died, but it’s being taken away from us. So targeting doesn’t matter. The targeting is your own data that you bring back through. So we use tools like Leads Bridge or customers do AI to pass back the data of who we want to reach.
Eric Shanfelt:
Agreed, agreed. Alright, so let’s talk email lead gen. So there’s several types of this that we’ve done. Depending upon if you’re a consumer enthusiasts kind of publication, kind of a hobbyist or sports enthusiast. If you’re B2B National Trade Journal or if you’re a regional publication like 5280 here in Denver or D magazine in Dallas. This is one that was a regional sports publisher up on the northeast region, and I wanted to share this one here. So, just January through March, just the last two and a half months basically, they generated a thousand email sign-ups, 85 since lead. Now interestingly, this campaign has been running for a while, but they took a little hiatus to focus their Meta spend over the holidays, November and December into conversion down into here. But what was interesting about this campaign is they didn’t really know what to put together for a lead magnet.
So we decided let’s take their actual product, their current issue that they have out there and now their editors are getting this. So before it was us working on this together as with their audience team, but now their editors are getting into this and their editors are actually putting this up and writing these and saying, hey, basically you get the checkout Lacrosse journal, see the prep schools here, and this is the actual cover that they’re running. That challenge with this campaign, Dennis, is because it is based upon their current issue, they’re actually swapping out the creative once a month.
Dennis Yu:
So it has to die every time. Yeah, you keep…
Eric Shanfelt:
So it’s rebuilding a learning, but nonetheless…
Dennis Yu:
Make it evergreen though…you can. It doesn’t have to be, you can still have stuff where you boost the current issue.
Eric Shanfelt:
We’re going to do that. They’re moving to…
Dennis Yu:
Evergreen is what works because then you can keep multiply. It’s like compound interest. The other thing too, did you notice that? Go back. It looks like a magazine cover. It looks like if I’m in the Lacrosse and clearly these, I care about the rankings, this looks good, I’d be interested. And it already sets the expectation that you’re going to fill out a form and get a free trial. But here’s another thing too, Eric, this might make you cringe a little bit. So the first thing people are going to notice is these two guys playing lacrosse and then below in bold: prep school, preseason boys, “rankins”…
Eric Shanfelt:
Misspelled, Lord almighty, they have a typo…
Dennis Yu:
But this is really good. I’m wondering if you did this intentionally. We’ve done this before.
Eric Shanfelt:
No, this is not intentional.
Dennis Yu:
No, but let me tell you something. So when a brand or the client sees this, normally it’ll be the director of marketing. They freak out and they think the agency is making mistakes. Maybe you have VAs in the Philippines that are making mistakes. What happens? I make mistakes, I have typos, but I’ll tell you a story. And then here’s the bigger learning. So we did this for Ashley Furniture, which is the world’s largest furniture retail. They have stores, they’re bigger than Ikea. And we were driving sales, we were trying to get people into the stores to buy furniture. What a surprise, using Facebook ads just like this. And we had, one of our best performing ads was of a couch that they didn’t even offer for the past five or six years. It’s not even in inventory. And that was the thing that was getting people into the stores and by the time they got into the store they’d say, hey, I’m looking for a cheap couch.
Is this one for $299? I saw on Facebook? And the salespeople will be like, but let me show you to the couch section and here’s all these other couches. And we never got a single complaint over it, even though that couch has been out of stock. And initially when the Ashley Furniture people saw, hey, these guys, they messed up. They’re selling couches or like here, missed the rankins instead of rankings. It actually is, if it’s working, it’s usually okay to let it keep running. Some people will take this image and they’ll turn it upside down intentionally like they’ve made a mistake. I’ve
Eric Shanfelt:
I’ve seen that before as well.
Dennis Yu:
And it causes people to comment and all this and then that tells the algorithm, wow, people are commenting on it. Now of course, don’t do it in such a way that it makes your brand look bad, but sometimes it’s okay. I bet you didn’t notice wish, say this was planned, your
Eric Shanfelt:
Not planned.
Dennis Yu:
Your brain read the thing and it already said, oh, preschool preseason rankings for lacrosse. I’m a coach or a player. Yeah, I definitely want to see where we’re stacking or who’s ranked. And so you’re scrolling so fast. Now imagine this is the freeway I-15, but not with a ton of traffic. When you’re going down the freeway and you’re reading billboards, you’re seeing billboards. It’s going by so fast, you don’t even notice that rankings is misspelled. So it’s not even a big deal.
Eric Shanfelt:
So I’ll share you a little bit of insight. So what’s interesting is my organization, we’re not an agency, so we are not actually doing these for folks. We’re a consultancy. So what we do is we work with media companies. We teach them how to do these and then let them even better do it themselves. So this company we’ve turned loose. They’ve got six different brands, and so they’re running their own lead gen ads.
Dennis Yu:
But you’re telling me in spite of this, it still worked. So what does that tell you?
In spite of this, it’s still working, dude. It’s still work
It means if you do things right, generally the system’s very forgiving.
Eric Shanfelt:
It is. Nonetheless, I think they’re going to want to somehow adjust that, but I think you got a good point here. It’s saying, hey, they’ve been very focused on saying, hey, we only want to put out the current issue. Then we’re going to take off the old issue. But as you and I know, evergreen is the key. We did one, we’ve done this for a couple of city and regional publications where they usually have, what is Minneapolis, or St. Paul or what is Denver or where is Dallas or whatever it may happen to be. They have their best restaurants, right? Oh, that’s Evergreen. Best restaurants. It is the killer lead magnet in the city of regional magazine space we are getting, and it’s evergreen because they can update that whenever they want to. But the ad stays the same. We have multiple creatives, right? Right. It’s that evergreen content that we’re looking for on this. I think with this example of the current issue, maybe we just keep these past issues still running and new issues and let’s see which one builds up the best over time.
Dennis Yu:
Right, but just make sure the topic is evergreen. So this preseason rankings thing, well, once you’re in the middle of the season that’s in the playoffs and all that, that’s not going to matter so much. So the topic has to be evergreen too.
Eric Shanfelt:
And that’s one of the challenge with sports. It’s very seasonal.
Dennis Yu:
But you can always highlight, I’m a big fan of the NBA, partly because they paid us a lot of money, but when you focus on Kobe or other stories, that’s evergreen. Whenever it comes down to the story of a player and what they’ve done and how they grew up and the challenges they overcame and they negotiated this big contract, there’s all these stories. So stories are the most powerful thing in social media, and they also happen to be Evergreen. News is …
Eric Shanfelt:
Fleeting
Dennis Yu:
News. Okay, fine. I’m not saying don’t publish news. If you’re an industry publication, even a B2C, enthusiast, sports, media, whatever, that’s great, but the more you can tell, do both. But the more you can get on stories which are evergreen and drive the engagement and show the quality of the writing and editorial versus news. News, they can get on Twitter in these other places. But stories, the curation, the editorial, that’s why people pay news. They can just get that for free anywhere, and then it’s all about who breaks it right away and now you can compete against TMZ or whoever. Right? Right. Exactly. Against the other consumers who were there with their cell phones. Now they’re competitors of yours. They’re a publisher too. Yes.
Eric Shanfelt:
It doesn’t have to be a big fancy thing, either. I mean, one of the best ones that I ever worked with was with a general aviation, so a pilot publication and the dude put together a one page little checklist for a knee board. If you’re a pilot, you know what a knee board is. There’s a one page checklist that kind of took you through every single step of the flight on one single page. You got thousands upon thousands of downloads off of that one page little PDF download that took him 30 minutes to put together, and it is still generating leads even to this day. That was four years ago.
The signal is there, right? The signal of engagement and downloads and whatever clearly is showing because the users actually like it. Yeah, you can turn it into an ad.
Yep. I wanted to show you then one little quick thing though, so the email lead gen campaign, this is what the email sign-ups were like before the email lead gen campaign started, and this is the email sign-ups after the lead gen campaign. So huge quantum difference in terms of growth in the email list since that lead gen campaign ran. Talk briefly about some keys to success. We’ll have to start wrapping up here a little bit. You’re a big fan. You and I both use the same word: plumbing. You’re a big fan of the plumbing, so why don’t you touch briefly on the fact that you got to get the plumbing set up correct.
Dennis Yu:
We’ve been working on this thing for 17 years. Digital plumbing is having all the tracking and data together, so it starts with your Google tag manager, which seems really technically complicated. It’s not. We have a whole 200 page guide on it with videos. The one tag that ruled them all contains all your pixels, so your Facebook pixel, the ad pixels, the analytics pixels, the pixels from all the other networks. Even if you don’t use those other social networks, might as well put their pixel on there. They’re always synchronous anyway with the JavaScript..
Eric Shanfelt:
Build remarketing audiences
Dennis Yu:
Which then informs the, because the whole point of plumbing is, I need to capture the signal of what’s working so I can pass it back to Google and Facebook to optimize for me. Then the part of any doing ads, whether it’s your team being coached by Eric or what-not, is then largely around the creative strategy, but without plumbing. Then you have this quality/quantity kind of issue. So this includes conversion tracking, conversion, API, or if you don’t know how to use a conversion, API, if you’re using a HubSpot or a leads bridge or whatever, there’s usually an integration there to pass back conversions, especially with the whole third party cookie death and all that. UTM parameters is part of this as well. Even if you have all the cookie tracking and tags, you still need UTM parameters because there’s direct none. There’s traffic that comes from apps.
There’s other things where you still want to be able to assign a source, whether it’s by the campaign or the channel or the keyword or whatever it is, where you can use these different UTM parameters, right? The beauty is once you set this up properly, then as you launch more campaigns, you use the naming convention or you have links that direct from your newsletter. The biggest fail I see is people sending out their email newsletter and they’re not using UTM source equals email, and then it shows up as direct none, and they don’t know the difference between web traffic versus email traffic, and then if you have no idea what you’re dealing with, like if you can’t see if you’re blind, you get the blindfold on, you can’t freaking see what’s going on, so everyone wants to jump to the latest TikTok thing or this new strategy or this new tool, and they forget about the fundamentals of plumbing.
Eric Shanfelt:
Yeah, it’s like driving down the freeway with your lights off.
Dennis Yu:
Yeah, I mean, I know plumbing is not sexy. I know there’s all these cool, red shiny things. You want to go look these cool new toys that came out, all these AI tools, start with the plumb. Eat your meat and vegetables first before you go to the dessert. Okay, just promise, just trust me on that one. I’ve seen a lot of fashion…
It’s also setting up your meta business manager account properly, your AI account properly, your LinkedIn ad account properly. It’s the audience…
It’s plumbing
Eric Shanfelt:
Properly…
Dennis Yu:
It’s all foundation
Eric Shanfelt:
Where you can take your email names and bring them into the ad systems to create matched audiences…
Dennis Yu:
All the custom audiences…
Eric Shanfelt:
It’s that integration and connection. It’s getting the leads back from lead gen campaign back into your email CRM marketing automation system. It’s all that plumbing that you and I you geek out on, but a lot of publishers just glaze over like, don’t get it. I don’t like it, it’s not sexy,
Dennis Yu:
But the publisher does understand and needs to understand. If I get all the data in one spot and I see where my visitors and emails—if I can actually see the path inside Google, if you’re plumbing is set up properly, you can see everything in Google Analytics and the data is accurate, right? Then whoever’s running ad campaigns or doing anything, you can hold them accountable. If you’re a publisher, you don’t know what’s actually working or not, or you’re paying all these other people for random things, or you have a team of people that they just send you a report every week that has these charts, whatever, and you don’t know what’s working from a business standpoint and it’s not in plain English … you actually have a plumbing problem.
Eric Shanfelt:
You do. Second key I think, is using the right campaign type and right campaign structure. I don’t want to go too deep in this, but I see very few publishers actually using lead ads where they can be very, very successful. Instead, they run a click generation campaign to go to a landing page, and then that’s okay, and that can work, but using the right campaign type for the right campaign objective set up the structure. You mentioned about the campaign, the ad set, the creatives and targeting different ad sets to different kinds of demographics or targeting to optimize that, right? You can touch on that if you want to real quickly.
Dennis Yu:
So you have different stages in your customer relationship, so you have different content that fits it. So imagine the dating thing, like you first meet them and you have a date, first date, and you go out multiple times and you kiss them. There’s this whole stage in building a deeper and deeper relationship if you have the right content that’s built along on this path as you nurture people. That’s what we’re talking about with right campaign structure, and so the goal selection too, when you choose, I want clicks, I want views, I want emails. This is the thing that everyone seems to misunderstand. When you choose that objective, that actually drives your targeting because when you say you can take the same piece of content and say, Meta, I want email subscribers, or I want video views, or I want whatever, think of this big pond that it’s phishing within, it’s going to know where the different fish are, the video fish or the link click fish. The guy who’s in charge of conversion optimization at Facebook, we had lunch and he said, my niece is 19 and she doesn’t have a credit card, so she can’t buy anything, but she keeps being shown all these ads from e-commerce companies that don’t understand that because they’re just trying to go for link clicks or website visits or whatever
Imagine back to the pond, you have one part where the salmon hangout or one part where the trout hangout or different kinds of fish hang out in different parts of this big pond. The objective you choose determines where Facebook is going to go seek on your behalf, and you get a completely different set of results, and to your point, if the content’s good enough and you choose link click, but what you really want on that is people to click yes, this is my name and phone number, you get a different result. So if you have great content, you choose the right objective and you pass the data through, that’s all you have to do to win. It’s not difficult. You don’t even need this complex campaign structure like we used to.
Eric Shanfelt:
Right? Right. Yeah, and frankly, letting the algorithm help you to find out which creative is going to resonate and then using that data to tune is awesome. I’m not going to dive too much in this, but finding the right hook for the audience. This is finding the right thing that’s going to resonate with your audience. Hey, maybe it’s preseason, boys rankins, or maybe it’s this particular story. You’ve got to do some understanding into what your audience is. This is what your editorial staff and your audience staff they’re going to know. These are the basic things, but look at your Google Analytics, see what resonates with your audience, but you got to find the right hook for your particular market that’s going to draw people in.
Dennis Yu:
And to draw people in, fine. Go into your analytics and see of people who are first time visitors, what did they come in on? What was the keyword that they searched, which you can see inside your Google search console. Your Facebook will show you when you log into the business manager and look into the analytics, what’s working, and so the hook is less about how do you capture their attention. It’s more about what’s that first stage in the journey? How do you tell the story deeper than the how do I distract people, which is what happens when most people think about hooks like, ahhh, doing this, but better is like, what is that pain that people have that, oh, I have that pain or this curiosity or this thing that I need to have this itch. What is that? That’s the deeper strategic view of this hook.
Eric Shanfelt:
And I’ll tell you right now, if you’re a city and regional magazine publisher, best restaurants, I’ll tell you bar none, that’ll be the best performing one you’ve got. Then it just comes down to execution of ad campaigns, which I don’t think we have a good amount of time to dive into—this might be an interesting chat for you and I to have another time, but when we execute on Meta Facebook and Instagram for instance, we don’t use every single stinking ad placement that they have. Frankly, we see the vast majority come through the feeds and not through all these other places, so we really kind of tune a lot of those down and then there’s just the execution of it, like you said, using the UTM parameters, executing on those things, so just the actual good execution. Curious, do you use the audience network much?
Dennis Yu:
Okay, so Facebook wrote three papers on this because there’s been this myth that has been very hard to kill the last 10 years on, oh, well, the desktop newsfeed is better than audience Network is better than WhatsApp or whatever placements. There’s like 30 placements. Here’s what’s going on. If you choose the right campaign objective, then the system will allocate to wherever the performance is. Okay? It doesn’t matter what you choose, unless you’re only spending a few dollars and there’s not enough data to find a winner, but if there’s enough data to find a winner and you can exit the learning phase, then you should just choose all placements. But yeah, it’s true. Audience network is lower quality.
Eric Shanfelt:
Audience network.
Dennis Yu:
You don’t need to exclude, this is a myth that’s gone on for 10 plus years.
Eric Shanfelt:
See, I’ve actually seen, and I’ll have to look up the data sources again, where audience network is one of the places that’s a little more rife with fraud, a little more click fraud on the audience network then…
Dennis Yu:
But if you choose the right objective platform
Eric Shanfelt:
Itself
Dennis Yu:
You’re choosing a real subscriber and you’re vetting for the quality and you’re passing back through conversions through the conversion API, and if you have all these protections in place and you choose the right objective, then that’s okay. We know that. Yeah, there’s a lot of fraud. Yes, you can exclude publishers just like you can on Google. Yes, there’s click fraud everywhere. It’s not just on audience network, but we’re trusting the system to do that for us, and we’re also taking three or four steps, like I mentioned, as part of digital plumbing and what-not to protect ourselves. The fact that people can argue click fraud is between five and 15% depending on who you’re talking to. I consider that just a tax. If the campaigns are performing clearly, they’re doing really, really well. If I’m paying 7 cents a click instead of six, but I’m really happy at six, I’m still going to be happy at seven.
Eric Shanfelt:
It’s kind of like inventory shrink at a retailer, right? You’re going to have some.
Dennis Yu:
Exactly. That’s the way to think of it.
Eric Shanfelt:
You’re going to have some stuff that’s going to be fraud, but you don’t want to throw the baby out with the bath water. You’re got a campaign that overall is performing well. You just know there’s a little bit some issues here and there, so what?
Dennis Yu:
Yeah, but it’s very easy to think, I’m going to exclude the entire thing because there’s some fraud there. One of the friends that we did ads for the United States Concealed Carry Association–don’t judge me because they’re a gun thing–eventhough I’m pretty good at shooting guns. We took them from 6 million a year to 65 million a year in revenue, largely off of Facebook ads, and they said thinking, you know what? Our avatar is the American male father, first amendment, middle class suburban, and this male, and his name is, I think Bobby is what we gave, what was his name, Tommy? I don’t know. It’s like Bobby and he’s 35 years old ,and he’s got a wife and two kids, and he lives in the south and he goes hunting on the weekends. They told this whole story and then they did targeting against just males who also have these conservative interests and what-not. That makes sense, right? That is the key persona. That is 85, 90% of their audience that will pay the $400 a year, but you know what? There’s women that shoot guns too. There’s younger and older people that also shoot guns. There’s all sorts of other people, so even though you do have a core avatar, your ICP or whatever you want to call it, don’t put those constraints on Facebook unless you’re targeting senior citizens and they have to be 62 years old to be able to be a member, right?
Eric Shanfelt:
Right, right.
Dennis Yu:
Just don’t do it. Then you’re putting Facebook in handcuffs for the same reason that you would say, well, I’m only going to use desktop or mobile newsfeed. You’re putting, the system can’t go and optimize fully on your behalf.
Eric Shanfelt:
That’s a good point, and I do see that a lot where media companies will try to over target, but this only works if you’ve got an evergreen thing that you’re running long enough to let it begin to optimize and come out of the learning phase, and it’s going to start optimizing the placements, the demographics, the right buckets of people in the right places…
Dennis Yu:
That’s the key…
Eric Shanfelt:
If we let it do it.
Dennis Yu:
But you have to have a winner first, so if it’s not a winner, then I, because a lot of people say, oh, well, the ads didn’t work for me on Meta because it’s like targeting the whole planet. That’s wrong on many fronts. If you have a winner and the data of conversion is being passed back, you can ask anyone who spends a lot of money and not just to people that spend 10 grand a day, but they are running targeted. This switch happened maybe six or seven years ago. All the people that do really well on Facebook, Meta, whatever, they run untargeted. There’s no need for us to specify the targeting. Our data and the creative itself specify the targeting. That is a huge shock for most people that are not performance marketers.
Eric Shanfelt:
Absolutely. Hey, we’ve got to wrap this up here, but I wanted to just throw out this…that we’ve been talking about Meta, which is Facebook, Instagram, WhatsApp, but there are other platforms out there. We’ve run campaigns like this onTwitter/X on Google ads, primarily in the search network on LinkedIn for B2B on Taboola. Haven’t done TikTok yet, but Dennis, I know you’re very hip on TikTok and you wrote the book on it.
Dennis Yu:
Well, Perry, Marshall and I have the bestselling book on Google ads, and the bestselling book on Facebook ads–sold 800,000 copies, and we have the bestselling book on social media on Amazon, on TikTok.
Eric Shanfelt:
We’ll put the links to those in this as well. Check it out, but I’m very interested to start running some ads, especially for some of my more consumer oriented publications on TikTok as well.
Dennis Yu:
No, it works for B2B as well. It works. It works for anything, so here’s the thing. If your goals, content, and targeting are in alignment, high quality content, passing through that clear conversion signal, the other things we talked about are true knowing what the content is at different stages of that journey. It doesn’t matter whether it’s YouTube or TikTok or Facebook, it’s all the same if the quality ingredients are there, things that we’ve been talking about here. It doesn’t matter what channel you’re on because you load that into the system. Now, the only exception is Twitter, which is X and LinkedIn because those two systems are so barbaric that they can’t optimize, and I’ve told them to their face multiple times. I was one of the first five advertisers on Twitter. I’ve told them that. They haven’t even fixed it, even since Elon took it over, they haven’t fixed it. I talked to somebody from Twitter last week about it, and I said, you still have your job there. No, actually, I was one of the new people who got hired after the 90% got fired. Their systems are so barbaric that they can’t optimize for you.
Eric Shanfelt:
I haven’t seen good results on X. I have seen some decent results on LinkedIn for some of…
Dennis Yu:
They’re great with the, here’s the thing,
Eric Shanfelt:
My B2B folks
Dennis Yu:
For X, you use to target when you have a particular handle, people who follow that particular organization, that industry, that manufacturer, that whoever, usually a person, an actual person is what performs the best. So because the system is so just unsophisticated, you use the fault or look-alike. On LinkedIn, similarly, you go by job title, the company they work at or some kind of company specific thing, and you have to take over the targeting on LinkedIn and X. You can’t just assume with Google and Facebook and TikTok—we’re spoiled lazy because the system just does the work for us. If we put the things in, it’s like a Keurig machine, put the pod in, press the button, it brew. Not even worried about it. It works every time. It does. If you put in the right ingredients and you press the button the right way. But with X and LinkedIn, it also works every time. But you have to do the targeting yourself. Yes, they have remarketing and all that, but it doesn’t work. You have to specify the targeting.
Eric Shanfelt:
A lot more manual on that. So it’s interesting you said that TikTok, you’ve seen also work for B2B. That’ll be interesting to try. I haven’t diagnosed mu
Oh, it’s fantastic. Very much. They literally copy…
Meta actually works very well for B2B to a lot of surprise for a lot of media I work with. They think, oh no, no, we’re a B2B, we’ve got to be on LinkedIn only. I’m like, no, no, no. You’ll get great results on Meta too.
Dennis Yu:
Meta and TikTok are fantastic for B2B. The more boring and plain and niche the industry is, the easier it is to win. If you set up your plumbing and have great content and choose the right objectives, like we talked about. So, the Byte Dance, which is the parent company of TikTok, full disclosure, they paid me a lot of money and I wrote a book and all this. They told me they’ll probably get mad if they hear this. So hopefully nobody from TikTok Byte Dance is listening. They told me they intentionally copied Facebook lead ads for TikTok lead ads. So you found success in getting emails from Facebook lead ads. You can copy that. It literally looks the same. There’s no extra thing to learn except for the business manager. A little bit different, but you copy it and it’ll perform better than on Facebook.
Eric Shanfelt:
Nice. Well, now you get me something excited to work on here, Dennis. So we’ll… .
And it’s not singing and dancing because remember we’re looking at it from the standpoint of performance, not trying to sing and dance and that kind of thing.
Exactly. Exactly. Well, hey Dennis, thank you so much for your time today. I know your time is in very high demand, but always enjoy chatting with you. I learned a few new things here today. I’m going to apply and hopefully y’all watching here learned a little bit from what we’ve been doing with media companies and from what Dennis has been doing as well. Make sure to check out his books. We’ll put the links to that in the description. Also, going to put the link to your dollar a day program, Dennis.
Dennis Yu:
Yeah, metrics.com/dad and any questions, happy to answer them. Hit me up on Facebook or whatever your favorite platform is. But the main thing is I want to see you as a media company adopt into your internal processes. Your in-house team, what Eric is teaching you here. These are fundamentals that will work five years from now. It’s not just some latest trick.
Eric Shanfelt:
Yeah, it’s been working for years and the specifics may change over time and specific execution may change, but the fundamentals are still going to remain the same.
Dennis Yu:
Yeah, I was doing this on Google Ads 20 years ago and now it’s the same principles that are working on TikTok. Exactly. Same strategy. Exactly.
Eric Shanfelt:
Yeah. Alright, my friend, thank you so much. I hope you have a fantastic day and gang, we will see you in the next session. Cheers.
