Inbound Marketing & Sales Development Inspiration

5 Strategies to Improve Display Advertising Results [FB Live]

Posted May 3, 2017
12 minute read

Digital ads are the oldest digital channel, but haven't always been very effective. In the last episode of WebStrategies Live, we covered targeting strategies that can dramatically improve digital advertising results.

We covered five specific targeting strategies to reach your target market at the right time with the right message. From reaching audiences similar to your current customers to people who have visited your website or who are looking for a product or service similar to yours, precision targeting allows you to effectively get your message in front of those most likely do business with you. 

Watch the broadcast below for specific digital ad targeting strategies you can implement using Google, Facebook and more.   

If you have additional questions, feel free to reach out to me at




Chris: Hello everyone and welcome to the eighth episode of Web Strategies Live, the video broadcast for the modern marketer and business owner where we share with you how to generate leads and sales online. 
Chris: I am Chris Leone.
Phil: I'm Phil Woods.
Chris: And we are broadcasting live from Web Strategies headquarters just outside Richmond, Virginia. We've got a great topic to share with you today. We want to talk about how you can achieve laser precision targeting with your display advertising. So why are we talking about display advertising? Well, it is the most invested in, probably one of the most invested in digital channels, right? It is the oldest digital channel. And it's not always been the most effective. In fact, it was usually the least effective, but that's starting to change, and we want to share with you some tips on how we can do that.

Chris: So we're going to talk about five different ways that you can target people online with display advertising. Typically with display advertising, if you haven't done this a lot as a practitioner, you may think there's only really one way of serving up ads on websites or social media or whatever to hit your target market. But it gets a lot more precise, a lot more sophisticated than that. We're going to talk about five ways to do this, matching display ads based on behavior, using email addresses to target people that you already know, using email addresses to target people that you don't yet know, which is a really powerful and interesting way to do this, matching ads based on someone's interest and behavior, and then matching ads based on search behavior.

Phil: One of the most common ways is to target people based on their onsite behavior. So, for example, if we have someone who has been to a specific page of our site, we can easily serve them ads again. So, let's say they visit our pricing page. That's one of our most important pages of our site. We can kind of collect the data from all the users who visit that page, and essentially put them on a list and then serve them advertising out on the web no matter where they are. Generally, the ad networks are going to call that re-marketing or re-targeting. Those two phrases tend to get used interchangeably but essentially that just means you are serving ads to people who have already been to your site. And I think one of the biggest mistakes people commonly make is just collecting all of your site visitor data, putting them all on one list, and serving them all one ad. If you want to take one point away from this, that would be to break up your campaigns and serve different users, different ads based on different activities from your site. So those who visit your products page or a pricing page could get one ad. Those who visit or add something to their shopping cart could get a different ad. Those who visit a contact page could get a different ad. I think if you end up doing it that way, you're really putting yourself in a state of mind where you're thinking about who you're targeting, and what it is they're really looking for, and you can create an ad that might resonate with them.
Chris: We've created two very generic personas that we're targeting here. One is James. The only thing you really need to know about James is that he's a current customer, somebody that we already know. Bill is somebody who looks just like James, makes similar money, similar, you know, life situation, but he is not currently a customer of ours. And what we're going to talk about with this target is how to reach both the James and the Bill's of the world, right? So, the example that Phil just went through is how to reach James, somebody that you know who's on your site or at least somebody who has already made some kind of connection with you just by the nature of visiting your site. And that's how we reach someone like James. 
So let's hit the second way that we can reach people with display advertising online, and that is using email addresses to match ads to users. And this is how that works. So here's an example. Let's say you're trying to get ads in front of people who are currently customers, former customers, or old prospects. It could be any one of these or even more. The point is that you have their email address, okay? You could be any business that is trying to sell more to an existing customer base, or you have, maybe at least a few 100 emails of people who are not actively buying from you, and you want to find a way to kind of stir that pot and bring people back in. So this is what you do. You take an email list, you export that from whatever system that you're using, and then you upload that into a Google or Facebook. In this case Facebook. We take that email list, we upload it to Facebook, and if Facebook can match the email addresses that you've provided to the email addresses people use to sign into Facebook, it can deliver ads to those individuals. All right. Now, because Instagram is part of the Facebook family, you can also serve ads on Instagram as long as they're part of that customer list. And then finally, Google offers this capability, too. So if you upload it to Google, you can have it display on Google properties or any website that sells Google advertising which is a very large percentage of the web. Chances are if you see display ads online, they're usually being served up by Google.
So a really effective way to hyper-target that advertising to people that you already know somehow. The only condition here that's important is that the email address that you're using is also the email address that they use for personal use. And we see that that success rate is a lot higher for B2C than it is for B2B because in B2B you're typically giving like a work email or something like that. With B2C, we usually see a match rate of at least 50% if not higher. So if you upload a list of 2,000 people, there will be probably at least a 1000 people who you can then serve ads to. The only thing that I would caution here is take advantage of the fact that you know who these people are and their relationship with you and make sure the ads match that situation, and you're not just running generic ads. But if you're uploading people who are current customers, cross-sell. If you're uploading names of people who used to be a customer but canceled service, create some kind of come back and join us offer or something to that effect.
Chris: Third point, use existing customer demos to match ads to new potential customers. And this is pretty much just like the last example. But we take it a step further. Let's give a real world example here. Your target market is someone who might be interested in buying a home security system, and you are an alarm and security company, okay? And this is what we do. You know, we take our email addresses of current customers, right? The customer we already know which is that first persona. And then we can upload it to a system, and it will kind of reverse-engineer and figure out who that individual is and what they look like. And again, we're doing this at large scale, like thousands, maybe, of people. And these social profiles or social networks, I should say, or Google can start to paint a picture of who these people are, and what they're about, and what their internet habits are, and all that kind of stuff. And then we can layer on additional criteria, like let's say we upload our email list, and we say, "I want people who look like this but are in this region, and then maybe also have these other interests." So we start to build criteria on this other list of people who are already customers with us because we know that people who buy from us might look like someone who might buy from us in the future, okay? So we upload that email list to a Google or Facebook, it runs its algorithm, matches an additional criteria and next thing you know, we are serving ads on Google properties, on Facebook, and on Instagram. So, very, very powerful way to match people who look like your current customer base. 
Fourth way that we can reach people, matching display ads to someone's interest and online behavior. This is how a lot of display advertising used to work either automatically or we were trying to infer this, right? So if we were selling golf clubs, we would go out if it was the '90s, we'd go out to any website that talked about golf and try to buy ad space, you know, on that website. Then display advertising started to get a little bit better. And we could say, "Just go out and find these sites for me. I don't want to look for them individually. And if you find a website that talks about golf, put my put my ad on there." But now we can take it a step further. So let's say our target market is someone who is actively shopping for a car, right? And we could be a car dealership, we could be a bank or credit union, you know, who offers car loans, or we could even be a website that sells car accessories. So what's Matt's behavior? What is this persona's behavior if he's in the market for a car? Well, he's going to go out, and he's going to use Google. He's going to use other resource sites. You know, how to buy a car, what to look for when buying a car, things to know when buying a car, all that type of behavior. And because Google and Bing typically are tracking behavior across those sites, they might have like a pixel uploaded, you know, or something like that. It starts to track all that information and understand what's going on, and it paints his picture of, "Hey, Matt based on visiting all these sites, Matt, I think, is looking for a car." Right? He's done Auto Trader. He did a search in Google saying he was looking for a car. All these other things go down the line. And so then what we can do, is we can start serving ads based on that matched behavior.
So what we're actually showing here is the screen that you can set up in Facebook for detailed targeting for your ads. This is an actual window that opens up in Facebook when you're building ads, right? So we went to the automotive segment, and we did a drop down. And you can see here, there are all these different types of car shopping behaviors that we can start to segment on, motorcycles, new vehicle buyers, people who just bought new vehicles, people who are shopping for new vehicles, people who own these types of vehicles. We can even go a step further in Facebook and say, "If you just purchased a crossover or a full-size SUV or compact car or whatever." And you can select and build audiences based on this behavior, right? So it's pulling in this data from both the user activity across the web and other third-party sites that partner with Facebook and Google to send this information back and forth, and then we can create super hyper-targeted ads. So in this case, this persona might go then off to sites on Google or go to Instagram and Facebook, and because we know based on his behavior across the web that he's looking for a car, we can have our car ads show up to him.
And now the fifth way we want to talk about showing really hyper-targeted display ads is based on search behavior, so same persona, same situation. This is someone that we don't know who is actively shopping for a car, and we want to try and get in front of them. So he might go to Bing, and he might type tips for buying a new car. We can then, based on that activity say, "Anybody who types something like that into a search engine, I want you to serve ads to that individual." So obviously, this doesn't just go for car shopping. This goes for, really, any industry if, you're trying to travel to, traveling to San Francisco, and you're doing rental cars or Airbnb or a hotel or anything like that. You can start serving ads because based on their search behavior you understand what they're trying to do, right? So there's just millions of different ways you could do this no matter what your industry is. And we found that one to be also very effective, although the downside to that, as far as I know, is that the third-party ad servers who use this technology, I don't believe work with Google. So you have to base it off of Bing and some other smaller search engines. So it's certainly not as, you know developed.
Chris: Not as developed, not as comprehensive as you generally like to see. But certainly, even if it's a very small group, if they're doing those searches and you can get in front of them, you know it's going to be hyper-targeted.

So the five ways that we can hyper-target people online with display advertising. Phil went over the first one, just basic kind of traditional re-marketing which is based on behavior on your website. I definitely recommend starting there if you're not already doing this. As Phil said, don't just throw one generic remarketing ad up for everyone who's on your site but get a little creative with it, maybe pick out three to five scenarios about someone who logged in, someone who looked at pricing page, someone who looked at this other page, people who have come back to the site multiple times. Create a few scenarios and then build very specific ads to match that behavior. Matching display ads using email address. If you have those email addresses, that can be just tremendously powerful, right? You can also use those emails to find people who look like your current customers. That's point number three.
Method number four, matching display ads as to someone's interest and online behavior. These platforms are so good today and so developed. And they're now partnering with so many other data providers that you can import this type of data, and then finally you can target ads based on search behavior. So, we just burned through a lot of stuff pretty quickly. I want to open it up to Q&A, ask what questions you guys have about how to try any of those techniques, any situations that you're in currently and you're trying to solve for. 

Chris: Okay, so here's a question I had a couple people email me before. How do you measure performance for display advertising? So, display advertising isn't always direct response. They're not always going to click. In fact, most of the time they won't click that ad directly and come back to the site. But what we do know, and we've seen this, and there's a lot of other reports that consistently find, you know, the same stuff and draw the same conclusions, is that when you run remarketing, just remarketing which is the first method that Phil talked about, you see increases in brand searches, you see increases in conversion rates, increases in clicks on ads, both branded and non-branded ads. 

They could be really effective even if people aren't clicking directly. So I would be careful about how you gauge success on these because if you just look at click-through to purchase or click-through to conversion, you might be disappointed. In fact, you almost certainly will be disappointed. But when this has been isolated at very large-scale and tested, they'll typically see lifts in a lot of other metrics that this is running. And it makes sense, right? Because if you're in a shopping mode and you're going out and you're looking at stuff and you're not ready to buy right now, but you can get just that little tickle, that little reminder over the next week or two that you, hey, you were looking at this thing or you're looking at this pair of shoes or you were considering calling this company about kind of partnering with you on something that you're doing, it's just a great reminder, right? Otherwise, they just might forget. So I would be careful about trying to measure it on direct-response because I think you might be disappointed. But there's a lot of data out there that says this is super effective.
Phil: About the Facebook pixel and custom conversions, what else can we do with that data? There's certainly a lot of segmenting that could possibly be done on the re-marketing front. So I would explore that first and foremost. Beyond that, it partially depends on the volume. But let's kind of assume that we're getting some volume to do something with. There could be some opportunities, kind of explore some niche targeting. We've had some pretty good luck with that. So if you can identify through the custom insights, interests, and behaviors that maybe don't directly align with your business, but it's a common thread among your audience, then you can go out and target that specifically and that could present some opportunities.  

Topics Digital advertising, Display Advertising, Digital Marketing

Agree, disagree, or just have something to add?

Leave a comment below.