Drive and Convert

Ryan is excited to dive into the often overlooked remarketing options for re-engaging those prospects who don't convert on their first visit to your commerce website. Contact Logical Position for remarketing needs at https://www.logicalposition.com/contact. Reach out to Ryan Garrow on LinkedIn: https://www.linkedin.com/in/ryangarrow/ Transcript JON MACDONALD: Ryan, good to talk to you again. Today we're going to talk about remarketing and re-engaging your audience, how does that sound? RYAN GARROW: Man, nothing gets me more excited, John. [laughter] RYAN: Marketing is a huge piece of online marketing driving traffic, it's an often-overlooked piece of digital marketing. I'm excited to drive into some of these details with you and hopefully shed some light on the mystery that is remarketing for most companies. JON: I really want to talk about a few things today, what do you do when someone doesn't convert on your site? You spend a ton to get folks to a site but then when they don't convert on that first visit because let's face it, most visitors aren't going to convert on their first visit, how do you keep marketing to them and close the sale and get that conversion? As my understanding, and hopefully, you're going to school me on this today this is typically what is called remarketing and it can be extremely powerful when done right. I do know that, tell me a little bit about how you and the team at Logical Position define remarketing? RYAN: Man, nothing as a marketer can make you more frustrated than somebody not doing what you wanted them to do when they came to the site because he spent all this time and energy sculpting traffic, eliminating waste saying, "All right, they are searching for my exact product and service. They are ready to buy, they're ready to put a credit card in." To get you into the right category or the right product page and then conversion rates on websites dictate that for almost every company we work with over 90% of that traffic goes somewhere else to do something and not take an action, and it just is frustrating. Especially if you do some of those real-time heat map watching and watching people on your site behind the scenes, you just get frustrated like, "Why didn't you click that? What's wrong with you? You should have just gone and clicked add to cart and buy?" Remarketing ends up becoming the step in the process next. You almost need to look at remarketing as a bunch of different layers. It's not just one simple, "We're remarketing, we're good." There's search remarketing, there's remarketing through email, there's remarketing through display ads. There's so many different things you need to be doing and be aware of in the e-Commerce space to help bring those people back to the site and take the action you want. Each one of them needs to have a lens of what's the return, am I doing it properly, is it generating the type of return that I need it to be as its own entity? When you look at driving traffic through paid search, and I think one of our earlier podcasts we talked about all traffic is paid traffic. At this point, it is requiring some level of investment to get that traffic to your site. Whether that's just time, energy, money, thought, something's happening that you're putting out there to bring people in. Paid search, hopefully, there's a return that's making sense with paid search as its own entity, search shopping, Bing, Google, Yahoo, whatever that looks like for you hopefully there's a return that makes sense. An additional marketing piece needs to be re-marketing, and it needs to have a return that makes sense for the business and for the products and services that you're selling. Within that re-marketing entity, there's different layers within that that say, there's this type of re-marketing and this type of re-marketing and this type of re-marketing and each one of those has different expectations for return. It may be five years ago re-marketing was vanilla and now we have the Baskin Robbins if you're in the Oregon area and you know Baskin Robbins, there are 31 flavors. There's all these different things you can be doing with re-marketing that for some brands it's overkill, you don't have enough traffic to use all of these wonderful different things. Other brands are using just the most basic remarket and they should be using a really complex additional layers of re-marketing to help drive different types of traffic, different ways, with different expectations. Brands should be doing it and they should be doing it more than likely more complex than they are. I think on average, most companies are not utilizing all the things they can be doing, even just through the simple Google platform of re-marketing, there's a lot there. JON: I look at this as two sides. One is what is the data that you need to be tracking or where are the points where you can then have data to know who to remarket to and then on what channels can you be re-marketing? Maybe we break that down. What events or obviously, there's the simple page view. Somebody views a particular page like a product detail page, you can then start re-marketing that product to them. What are some other options there for how to get data to know who to remarket to? RYAN: For simplicity purposes right now let's just focus on the Google re-marketing platform. Most companies are at least familiar with it, most people probably understand conceptually what the Google re-marketing is inside the Google Ads platform. All re-marketing is dependent on the data you're putting into these lists that you're re-marketing to. If you've got really crappy lists, you're probably going to get crappy re-marketing results. Step one is understand how you are breaking up your data. I would say as a general rule, more granular is better because you can combine those audiences into bigger groups. If you have really granular data sets within your lists in audiences, wonderful, use those, make sure that are in there. You can always make bigger groups but if you don't have the small granular groups you can't get to them. At least set them up there whether you use them or not, at least get them in there. By granular groups I'm talking about you should have a list for shopping cart abandoners. When I say list, it's an audience within Google ads. Let's have one for shopping cart abandoners, let's have one for product viewers, people that have viewed a product page. Let's have one for people that viewed a category page and people that only view the homepage, what's on the homepage and left. Site depth would be a good way of looking at that, the deeper they go on the site, the more likely they are to convert through remarketing and the messaging is probably different. What we see when we do this, when we add these in, and we add the audiences in for not only display ad re-marketing, which is an important piece to follow people around appropriately, we'll talk about some of the details on what's appropriate and what isn't later, but also re-marketing lists for search ads. If people go back and there's a heavily researched product that, "Hey, I found this one, it looks like it's good, then let me go back to Google and do a couple more searches to make sure that I'm not leaving a lot on the table as far as options or price point," you can bid on those people differently based on where they went on your site now. What we see, generally speaking, you're going to have a higher return on re-marketing the further down into the site they went. For example, people that were a shopping cart abandoner that you're re-marketing to are probably going to have a higher return to re-marketing than the person that went just to the homepage and bounced, fits the logic generally too. JON: These are all intent signals, you're looking for high intent? RYAN: Yes, to a degree. If somebody comes through a shopping ad they're obviously going to land on a product page. That same searcher could also land through a text ad on a category page. The search is exactly the same, I'm looking for a purple widget. If I click a text ad, I should land on a category page for purple widgets but I could also have clicked a shopping ad and landed on a product page. Same search, same intent that we saw on Google went to different spots on the site, but if I landed them on a category page and then they went and clicked on a product page we're probably getting the same level of intent on a purchaser as word there. If I clicked on the product image with a price point off of Google to your site, probably likely that they're looking to purchase one rather than I'm clicking on a text ad because I'm going to go to a category and do some more research. Then if I bounce re-marketing maybe a little bit different than someone that took an extra step. JON: That covers what you should be looking at to form that audience as you call it. It's not a list because you don't know who's part of that audience specifically, you can't get the individuals. Google masks that and tells you generally how many people are in that audience, is that correct? RYAN: Yes. Then each audience will tell you how many people you can re-market to on the display network and YouTube versus on a remarketing list for search ad. Because then you might have a 100,000 people in that audience but maybe only 30,000 of them you can remark it to as a remarketing list for search ads type thing. There's always going to be different numbers around those because of how Google is collecting data and allowed to present information around that as a re-marketing list. If you have medical devices of some sort or you're treating some medical problems, you may not be able to remarket at all on Google because that could potentially be personally identifiable information. If you share a computer with somebody and you don't want them to know that you may be feeling sick, you may have these symptoms or you may need-- I'm older than my wife so maybe if I need a knee brace, maybe I don't want her to know that I'm getting old and fragile. [laughter] RYAN: Google won't let people remarket it to me cause my wife might see that. That type of thing is a limitation within remarketing based on some products. JON: We both know we're getting old and fragile so we'll leave it at that. In terms of the platforms of which you can then take that audience and market to them, I heard you say Search terms, search ads, I heard you say YouTube, obviously display ads. What other mediums can be used for remarketing? RYAN: YouTube is a big one that I think most companies overlook, and there's this general opinion or thought around YouTube being complicated or difficult because you have to create a video and there's that barrier to entry there. Creating a remarketing video is really not difficult, it can even be very basic, it doesn't have to involve human or video, it could just be static images with voiceovers or image textovers. There's a lot of things you can do to make a pretty basic video on YouTube that can still be effective but the cost of YouTube is so low that your reach can go so far. What I tell people about YouTube is you do remarketing true view, which means they have to watch a certain amount of your video to be counted as a view that you actually pay for so they skip it. Everybody's seen those videos on YouTube that you skip before watching the cat video your mom sent you that you have to watch because she's going to bring it up when you're at dinner tonight. If you skip it, they don't pay anything. There's still some forced branding though because you have to watch five seconds but if you watch it, and you can do all kinds of overlays and allow people to click and go right to the site and the product that they were viewing, but if you watch it, it usually costs $0.10 to $0.12 for that view. The average click on the display network can sometimes be up over $1, $1.50, and so lots of opportunity on Youtube that people don't pay attention to as much. YouTube, email, you can remarket through email. In certain areas, you can remarket on the display network through dynamic ads or static ads that show the product you're looking at, and you can also do a list that allows you to bid differently on them based on their behavior of your website in the past. JON: That sounds interesting, it can get really complicated then. Of those, what do you think is the most powerful or has the highest ROAS, return on ad spend. RYAN: That is really going to vary, it's my most often response but my least favorite response, is it just depends. Every online company's probably going to have slightly different numbers based on their products, based on their website, and their conversion rates. Generally speaking, I really like the remarketing list for search ads, especially if you have a research product. People are going through the process are going to do a lot of searches before they buy something, finding out if that person is valuable or not to be bidding on again, and so many companies don't look at this. You can actually take this list of all the people that have visited your site before, and let's just say you sell coffee cups-- I'm staring at one in front of me. You sell coffee cups and this person searched for white coffee cup, came to your site, didn't buy, that's a list I can put in there. If they go back to Google again and they search for gray coffee cup or red coffee cup-- you don't even have to change your bids in there and you can see them because normally you'd be bidding on them again anyway because they're searching for another coffee cup, which you sell. You can see, "The people that have been to my site that searched this term either perform better or worse than the average person searching for this product off of Google." If it does better, you catch them again in the research process, and they perform better you bid up on them, be more aggressive because the return is higher, you have more you can be spending on them. If it's worse and once they've been to your site and they leave they're not going to buy on their second visit, then you just save the money, don't bid on them at all. You can actually spend that money on acquiring a new person to the site rather than somebody that's already been there and you know is not going to buy based on the data you see. There's a lot of levers that Google is now giving us as marketers that really can get creepy to a degree, but also really gives us actual leverage to push more return when it's necessary and then pull back when it's not. I think that those search ad remarketing lists can be super powerful. You can even do your past customers. Google does have an email match, you can upload your past customers into Google and create an audience around them and say, these 100,000 people have bought from me over the past decade. I upload them to Google Match generally, it's about 50%. Let's just say you have 50,000 people in an audience, you can now bid on them differently based on their search. If you sell mattresses, they're going to buy a mattress one time every 10 years on average, even though we're supposed to buy it more often than. That list of people, if they're going off and you know that they've bought in the past and they're searching again for another product you sell, maybe you can bid more aggressively on them. Because now you sell pillows, for example, and they're searching for a pillow and are like, "I really did like that mattress and they also sell pillows. That's great, I had a great experience on the mattress, now I can buy pillows." You can bid more aggressively on them even though it's a general search on pillows that maybe you wouldn't spend a lot of money on just by itself, but knowing something about that customer gives you the ability to be more aggressive than your competitors. It's a phenomenal subset of the remarketing world because you have some information about them. JON: This sounds like its own specialty. Just running ads is a specialty that you have to know a lot about, but even just going into the remarketing this is way deeper than I would have thought. RYAN: Man, we have an omnichannel strategy group now internally at Logical Position and the reason they came about largely is because of all the demands of our clients trying to figure out how to get all of the use out of remarking that they could. They were having so many of these conversations with our clients saying, we need to be doing this, we need to be doing this and the strategists having to work with all of our teams like, "Hey, we just need a department that can do this." So we've scaled internally, we've got a whole group that handles strategy across all of our accounts and a lot of it is around remarketing. JON: That's great. RYAN: Just the things you can do that are-- again, I say it's creepy [chuckles] but really fantastic as far as the marketing person's concerned on the things you can do with these levers that Google gives you. JON: On the flip side of creepy, it could be helpful because you're reminding folks of where they were, what they were doing. In terms of conversions, we see at The Good that you only have a certain amount of time to convert someone on your site and that's a couple of minutes typically if that. If they get distracted they end up onto something else. It's often even just personally I forget what I was looking to do and then I see a remarket and I'm like, "I really needed that." Then I end up going back and completing it. It seems to me, I'm hearing, that there's a ton of data that can be collected to ensure proper remarketing, and you just listed off a ton of them. It gets real intricate pretty fast. JON: Folks usually default to thinking about remarketing only on Google, what other kind of tools do you think could be used for remarketing? I know there's other providers out there, there's other toolsets that aren't Google, display ad networks, and things of that sort. Am I right in that thinking? RYAN: You are. Google is probably one of the most well-known just because if you're spending money on ads it just becomes very easy to see there's a section for remarking on display ads in Google. Everybody tries to claim that they have the largest display network on the market. When you look at remarketing a lot of it is done on the display network with display ads. As you're reading an article on a new site or something you're going to see images that are ads all over as you're scrolling down. We've been doing that for a while, most people know about it, most sites, not all, but most sites aren't going to say no to a display network that wants to put some ads on their site. A lot of these ad buys are done behind the scenes on a secondary market that most people don't even-- it's like the matrix, most people don't even know it's there but all of these ad buys are happening in this big secondary market that's an ad desk that all these networks plug into and they all compete in real-time to whose ad's going to be where. It's crazy when you start looking under the hood of the display network world. From a remarket perspective most companies are aware of Google, not a bad place to start, it covers almost everything. There's a couple of other competitors in the display space of note that we come across most often, it's Criteo and AdRoll. Both of them have some good things and some drawbacks and they have some exclusive sites that may be Google doesn't have access to. By and large, most companies should at least start with Google because it's already there with every context that gives you an idea of how remarketing is going to work for you. AdRoll and Criteo can do arguably better ads. Google's use of the HTML5 stuff now can make pretty darn good ads if you have a good creative team but from what I've seen, you do go into buckets within Criteo and AdRoll. They say, here's the return you need, this is your bucket. Here's what we're going to do for you. Done. It goes and it just does a lot of work. You don't have a lot of extra leverage to push and pull within their platform. It's kind of a, "We do it for you", scenario and it can be good. The only thing I caution people to look at when they're at Criteo or AdRoll is generally those companies out of the box are taking full credit for a sale through remarketing for an impression. If I'm scrolling through and reading a website and I happen to see a remarketing ad from a company who I was on, I didn't click on it, I just kept scrolling and finished the article and then later that day I went and bought direct. I just typed the website in, went and bought and finished my transaction, generally, the pixel for AdRoll for Criteo will say, "Hey, Ryan went and bought and we did it. Our impression generated it and we claim 100% credit for that." Their report will generally say, and most marketing platforms take a lot of credit so you just have to understand how to look and filter that data to get a real number. Because you can see in Google Analytics the click data, "AdRoll and Criteo they had 50 clicks and this much revenue." Not hard but trying to assign a value to that impression becomes complicated and usually requires conversation. All that to say AdRoll, Criteo for the display products out there for remarketing along with Google. Yahoo in the past has done a lot through remarketing but they've shrunk down and they're not as impactful so generally, I would say for display, those three. Video remarketing generally lives on YouTube, there's not a lot of video re-marketers that compete yet with YouTube and the scope and size of them. Email remarketing has become a bigger and bigger player in the space. Most people that come to your site, you can get an ad in front of them through a display network. Regardless of what you choose, you can probably find a website that they're going to be visiting and you can get an ad on. If they have an ad blocker on you're probably not going to get that cookie, you're not going to be able to remarket to them. There's been recently over the last couple years, a couple of companies that have come up with the ability to send emails based on the pixels. They have a list of-- I think the latest I heard is somewhere around 300 emails, they have permission to email these people. These people have opted in and said, you can email me and they've been able to combine that with a pixel or a cookie saying, "Hey, they went on your website, they didn't buy. We can filter that against converters. Let's send them an email and say we'll give me 10% off if they want to buy." They email themselves on behalf of one of their partners, which would be you. It's actually been gaining some good traction and it's another opportunity to get in front of people that you couldn't get in front of before possibly or in a way that they're actually going to pay attention to. It is a little bit promo heavy so you might give a 10% off discount coupon code, but for many of these companies, they're willing to do it for free for a rev share. It's just an add on, "Hey, if we're going to get a 10x with a 10% off, we're going to get a 5x at the end of the day, that's not terrible." JON: I'm pretty sure this happened to me the other day because I was on a site, I didn't give them any information and I left. Then I got an email from them about an hour later just like, "Hey, noticed you were on our site and didn't buy, here's a discount if you want to come back." I was like, "How did they get my email?" That's pretty interesting. RYAN: If you looked at the bottom in the fine details of that email, you'll probably see that it came from a company you've never heard of on behalf of that company. JON: I see. RYAN: I think if you haven't tested it and you've got enough traffic, and enough traffic would be you've got 30,000, 40,000, 50,000 visitors a month to start generating a real number through remarketing, I would at least consider it at least testing it. That's my general rule with most marketing test and measure. If it's working do it better or more, if it's not working stop it and then maybe come back and revisit it later once you understand some of the things that may have caused it to not work well. As we live, especially as Americans in a very busy, complex society, our attention is being drawn all over the place. You and I both have kids, John, we're not sitting down in the evening doing anything without an interruption, we're not alone. Understanding that as our day - as I go between three screens and I've got phone calls and emails and SMS and text messages coming on my phone that's on the other part of my desk, I'm going to get interrupted in the process of purchasing something more than likely. Remarketing, I believe, is going to have a bigger and bigger place in the overall online marketing scheme over the next few years. We're not going to get less distracted as Americans, I don't think. Understanding the appropriate way to get in front of people in the way they want to be communicated with. Text remarketing is becoming a bigger, bigger thing, especially with millennials that want to be communicated over on their phones almost exclusively. It's fun but complex. JON: I was going to ask you SMS like Facebook or Twitter or Instagram, all those are platforms you can do remarketing on too, is it not? RYAN: You can. Social remarketing has been a big one. In a lot of companies, I suggest that that's the way they test social marketing. If you've not done ad buys through Facebook and Instagram yet, an easy way to test it and the validity and what return you can get is start by remarketing on them. Again, we'll go by my coffee cup example, if you're selling coffee cups and you're going to remarket them on Facebook, these people have been to your site, they've seen your product, they know who you are, they haven't bought yet. If you put an ad in front of them on Facebook and Instagram to try to draw them back and that ad does not work at a profitable level or rate that makes sense, the chances of going out and finding a broad swath of customers that have never heard of you before to buy at a rate that makes sense for you is much lower. If remarketing works, that tells me that there could be some potential to draw in or prospect new customers off of social channels through ads. There's always some ways you can do without ads, but at least through ads. It's a good way to test it all and see if that can make sense for your brand. JON: Okay. This has been really educational for me, I appreciate that. I'm now wondering what budget should be reserved for remarketing versus driving initial traffic? Is it maybe a percentage that someone should be setting aside of their total budget spend on traffic? What do you typically see there or recommend? I know it varies but is there a general rule of thumb? People should be at least trying a certain percentage of their overall budget towards remarketing? RYAN: It's my wonderful it depends answer again, unfortunately, but the bigger and more complex and the more research that is done to make a purchase of your product or service, the more you're probably going to spend on remarketing. You want to be staying top of mind. If you're selling $10 coffee cups, you're probably not going to have a huge portion of your budget on remarketing. There's probably going to be somewhere between 5% and 15%. Again, there's going to be outliers on that. 100% of the time, there's always going to be outliers. Step one is just start doing it if you're not doing it and then understand that you need to have frequency caps so that you're not blasting somebody and spending $20 remarketing your $10 coffee cup in one day because you won't stop being in front of them. Understand that you're going to test and measure your promotions, sometimes you're going to need promotions to get people back to the site to buy something. Sometimes you won't, it's just timely like, "You forgot that something was in your shopping cart," or, "I was looking at that red shoe." Also, when you cross apps, bring apps into the space, that's another big one. I tend to be a sneakerhead and so I buy stuff on Nike's app and the remarketing potential within that space for them is even more creepy and awesome. Because they have all the first-party data around what am I doing on their app and on their website and what type of products do they want to prospect me with, "Hey, you've never seen this shoe. You should think about it." I'm a huge fan of Jordan 4s so anytime there's a new Jordan 4 release coming up they know they need to let me know about that before. Even though I've never seen the product they are remarketing to me with a brand new product because they have that data of what I've purchased in the past, so much opportunity in the space that you probably need to be a little bit uncomfortable with your remarketing budget. I love saying that to people because they're like, "I'm going to dedicate 10% to remarketing." That's good but if you see it working don't be afraid to go to 20%. Why not? If it's there and you can put money in front of it and you're getting a return that makes sense for your brand there's no reason not to do it. JON: It's an awesome way to just attack the people who are already interested. They're obviously already interested so get your ad back in front of them because they were already on your site. That's a great way to attack that marketplace that is there for more conversions getting back to your site, for sure. What's the biggest common mistake you see around remarketing? RYAN: Oversimplification. I think we've got the data, we've got all the levers we can push and pull to get the appropriate message in front of the appropriate person at the appropriate time. Often I see just blanket remarketing as the name of their campaign in [unintelligible 00:28:59] one campaign, all visitors in there, they have no frequency cap then it's like, "We spend $1,000 a day on our general marketing and we spend $100 a day on remarketing, done," and they just let it run. They just think that it is what it is, like we randomly get a sale, we don’t… It is probably necessary to build multiple campaigns for multiple audiences and change the messaging throughout the time to K model. Day one you have one message for them when they didn't buy. When they don't buy. If they don't buy you move them to your three-day remarketing lists with a messaging change or the ad changes, you're staying fresh. If they don't buy then they move to their seven-day remarketing list. You're always moving converters off of these lists as they go down, you can store pixel on somebody's site for 540 days. We have a long term pixel list that we use for friends and family events or discounts online or when they want to run a big promotion. We can get more aggressive with those people that have been to the site in the past when they're searching for a broad non-brand term or something. Not getting complex enough is one of the biggest issues most companies are having, there's a lot more leverage you can put into your re-marketing starting just with all that data that you're tracking. Having the granular list that allow you to do some of that really cool/ creepy re-marketing to make your company more impactful in this digital marketing, is traffic generation. JON: Ryan, this has been a super educational experience for me, I really appreciate you sharing your experience on remarketing and re-engaging your audience. I know this is something that you have that whole I'm Your Channel team you mentioned at Logical Position that handle this. People wanting to get a hold of you, ask you any additional questions, we'll put into the show notes how they can email you directly. It's been an amazing experience, I look forward to our next topic coming up here next week. RYAN: Thank you so much, John, I appreciate it. Any time I can get into the weeds on helping clients get more success from all of their marketing in general, I'm in for it. Having such conversations individually, with anybody listening, I'd love to because it's just fun for me. JON: All right. Thanks, Ryan. Talk to you soon. RYAN: Thank you.