Closing the Loop on ROI: Paid Marketing + CRO

Gary Marx:

Thank you. Good morning.

D:

Sure. Before starting with the actual discussion, I want to let our attendees know that you too can participate in this discussion. Go-to-webinar does not allow me to switch on your cameras, but I can switch on your mics. If you want to share your thoughts on the questions being discussed, do let me know. Send me a request using the chat or the question box from the control panel and I will be happy to unmute you. Gary, please take it away.

GM:

Awesome. Good morning, and welcome to closing the loop on the ROI. We’ll be talking about paid marketing in combination and coordination with CRO today and how that best drives, a higher return on the ad spend, return on investment, whichever way that you like to abbreviate and talk about your return on the money that you spent here. 

My name is Gary Marx. I’m the Senior Director of Conversion Rate Optimization (CRO) at Roundbarn Labs. I’ve been in the optimization space for over a decade now and loved it. It’s been a lot of fun. I’ve run tests on 7,000,000 visitors per hour on sites like CNN, down to not even maybe tens of thousands or just single-digit thousand on some B2B sites or startups, on Shopify e-commerce stores, and a lot of things in between. 

So we’ll talk about how I see this working at Roundbarn Labs, I’ll talk to you through two different sections, really. We’ll go over a contrast on paid-for CRO, what the similarities and differences are, how the user journeys continue from one to the other and overlap, and some segmentation that can come out of paid. 

And then I’ll do a separate section, which is really how you can take action on this. So I’ll give you some points on how we do it here that we found really drive collaboration and success across the programs, and give you five tests to walk away with that you can implement tomorrow or today, depending on how fast you can get your designers and your developers up. 

So first, I wanted to go over a little bit of who Roundbarn Labs is. We’re a high-touch performance marketing agency. We focus on full growth and ROAS. So that is very, very key with pairing paid with CRO to focus on the ROAS there. We’ve worked for some house name brands, such as Amazon, Next Door Access, Iflus Quest, and others, and that drives some of the experience today.

As always, we’ve gotta have some social proof our customers love us. And lastly, on this little discussion about what my day job is, we really believe in growth loops, and conversion and optimization pairs with affiliate, and affiliate pairs and supports paid marketing, and paid marketing then goes and supports conversion optimization. The big thing here is not to not to keep things in silos, but to break down the walls in order for things to succeed and become more than the sum of their parts. 

And in the way we talk about it, it’s kinda like adding chocolate and peanut butter together. Two unlikely things that no one in their right mind would have combined before but they’re fantastic on their own. And when you put them together, they’re more than the sum of the parts.

And so for today’s conversation, somebody might be thinking, well, I don’t like Reese’s, that’s fine. It might not be for you, and the combination of paid & CRO might not be for your organization. And others would be more like, well, do I eat around the rim? Do I take a big bite? Do I eat out the center first and eat the edge second? And that depends on the culture of your company. And happy to talk to you further about that. but here we’re gonna focus more on the results and how you can get started. You’ll have to figure out the way that you like to eat it afterward. 

Alright. So let’s dive in and talk about the differences and similarities on paid vs CRO. We’re first gonna start at the high level when we’re setting expectations for our customers. Our paid team will present something talking about the ROAS that they believe they can return. 

Here’s an example, on half a million in ad spend a little over, driving a 5 and a half ROAS. So they’re just under $4,000,000 in revenue driven from that half a million in assets of $650,000. So for every dollar they put in here, they expect to get $5 a half back. $5.60.

This is different from how we talk in the land of conversion rate optimization, because we’re talking about the lift and how if we can increase the conversion rate, what does the revenue do? And there are ranges for those revenues. Here, it’s very tied directly to a dollar in, a dollar out, which is some of the difference there. And as you see on the left in pay, this is the language that my pay team uses on a daily basis. They talk about ROAS. They talk about click-through rates. And they have their partners in CRO Land. The bullets are even on the right. So ROAS, they talk about just as much as we talk about conversion rate. They talk about click-through rates, just as much as we talk about lift. Impressions on their end just as much as we talk about unique visitors. And there you start seeing some of the differences, but some similarities. Creative, we talk about images. But the impressions for unique visitors is a different way of measuring and actually a different metric altogether. and we have to take their metrics, and we have to work with them for the way that we need our metrics to also work.

The wonderful thing about this is there is a huge overlap They’re both systems for gathering learnings and getting closer to the customer to understand what drives them and what can convert them. So we’re both focused on adding revenue. And the big thing is when CRO is supporting paid, we’re focused on the cost of acquisition and driving that down, especially because we understand what that cost of acquisition is. 

So you guys are all probably familiar with ICE frameworks – impact, confidence, and effort. Second, three columns after the channel’s one, and then we add cost and score. This is another thing that our paid team will do when they’re talking to customers. Whereas, with CRO, you’re probably used to seeing these for tests and different test ideas and test pages. I wanted to share what this looks like for the different channels that are paid team is working across. 

As you see, TikTok is on the bottom here, or second to bottom, and it’s one of the most popular, if not the most popular places to go and spend money right now. In fact, I was at a conference last week, and they were saying, if you’re not on TikTok, you gotta get on TikTok. It’s you’re losing out. There’s so much going on there. It’s the place of the future because of the way content is now digested, and also film in a vertical format, first and foremost. So those are some fascinating things, but I wanna encourage you to take a look at what you’re doing today in relation to where to get started on CRO.

If you have your bulk of paid marketing and dollars in Google, Display, or Facebook, focus on that first. Don’t try to spin up a new channel because it takes so much effort just to spin that channel up and get it right. If you’re trying to optimize and set up a new set of frameworks on top of that, then you’re adding more challenges to yourself. You can do it, but it could be more successful quicker if you start with an existing channel. So take a look at where you’re spending your dollars currently and how much of a rise you get for that, and then prioritize based on that.

So let’s take a look at CRO. You might be familiar with CRO roadmaps. Again, we just saw what our paid team does. This, we could do a teardown. And I say, hey, every other week, we’re gonna run a test of 2 weeks at a time, and there’s 2 different sections. The top row here is the landing page that coordinates with our paid. And so we’ll start tearing up that paid page, and you can use a tool like VWO to target based on the UTMs and query parameters that are coming from that paid traffic. 

So that’s why this first test is to remove navigation, and CTA scroll, so we can turn an existing page into a landing page without having to build a new page. VWO is a fantastic tool for that, and we can keep going down the test and roadmap here in coordination with paid and do our normal things but have a focus on pairing with paid here. 

And so now that we’ve seen some contrast and some differences and similarities, how do they overlap? We’re gonna start with a little look at the data. So in this case, the top row is a screenshot, a little modified out of Google Analytics, and the bottom is out of our Google Data Studio, where we’re pumping into a funnel. So the top section is our paid team, and they’re looking at impressions and the click-through rate. As you can see, we’ll do a major league baseball game. So let’s do a ticket center that we’re doing here.

A 1% or almost a 2% click-through rate in the top right box, is equivalent to about 40,000 or 3,500 visitors or impressions, so it’s not visitors. When we tag that onto the site, for the on-site funnels, we’re looking at almost 12 million visitors. It’s a different metric, but it shows how people are being funneled in from unique visitors to impressions. These are just some of the campaigns I’ll be running, but this is where we know they go in. And we could separate this to just show those 40,000 conversion rates come through. But this is how you can connect the paid work to CRO and start looking down the funnels that happened from where you’re pumping users into. You might not be pumping users into the top of the funnel, but more, but further down the stream, closer to the action, whether they’re in the checkout or the step before. 

Alright. If you have any questions before going through this, please raise your hand then we’re gonna start getting into some examples and deeper stuff here. 

So one of the things we look at here is matching the ad to landing page consistency. When we run a creative ad like you see on the left, which is very impact-focused, and very visual and we send it to a page, which is in the middle of the control, we we see a good return on it. We see a good conversion, but we don’t always see the best. And what we found is if we take a still image or a video playing in the background, that’s equivalent to the best-performing ads and put that as we see on the variation on the right, so there it is in the video. We got a 40% lift.

Now, this was a product in beta, but it still had thousands and tens of thousands of visitors, so this was at scale, statistically significant between both. And it can be done. And this was on Facebook’s channel. So another Facebook ad that we coordinated for with selling Yankees tickets, back to the example on the data. We have the stadium image, ‘Get my seat’ on the left, and the ad is a still frame. We took another stadium image that was a little different angle, but the same stadium, Yankee Stadium, of course. And even though there are other changes, it was the parallel in the imagery that drove that 4.5% lift as we’re going through and doing testing. So this is something key to keep in mind. 

And now I’m gonna talk a little bit more about these different segments. On the left, nothing should be a surprise. We added an affiliate down there since we’re very much an affiliate marketing for a part of our agency here. And the question is, do you run this as a relay race where there’s a handoff from paid to CRO and on-site, or you’re gonna copy and print exactly what’s done on the paid ad on the website? 

And the answer is – it depends. So if we go through some examples here, this was a Facebook ad on the left, and the videos were the same. But when the paid team changed the copy that was written and included no cost, it was underlined or did not include no cost, the no cost one performed better. Everything else, all things equal. So we figured on the landing page, it didn’t have ‘no cost’ there. So we thought, let’s put it in the headline. If it has that much of an impact on the paid system, maybe copying it will drive more lead generation here. In fact, the opposite was true. if we had a loss in lead generation, statistically significant. And so we took a look at this and realized this was the wrong scenario to do an exact copy. The users were already aware it was no cost. They wanted to take that next step, but they wanted their desire and their interests stoked a little bit. 

So another test we did, was on Google Ads, just after, so different channel. We’re taking a look at a landing page that existed and worked for other channels and take a look here. As they were going through, again, a lot of changes in the ad copy. Since this is just a copy, with no imagery, when we pushed it to the controller of the variation, we saw a 14% lift on the variation, and that decreased the cost of acquisition by 24%.

Now there are a number of things here that are happening, but the primary thing was the passing of the baton. We did not focus on the free again, even though Clear Captions Free Phone are the first 3 words that you see in the advertisement. We focused on the benefits. If you take a look at any of those frameworks like AIDA –  attention, interest, desire, and action, there were a few things that needed to happen here to drive this lift and, like, drive down the cost of acquisition, and that was to drive the desire a little bit more through the benefit and make the action very visible. If we hid the action, we still had a huge lift just by making the desire stoked here. 

D:

Sorry to interrupt you, Gary, but there’s a question, Andreas wants to know how does the data studio picture works with customer behavior around ITP, cut cross-device behavior, and the attribution for the paid social and Facebook.

GM:

So attribution would be a whole other conversation that we could do a webinar all by itself. There are some fantastic tools, and we do some great data analysis with our business intelligence analytics team. In Google Data’s, studio photo here, we are focused on the last touch attribution here, but you’ll notice we have prospecting and retargeting for this funnel here. So it’s taking a look at both the first touch, and then retired and getting you just to come back after they’ve seen it. Really good question.

D:

Let us know if that answers your question. I can also quickly unmute you if you want to, you know, kind of understand more about it. Yeah. We can go ahead.

GM:

Alright. Thank you. Alright. So here’s something for affiliates. You may or may not be doing podcast ads, but what I found, let’s say you’re listening on Spotify or any other podcast tool. If you’re not paying and you’re for the premium on Spotify, you do hear ads and some of those ads are really fantastic. The person doing this podcast is doing a read-through, and they have a very high level of trust with their audience. In fact, they become an authority figure. And something I found is we can get 15% conversion through the funnel. That’s purchased. That’s people landing on the page coming directly to these links, as long as we’re focusing on the authority figure. And that made a big difference for these funnels, the best performing reads on podcasts in these affiliate partners, it was the consistency wasn’t too necessary to the discount. It was there. But it was more for having the imagery and the call out to the podcast that the user was coming from rather than going towards the brand and the product.

So that was something with consistency. Again, it’s more it’s a pass-off, but here, the pass-off looked a bit different because we still had the advertisement as a heavy part of what the landing page looked like at the top. So as always, take a look at your mobile versus desktop. If you’re getting much, much more traffic on your mobile than your desktop, it’s worth considering doing that split. If you’re getting a faster different conversion rate between the two, again, the split. If you’re getting a lot of users looking on mobile, but then buying on desktop, that’s a different complexity. We’re not talking too much about that today, but that is worth looking at how you test on mobile to really remove the barriers, figure out where the user is in their journey, and if you have to remove some of the form fields or requests that make it hard for them to purchase on mobile. And then see if you can reduce the drive, the switch from mobile to desktop. We see that on a number of clients all the time and that goes back to the attribution question, but it’s really tricky to be tracking that without a logged-in state. VWO does have some neat ways that you can use the query parameters and the, the source to look at. 

So, let’s move on to how you can take this and work on it for yourself over the next few weeks, and a few days. The biggest thing here is communication among your team. We really encourage to break down silos. Don’t let them build in the first place if possible. Because the best way to grow is to have meetings on a weekly basis or conduct talks between the different teams and act as one unified team. And then here, so we meet every Friday, people can be optional if they wanna attend or not based on everything else going on. But we set up calls to show up, talk about what CRO is seeing, and brainstorm of outlet, what paid is seeing, and we make it fun. We look at data, we present the challenges, we go over what has worked, and what hasn’t worked, and then the floor is open, and it’s kind of a whiteboarding session, which, despite being fully virtual, works fantastically. 

And then align the messaging. Make sure that your team is aware of what paid is sending. They may be changing the messaging every day. But what is the core of what they’re having on the message? Are they talking about something free? Are they talking about benefits? Are they talking about social proof? What are they focused on in their messages and how are they iterating? If you can get that communication from your paid team, then you’re on-site team, your devs, and your optimization of on-site, can coordinate with that and align it. You don’t have to copy exactly, but make sure you’re aligned.

And then let your paid team tell you where they’re spending their dollars and what the return on the ad spend is. The more that you can increase a conversion rate for any of their channels, the lower their cost of acquisition will be. For example, we have inflation right now. So, if you spent $1 at the grocery store two months ago, that is now probably $1.20 or $1.10 for that same product, or more if we’re talking gasoline. Here, we’re talking about deflationary actions.

So $1 that your paid team would have spent yesterday, now that you’ve increased the conversion rate, they only need to spend $80c in order to get that same result. Those are huge things in a time like today that will make your clients, your teams, and your bosses very happy. It will make your businesses more successful and you’ll be able to compete. 

So feel free to interrupt me with any questions, if you go through that. We could dive more afterward about what makes this work. It’s gonna be custom to your team, and that’s the big thing about the chocolate peanut butter. How do you eat it?

This is a framework to get started, but how you set that up with your team and how you get the communication to go is gonna be up to you. So lastly, I promised you 5 tests to walk away with. So here’s a way to tune it up right away. 

1. Copy the language on the ads, CTAs, onto the landing page from the ads to the landing page. Just do a direct copy, and make it parallel. 
2. Make it continuous. Have that as a follow-up test. take and look at the next step. If you have the attention and the interest, stoke the desire, maybe stoke, go direct, direct to the action on the landing page.
3. Connect your imagery. Take a still image as you saw, from the video content, and get the best-performing ones. You don’t have to get the best still image the first time but iterate on that. You can also, if you’re doing still images, copy exactly the still image to the landing page. This shouldn’t be surprising in that sense. You probably see a lot of advertisements for clothing having a specific shirt or sweater. Then you click on that and get to the landing page, and it’s a different product listing page, and you can’t find the product you were looking for. 

Imagine how frustrating that is, or if you’ve been through it, you know, now expand that to other areas. You wanna have that consistency through the imagery just because you’re setting expectations in the advertisement, and the website is going to deliver on those expectations. 

4. Expose the CTA action. Here, instead of having a button, depending on how warm your users are from the advertising, if it’s that retargeting group and they’re ready to buy, or if it’s a direct response on the prospecting that you saw, earlier, then reduce the form fields. You don’t need as much information, and that could be a barrier for them to purchase. 

Whereas other channels, you might need more information to reach out to them. or have more information about them. If you expose form fields that are otherwise hidden, you expose pricing that otherwise is hidden, which could drive users’ interest down the funnel as well, or send them directly to the PPP or even to check out. 

Shopify allows you to send links directly to checkout from advertising and have the product in the cart. If the user is ready for that and they just wanna buy it, that is a huge way to do it. And if you’re worried about the average order value being too low on purchasing that, there are post-purchase options that you can do. 

But, focusing on getting the user through the funnel as quickly as possible, and exposing the action rather than just a call to action is a huge way to do that. 

5. As you’re doing this, your website will grow from one version of the website to responsive versions of that website. VWO is a great tool for doing this, especially using UTMs, sources, and campaigns. Building it from a static version of the site to a responsive version that serves a few different channels. And then as you grow in your sophistication and complexity down the road, to dedicated landing pages. That you’ve proven the value, you can now build on that, whether it’s unbalanced or developer, there are different costs associated. But making changes specific to an audience, building funnels specific to them, is the direction that you should go in. 

Alright. And if you could take a screenshot of just one of these slides today, I’d recommend taking this one. These are the last five tests that I’ve recommended, based on the channels that you’ll most likely be focusing on in whichever order is most appropriate for your paid team and your business. 

So with that, I want to say thank you. It’s been a lot of fun, and I’d like to open the floor up for conversations.

D:

Thank you so much, Gary. That was such a well-put presentation. I’d like to request all of you. If you have got any questions, do let us know in the chat box or, in the questions panel, and Gary would be more than happy to answer them. Andis, I’m going to unmute you and, maybe then you can carry forward the conversation.

Andis:

I want to thank you for the great presentation on the AB testing, and I like this, scoring method with different, adding screens with the TikTok snapshot to Google,  Instagram, and also the tips that we have in the end there with the AB test. I have a question about A/B testing. What tips and tricks do you have to get the the developer in IT department to prioritize your A/B Test? Even though we have a great hypothesis and, a high score method, it’s hard to get the department create to a variant, considering I don’t have the required JavaScript and web developer skills. 

GM:

Gotcha. So the the question is what advice do I have on getting things to be prioritized for development?

A:

Yeah, exactly

GM:

Okay, The business goals are a really great place to focus on if you’re making a request that’s in support of the business goal. Let’s say we could talk about ROAS here with an example. If we wanna increase the return on the ad spend on the marketing team from 5% to 5.6%. Then the request to have a developer do something in order to support that goal would be worth x amount of dollars to the company. When we’re looking at the revenue for paid and the way that they pitch it, this is part of that setting of expectations. Okay. We need the following resources in order to deliver this ROAS. that’s part of our pitch as well. And that goes into pricing, that goes into telling them how it’s staffed up, the coordination, all those things. It could help out very much with your developer. But if you have a spending budget, and you need a certain amount of revenue, but can’t achieve that revenue without the support of a developer, then outlining what those tasks are for the developer and how much they would drive potential value for the company could be a way of selling people on it. Otherwise, for a tool like VWO, there are ways to use it without having a developer or getting even a budget to hire a developer from outside. That can work in VWO so that they’re making changes only on the front end as the page loads rather than pushing code to the back end. which some clients and some organizations may have more challenges with.

A:

Yeah, great answer. Thank you. 

GM:

Cool. Yeah. Here’s my background, I’ve done full-stack engineering. For some clients, it’s a lot of fun. When you’re in the tool, when you’re technical, you value the work you do. When you talk to people in business roles, they look at development, not necessarily as something to always be doing, but as a function, a cost function. And the trick here is to make it make parts of that cost function become a revenue generator. And if you combine development in support of a higher return on the ad spend, you can directly tie their actions into revenue generation.

D:

I don’t know about others, but I agree 100% with that. I think there are no more questions, Gary. Thank you. Thank you all for joining in for this webinar. I hope you enjoyed the presentation for any other questions or any more doubts. You can write back to us, or you can connect with Gary. And there’s a handout attached, to your handout section. It is an ebook that contains a test and summaries from tests that were done successfully for brands across the globe. So you can download that for your reference. And, see you next time. Thank you so much.