Are you overspending because you are failing to use one simple tactic?
Using multiple ad sizes (specifically those with lower competition) is something many advertisers overlook, but it has a huge effect on campaign effectiveness.
Here is our virtual roundtable with our expert panel
discussing how and why using various ad sizes really works.
- Why sizing affects your ad placement and reach
- How to give your campaigns a boost with small changes
- Which ad sizes matter (and which don’t)
Watch the recording
Hey everybody, welcome. Thanks for joining us today. I’m Kathleen Davis on the marketing team here at Perfect Audience and today we’re going to be discussing how you can increase your campaign performance by 20% or even more by using one simple strategy that we’re going to get into today. But before we get started I just want to remind everybody that we do do a question and answer at the end of our webinars so if something comes to you or you have questions already, go ahead and throw those in the chat box at the bottom of the screen and we’ll get to those as many as we can at the end. So I’m going to introduce our awesome speakers for today. First up we have Todd, who’s the, he’s a returning speaker of ours, he’s a common guest. He’s the CEO of Ascend2 which is a marketing research company. And Todd’s actually developed their research-based marketing methodology so we’re excited he’s here today. Next we have Jalali, who’s the founder of Robauto and he also serves as the AI lab director for Perfect Audience. And then we have Eric of course, he’s our general manager. He’s spoken at lots of other marketing sherpa events, email summit and things like that. And then finally our guest speaker for today is Gaba. He’s bringing us 15 years of digital marketing experience including a lifetime spend of over five million dollars in ad spend on Facebook. So he’s bringing us lots of knowledge on the ad space specifically related to Facebook. So we’re just going to go ahead and jump right into day. So we’re recently done some research with Ascend2 that Todd’s very familiar with so I’m going to hand this over to Todd.
Thank you Kathleen, and thank you for everybody for attending. And I had the pleasure to work with Perfect Audience on some programmatic advertising research and actually that reports now available on their website, but one of the things we kind of found and kind of spurred this conversation for today was when we asked marketers how would you characterize your program, is it very successful, is it somewhat successful, or unsuccessful, and you can see the data results here. And you know we kind of saw this fact that there’s a pretty small percentage of companies who feel like they’re best in class or very successful at this thing we call advertising. And as we started, Eric and I started talking about this a little bit more, there’s just this thought of just, you don’t always have to hit home runs. Sometimes it’s just doing the little things exceptionally well, and that’s really what can make you best in class. Obviously you always want to be testing. But you want to look at these little ways that you can optimize your program and so that’s why we today we want to talk about ad sizes. Which at first doesn’t seem like maybe a critical topic, but as you start getting into the data, you’re going to see that actually ad size and thinking about strategies and what you’re doing when ad sizes, is actually a pretty critical part of your advertising program. So with that I wanted to pass it over to Eric. He has a little bit of additional data information to share.
Thanks Todd, yeah so it’s interesting to me that we got as many registers for this event as we did. We were actually just talking about this a few minutes before we went live, that it’s a pretty small niche topic. But what’s happened is I think people are realizing, just to your point, that there are significant focused execution related things that we can be doing really well that doesn’t feel like it’s a big strategy, but prime mover thing, but it’s a small repetitive, small repeat, consistent execution, that really is what, and testing you know, that’s what makes marketing work well. So with that this is just a real quick slide I wanted to sort of like, sort of couch the discussion so we were getting together as a group, I don’t know a couple weeks ago, and we were talking about some of the topics that our advertisers in Perfect Audience were having, some of the questions that were being brought up, and one of the things that really stuck out to me was how focused ad creation was around certain sizes and you know, then the question sort of arose is well, why is that, what’s happening, why is that happening? And we actually were able to dig into the data a little bit and talk a little bit that. And it actually was big enough that we were able to put it into the format that you’re seeing today. So we’re pretty proud of it. There’s some fun stuff we’ve got in here and some a ha moments. So we’re really excited about that. And then of course for anybody who wasn’t able to make the live event, we’ll have that available on a recorded version on demand here in a few hours or so.
So here we’ve got the IAB standard for advertising or ad sizes for web and mobile. I just really wanted to sort of talk a little bit really quickly through this. This specifically is web and mobile, this is not Facebook obviously, but you know everybody’s familiar with the big billboard, you know the 970 by 250, the leaderboard, the 728 by 90 right, and then you’ve got your standard 300 by 250, the 160 by 250 or 600 So those are all sort of like these standard ad sizes that everybody knows, everybody develops or designs for their campaigns. And so in the next slide this is sort of just broken out by usage. So in the, and sorry for the granular little numbers there it’s hard to see, but this is really just to show how these banner ads are being developed and then how often they’re showing up across the web. So the biggest one is the 300 by 250. that’s the medium-sized banner in the top right typically. Pretty darn close to a square. And you know, that’s the, of all the sizes that’s going to be the largest size. And if you go to the next slide real quick, the sort of this graph or this chart shows like how kind of concentrated that it really is. So when you look at it really close to 90% of all of the ad sizes are really concentrated around the top four sizes, which is a 300 by 250. The 728 by 90, the 320 by 50 and the 160 by 600 and that is a major concentration around sort of like these four sort of dominant ad sizes. And we were just briefly talking prior to this, Todd had made a really good point because i think Jalali was asking the question, he said well why is that, why is like a 300 by 250 like the sort of that’s big, dominant, you know ad size, so we were talking a little bit about that. Do you want to just talk a little bit Todd about why you think that is?
Yeah I think part of it has to go back to the pre-digital, to the print ad days and everything was set up with standard sizes that you used and all advertisers used those. And I think even though they’re not necessarily needed for digital, I think that kind of concept of having standard sizes kind of became a thing and so I’m sure you guys might have some other thoughts on that, but I just, I think that’s kind of how it started. i think I have a thought that might also be, because sometimes people, some of the clients, they ask why are you running my ad with this image with this tax displacement doesn’t make any sense. It’s because it has the best return on investment. So let’s say I made all the ads with all the different sizes and the one that’s generating the most revenue for my company at the least amount of spend, that’s the one that’s going to stick, that I’m going to stick with. So maybe that’s influencing why people are focusing on it so much.
Yeah I think that’s, I think that’s a really great point. I think also there’s static display ads, banner ads, and video now are really it’s you know, another point to this is they are all very much designer centric right? So there are these ads that are focused specifically around certain ad sizes and anecdotally, I don’t have any data on this, so don’t quote me on it, but the focus around when you send these creatives off to a designer, what you’ll get back is something that’s actually really convenient for them to work off of right? It’s hard. I mean if you think about it it’s really hard to do small text on a banner ad and that’s like super awkward to place it and it takes some work right? It takes work to place them around and make sure that the text looks right and the call to action looks right and if it’s coming from a designer, obviously the way that it looks visually is going to be sort of their top of mind thing, but we know as marketers that there’s a lot more to it right? There are things, like I’ve seen some of the ugliest, weirdest looking banners on the planet that have outperformed two to one over something that was really gorgeous looking and really well laid out and just looked perfect. So that’s why we test right? That’s why we test, that’s why as marketers we come in and we test these things and so I want to talk a little bit about that but this graph was actually really surprising to me. And then if you look at the next slide what really stood out to me is like okay well, if everything’s concentrated on these areas over here, then what about all of the rest of the ad sizes like where, like are they getting any kind of visibility at all? What’s happening with them? And so this is really like the core discussion right? It’s the whole point of the topic, of the webinar is, we’re looking at these, like the core focus. And if you think about programmatic and RTB and the way that we do digital advertising today, we do everything programmatically right? So it, but at its core it’s still a marketplace right? So there are buyers and there are sellers of inventory and they meet right at the optimal price right? So you know when you say I want to bid at this CPM, a publisher comes back and says this is what the CPM is and then all of the other advertisers are in bidding for that exact same inventory slot all within 50 milliseconds and that decision is made, that auction is closed and your ad is displayed all by the time that somebody loads their page on their browser.
So having all of this done in real time like this has really sort of shifted all of the available inventory. It’s increased the number, like the actual available inventory and it’s sort of right sized, what CPMs are worth to us as buyers and sellers or advertisers and publishers depending on how you look at it. So when we looked at this we said okay, wow, there’s all of this core inventory that everybody’s bidding on and all this stuff happening. And then there’s all of the long tail right, all of the long tail that as you can see from the graph hardly you can even show like towards the tail end of this long tail, you can’t hardly even see a little blue line there because very few ads are being served in those areas. And so who gets that? A lot of times it’s showing up on Adsense or sold as remnant space or what have you, and what it ends up being is really low-cost CPMs for those ad sizes for that particular inventory. So you can see from the chart here at the bottom, the bigger ones you’re gunna pay more for higher CPMs because it’s more premium soace. It’s what you would think right, but then all of the rest of it is close to a third of the cost on a CPM basis as the core, or sorry as the premium inventory. So that to me was sort of like the aha moment when I saw it in this chart. As I saw, wow okay you know this is, anybody want to add to that?
I have something. This is something that we see a lot on Facebook is frequency by which something shows up to the other marketers, because it’s a competition, it’s an auction house we’re beating against each other. So we cannot ignore the targets that show up when they’re looking for a target for them to find the audience and I’m also going the same thing if we both choose the same exact target. It’s going to be more expensive, so what we do is pull up the 20,000 targets that might be related and we make a graph that looks exactly like yours for frequency. So we can measure the frequency by which those targets appear on Facebook. And every time we choose the one that’s highly relevant and has a very low frequency, we get better results than if we were just regular targeting like everybody else. So I think this makes a lot of sense. Yeah for placement as well. That’s it guys, anybody else?
Yeah I think just, this is really interesting and it’s I think, one thing that sometimes I forget when I’m doing this is, this is a real-time auction that’s going on. I grew up going to these farm auctions. There’ll always be the tractor and the nice conditioned truck out front and then all these tables of tools and different things right, the deals are in the tools and the stuff not everybody was bidding on. So it’s the same thing going on here is one and two, there’s definitely a collect, like this frequency thing, there’s definitely this impact of showing people in different places on different devices as they move around from the web to their Facebook, to their cell phone right? And that’s another aspect of some of these differences. I think you get this kind of collective ground swell when you reach people in different areas. So I think it’s, I just, it’s real-time auctions going on, like literally the 300 by 250 is the most popular one. It also probably has the most inventory, but there’s just so much opportunity. What we’re finding as we kind of push down, out even to the end of that chart for little pockets of opportunity for conversions and then also just kind of this building a building report across there, the graph essentially of advertising with different sizes is what this does.
Yeah. I mean I remember Eric I think it was, about a month and a half ago when we did a webinar together, and Jimmy Ellis was on, and he said you know, I have my designer make every option possible. And I think it was that concept of, I want every tool in my toolbox available so it’s all available, it’s all been well thought out and, like Jalali just said, you have all those options available in the bidding process. Yeah I think that’s right, I mean so there’s an opportunity right in the long tail. So you know I think about my SEO days years ago when I was doing SEO, there’s everybody’s focused on the most competitive key terms had the head terms, and they’re trying to win those general head terms and that is extraordinarily difficult to do, but there was tons of opportunity on the long tail right? And this is the same sort of concepts right? So there’s tons of competition up front in the head and then on the tail end that’s very specific inventory or keyword terms, however you wanted to look at it, less competition, easier to get, and if we’re in an auction scenario because we have less competition it should be lower price, more competitive pricing.
The other thing I wanted to make a point of here too before we moved on was, we’re talking about there’s less inventory right, across that but it’s still pretty decent. I mean if you think about it in this long tail in this particular scenario that we’re showing if you look at the little chart off to the right when we ran this test, so you know core premium space generated around 92,000 impressions. And then all of the additional sizes added up to 57,442 impressions. So you know that’s, this is pretty decent, that is actually higher than I expected it to be. So there’s a real opportunity to be able to increase your reach. And so I was talking actually to an advertiser, it’s been a couple of months ago now, and we were just, he was asking and saying hey you know, so we sort of bid on everything we can get our hands on, the advertising that we’re doing is effective but we need more reach, like how do we go further and this was actually one of the topics that came up. It’s okay, well what are we working on today? Well, okay I’m doing really well on these ad sizes, on these campaigns, but you could tell and he was doing the same thing right? He was having a standard ad set and he was doing great with rotating the ads in, he was creating new ads and refreshing them constantly. So they were, he’s keeping his CTR up but what was happening is he was only focusing on those, but he really wasn’t spending any time on any of the other, you know the ad creative, sizes, and ultimately he was missing out right. So improving that is getting the same kind of reach or sorry, increased reach with expanding your ad sizes is something that he had success with. And I’ve seen time and time again.
So let’s move on, yeah thanks Eric. So that’s a really great segue into our next slide which Jalali was instrumental in running these tests, but this is kind of our example, these are the ads behind our chart and our numbers. So I’ll let you Jalali talk you guys through this one. Yeah so some of you may have seen these ads as you’re signing up for the webinar. So we always test our own stuff and try to figure out really what works and the reason we do that is it can have a tremendous different impact on your business, our business, or anyone’s business right. So small changes can equal a big impact. But we’ve been talking about all these different ad sizes and some of you might be thinking like okay, well it’s hard enough for me to get one banner, like how do I make all these banners? We have the same problem right? We have thousands of advertisers, we’re always running campaigns, we’re always changing campaigns. So kind of a strategy that we’ve deployed is to start first with the 300 by 250. So we run a quick blast of traffic with two or three different concepts. And what we’re looking for here is to see is there an outliers? Is there one design that kind of works much better? In this case they were pretty close right? It was pretty close. But one starts to pull ahead. And actually it’s a pretty significant improvement. I can’t tell you why, you can kind of look at it, I think the one on the right is probably a little busy. I really wouldn’t have guessed which one, but anyone so we did this initial test without a huge design lift. We just did three same size things figured out which one works first, then we put the rest of the sizes together. So we didn’t use all of them, we actually picked ones that were most popular or that we thought would work best, and you can kind of see the impact as we talked about before. It’s like 90,000 plus on one and then 50,000 additional impressions. But the gain, and this is what is fun, like even our own business, a 62% decrease in cost is significant for us. It could be the difference between this campaign not working or working right? Or it could be the difference between our quarter working or not working right. And that’s why we do this in Tesla’s.
But I was actually surprised by this. i thought yeah we’ll get some additional play, but the cost would be roughly the same. Kind of makes sense. But it’s really kind of even opened up my eyes. Some of these campaigns are running, how do we expand this reach and if I could get a 62% decrease in costs across the board, again it’s gunna have a tremendous impact on people’s bottom lines. So yeah, so just again tested it this could have actually gone either way. We had a pretty good idea, we’ve done some of these in the past but much cheaper and very significant reach. And I think the thing that’s now shown here is what is the impact to that person seeing the ad across all their devices right? like I think one of the things about Perfect Audience that’s really interesting, I get a lot, is to the end user, to the person that’s visited your website or that you’re targeting, it appears like you’re running a million dollar campaign across major channels right? They see it on their news site, they see it on their social media, then they open up their cell phone and they see it right? It’s kind of, it has an impact and that frequency and that recency really kind of can build up and make stuff work that wasn’t working before. So anyway, yeah. It’s just a quick test just so we can kind of see what we were looking at.
And then in terms of the breakaway, just so you kind of understand, so 90,000 impressions on that first kind of a, b split on the big square banners and then a couple days later is where we put the other ones live, and the result was this blended CPM, much cheaper than what our original, so 26% cost reduction across the board. And we’re just getting started. So we’ll expand what I would do next on this one is I would start to maybe look at where there’s some additional ad sizes we should add to it, to this set just based on which ones are working. So we’ll start to drill into click-through and conversion by size and then once you get into these very small ones, there is something design-wise we should do. Like we were talking about, before we get on this calls it’s hard to design little tiny micro ads right? The thing about advertising is you’re dealing with a very small people, like you have to explain it so quickly right? You don’t have a chance to when you’re looking at an ad to learn or process right? So as you get smaller you have to get better and better at making your point. What is the headline? What’s the call to action? What is it you want them to remember? Or you know, kind of learn from this and take subsequently take action so it is hard right? And we actually have a special here we’re gunna offer in terms of some creative but it’s hard to get all these creatives together. It does matter and if you’re gunna do this, like do it right. Start with your 300 by 250s, figure out what’s working, use that information, even drive some of your strategy in general, like we do the upfront message testing as a way to test some product theories right? Like does this resonate or is this a good way to explain this? You know that kind of stuff. So you can do some of that stuff and then just look at where there’s opportunities in that long tail, and we’re here to help you with that. Honestly, it’s that part of what we do. But yeah, very interesting, it surprised me.
It’s interesting. This whole webinar has kind of got me really thinking about what I need to do. The campaigns I’m looking at right now in terms of filling this stuff out, yeah yeah thanks for joining. If you want to leave and start working on some campaigns go ahead. Yeah, awesome. That, kind of like Eric said in the beginning, that only really applies to web and mobile, but we have a Facebook expert on here with us today so, Todd has got a lot of experience with Facebook ads and he’s gunna talk about kind of how this relates to Facebook and how kind of some of the same tactics apply.
Yeah Facebook, all you need, you have a couple of sizes that you can choose. It can either be a square or a rectangle. If it’s a square it’s 1080 by 1080. When you put that on Facebook, it’s going to give you options for placements. So you can choose where you want that square to appear, or rectangle, the square is better because it occupies more space on the screen when it’s on mobile phones. But the thing is, it’s not the same thing as choosing the size of the creative that you’re going to do because Facebook is going to redimension it anyways. So what you got to do is you have to try to get the image to be as good as possible, but as light as possible so Facebook doesn’t mess with it too much. If you make the best picture you take with a great camera and you put it on Facebook, they’re going to really mention it. You can get around it if you post the picture on your profile or on your page, a higher resolution picture, then you promote that as a post, and if someone clicks the image they go straight to the link, it’s even better than an ad because that you have to click on the call to action and you can add a call to action to the post anyways. And then you can get around Facebooks re-rendering the image. What you see on the screen here is a two, almost two million dollar test. This was how much we spent this year with very similar products right, so you can see all the placements that Facebook has here. You have the feed, Instagram stories, right hand column, everything that Facebook is going to give you as an option. However, the problem is, if for example, I just choose feed because I want it in search and then in-stream video, Facebook is going to give me a prediction of conversions that I’m going to get that’s really small. So I have to try every single combination of placements to find the perfect one that’s going to give me the most conversions. Facebook gives you predictions, you would not believe those predictions, just know that those predictions are correlated to reality. So if Facebook says 40 conversions that you’re gunna get, you might get two right, but if it says 80k, then you might get four. So you try all of the combinations of placement that Facebook has when, if you’re going to upload the image to the ad itself, you have to try to make that image as light as possible because they’re going to mess with it. I don’t want to take too long, I can talk forever.
So consistently, let’s see yeah, consistently inconsistent is that right for attribution, is that what you’re sort of saying? So you’re saying that the amount of conversions that’s being reported is going to roughly correlate to what you think is actually happening. And if you see twice as many you know it may not still be exact but it’s going to be as a ratio to the yeah, as a tip to make things easier, go with the feed if you’re trying to sell some impulse buy product on Facebook, go with the feed because it’s always the best. It’s been the best for years now. But if you add the support placements like we’ve been talking about before, you get further, faster, and cheaper. So if you spend the time to try to, you choose the feed as the best placement that you’re gunna use and then you start choosing the other ones that you really like, that you think your brand should use because it’s relevant to you if you want to go to instagram, if you don’t want to go to stories if you have a video you just try them out together with the feed to see which one is going to give you the best predictions. You shouldn’t believe the predictions, but it’s the direction that you can, you know think I’m going the right direction and then you go that way. It makes sense. Now the CPM, the CPM here is something that you can go to one back as well.
I just, real quick, the previous slide, the CPM does not correlate on Facebook to the return on investment at all because it might be really really cheap. Facebook’s gunna find you a lot of people that love to click. Let’s say that link clicks are super cheap. They love to click, they never buy anything, right? So those are not the people that you want, so you want to find the buyers with the targeting parameters that you’re going to choose. And the CPM is going to be higher. So you can ignore the CPM and look at the ROAs, which is the ROI, they just renamed it. And then you can choose the best ROAs and focus on that. So the CPM is not as important on Facebook. It is kind of important, but focus on the return on investment because that’s the best way to go. Yep, yeah. We were actually, we had a webinar, I want to say a month, two months ago, something like that, where we were talking about key metrics, and like ROAs and how does that differ from like a true campaign-based ROI versus other types of key metrics that we would want to, we ourselves, would want to be focusing on, and what we sort of preach every day to advertisers that are working through their campaigns and working to optimize. And we have a, it was actually a pretty big topic, such that you know, I don’t know, I can’t remember. Todd do you know? When it’s like a whole webinar on basically KPIs around.
Yeah I think our next webinar is going to be on KPIs. Oh is it? Okay yeah so that one’s gunna be fun too. And it’s exactly to your point Baga, I mean there’s a real focus around getting the right KPIs in place, knowing what you’re supposed to be measuring for, and then optimizing around that. So yeah, I have a list of the KPIs and their priorities, so if you have this one, forget about everything else, you have this one. If you get to the ROI you shouldn’t really care about the other KPIs in my opinion. You shouldn’t care about, they keep is because you know you put a dollar here, you make five. If you put a dollar here, you make six. Go with this one, spend as much as you can there. If it stops being six and it goes into four and then you go to five, and that’s why we have thousands of ads. And then it’s the whole methodology that’s a little different. But I don’t want to go too long in that engine here. Oh you’re fine, we’ve got some time. But I know you have one more graph, I don’t know if you can talk us through this one. This is just to show the point that the CPM and the cost per purchase is correlated in this case because the AOV is the same for everything. It’s so the cost per purchase, the lower it is, it means that the higher the ROI is. But against the CPM and the cost per purchase has no relationship whatsoever and that’s what I wanted to show on this graph here. Is that on Facebook if you get as further as you can and also get the cost per purchase and what’s your return on investment? That’s the metric that you should use. So this is just to show that the CPM on Facebook is not really a measurement of success.
Gotcha. Well that’s, thank you Gaba, that’s awesome. So this slide is kind of just a catch-all for what we’ve been talking about today. What we’ve been talking about with how you can get a more complete reach with using different ad sizes and how that’s super helpful. So Eric I’m going to let you talk through the rest of these points, but a lot of these things are things that we, you get with Perfect Audience, and that’s why we do these. And kind of you know, yeah right, that’s right Kathleen I think too that I would want to underscore is, we went through a lot of topics. Some of them are based in pretty in-depth algorithmic approaches, and you know there’s a lot of stuff sort of happening underneath. But I think what’s important to know is that it doesn’t have to be, you don’t have to feel overwhelmed by this right? There are simple steps, simple things that you can do in order to get a campaign running. And then as you optimize around like one specific KPI, one specific goal, you can start getting more sophisticated. And the idea behind Perfect Audience really when it was originally built, it was built by people who were using the platform themselves for their own campaigns and their own things. And so it’s really just evolved from that. So you have lots of granularity, lots of transparency in the reporting. We give you basically everything at your fingertips, probably more than you necessarily need. But i, you know in case you want to dive in, you’re able to do that. And you get the control and the kind of like refined targeting. That is really beneficial as you dial things in. So that’s really all this is supposed to do. I mean it just gives you that kind of capability and then again, we as a group we are very focused on just trying to make campaigns run efficiently, run well for advertisers. Sometimes that just takes a little bit of, sort of like us coming alongside and guiding an advertiser and helping them with their campaigns. Maybe it’s somebody who’s in the finance space and they want to know what are the best practices for finance related categories and of course we’ve got thousands of campaigns worth of data that show what’s working and what’s not working. And that’s beneficial right? So that’s the part of the AI, it’s the last bullet point there. It’s part of the AI plus experts.
We are of the opinion that AI and technology, it is irreplaceable, like you cannot compete these days without having that. And on the other hand, or sorry on the other end of the spectrum, there is focused goals and strategy and expertise that comes along with years and years of doing this kind of stuff. And if you put those in that, in and of itself these days is not enough right? You can’t out optimize algorithms. You can get really close but you’re not going to get a whole, all the way there. So really the whole concept behind Perfect Audience is putting those or blending those two things together right? So you have the strategy and you’re focusing on the right KPIs, and you’re blending that you know, marrying that up with technology that makes you faster and more experienced, or sort of faster and more effective in your campaign. So that’s the idea.
Awesome thanks Eric. So this again, we’re just gunna kind of reiterate, this is the reason that you use the different ad sizes. It kind of helps your subsidize your higher CPMs and you get more reach for your money. So now we’re gunna go into our questions so we have some, but if you still have questions go ahead and send those in as we’re talking. So our first question is for Gaba. For me? What Eric, I said, we love questions, yes we do. So if we get too many that we can’t answer today, we will do a follow-up video on Monday and that’ll be on our blog with this recording as well. So don’t be shy, go ahead and send in any questions that you have. But okay our first question for Gaba, is you mentioned that the feed on Facebook is the way to go and that’s the best use of ads, but do you run multiple ads per campaign and like different creatives per campaign? And if so, how many? So I do 200 ads per creative. Okay why? Because Facebook delivers ads by waves based on filter bubbles. Filter bubbles, I used to call them servers before I real the book that’s called Zucked, where they explain how Facebook has filter bubbles. But when we reverse engineered it, we called it servers right? So if I have a target, let’s say I’m targeting donuts. I want to find people that like donuts, and then I have the picture, beautiful picture of a guy eating a donut. I have the tax to have everything and I make five ads. One has one image, the second has another image, another image, another image right? One of them is gunna perform the best, and I’m going to say this is the best image. But you don’t know that because they’re in different filter bubbles. And each filter bubble has a cluster of behaviors. Facebook divides us into the behaviors that we have on their platforms.
For example they only show me videos of people working with wood right? I never know I like to watch this but because I spend so much time watching, they keep showing it to me forever right. So if I like donuts and I watch a lot of wood carving videos, Facebook is gunna find more people that are like me in the same filter bubble as me. So all of those people see the same type of content, they behave the same way that I behave online. That’s why I make 200. Out of the 200, 2% outperforms the rest by a lot. And those are what we call sweet spots. When we find sweet spots and then you measure sweet spots against sweet spots. So it’s 200 per creative, then you can have an idea of the overall performance of that creative, then you can start having a comparison right, then say oh this creative is really good, but it doesn’t have a sweet spot, it has a bunch of average ads. But this one that has 200 ads has 199 ads that are terrible, but one of them has an ROI of 12 and makes 100 purchases a day. So you have to play by their rules and their algorithm stuff that they have. And another thing, do not ever increase your daily budget by more than 19% or they reset the filter bubble and you lose it. So you found the sweet spot, you got excited, you double the daily budget, they reset it, don’t do it. The 90% increase in daily budget per day, that’s the secret.
Awesome thanks Gaba. There’s a follow up question to that kind of, so if you don’t have the budget to create 200 ads off the bat, what do you recommend? What would be the next best thing to do? Fifty. if you have fifty dollars, do fifty ads at one dollar a day, daily budget. Just so you can find out where your filter bubbles are. How many filter bubbles do you have inside of your targeting there are actually good from the 50 ads that you did, did you get one, did you get two, did you get three? That’s how you’re gunna find out. You need some money to play this game. But if you only have like fifty dollars, put a dollar a day, daily budget. The daily budget doesn’t have to be high, but okay you put a one dollar you’re going to have to be increasing 90% per day until you get to the budget that you want. It’s going to take longer, but that’s what you got to do to spend less.
Awesome so yeah, I think that that answered your question Ashley. But if you know, obviously if you have comments, questions, send those in. We have another question from Travis that, and this is just a general question for everybody so feel free to jump in. He asks, when I’m doing my initial ad test, is it better for me to test certain variables? I think he means like having similar ads with one thing different per ad or is it better to kind of have three completely different ads just to see what’s out there. For Facebook it’s the same answer that I just gave. I want to know what you guys think. Yeah I think so if I understand the question correctly, and you know feel free to chime in if I’m not, but if you’re first starting out and you have not run any ads whatsoever and this is really your focus is just the main ads that are like, you’re just launching a product or you’ve got a brand new website etc. And you’re trying to find what is going to work the best, testing radically different options are great right? it helps you hone in. But I think the important thing to know is that this is an iterative process right? And this is where people fall down I think over and over and over again. And this is like a huge opportunity for anybody who’s on this webinar or listening. Yes you know you really want to be able to find what works, test it, and then you have the mindset of an a/b split you know mindset where you’re testing iterative improve options or different types of call to actions, different types of headlines, different types of copy. And Jalali has talked a little bit about this and Gavin was just talking about this. So I think the mentality of somebody who is coming into this fresh, who doesn’t have any ads or anything at all, you want to make sure that you are testing over and over and over again because what you start with is not after testing, is not where you’re going to end up.
Where you’re going to end up is something like a two times or three times where you started, unless you just got ridiculously lucky right? And you can never ever ever build, you know be your control. So testing is an incremental and sometimes painful sort of process. But testing over and over you know, iterating over and over, finding ways to be able to beat the control or beat the, meet the original baseline is the only way to go. So a lot of times that’s what people do, is they’ll load up their ads and then call it a day. In one of the earlier conversations we were talking about how these ads over time, with Jalali’s test that he ran, he was focusing on trying to get a baseline and then he was slowly but surely coming up with, well quickly actually, new options to test and ultimately kept beating the original. That’s how you win right? is over time. And as you test and as you iterate. So anybody else want to chime in there?
Awesome, I’ll go. So you have to, you want to test three different options right? Make the first one, see what happens. And then you make a variation of it and see if it improves. If it did not improve, go back to the original and make another change. But if you did improve, now you have a new baseline. Grab this baseline, change it again with something that you think is going to work. if it works out, good. And then you forget about the previous one. If not, you go back and choose another, change to try to get a new baseline. Yeah something like that. Yeah that’s great advice. So we are going to move on here just because we’re getting short on time, but any other questions we will go over in our follow up on Monday.
So we have this free ad set that we offer from Perfect Audience and Jalali kind of helped us come up with this and so I’m gunna pass it over to Eric and Jalali. Yeah so first of all, we’ve covered a lot of stuff here and we get excited about things like click-through rate and conversions rates, partly because we’re pretty geeky and you probably are too because you’re on this, but it also has a huge impact right? It’s like, I was saying it could be the different between something working or not. So what this kind of comes down to is, it’s a mix of art and science. So the art is kind of what you as the marketer have control over. We’ve talked a lot about AI and all these automated systems and bots right? It comes down to you just can’t manage that anymore by yourself without technology. And that’s what Perfect Audience does. So there’s an underlying platform that automates it, and always is trying to optimize. But the ads are always an issue right, because it’s like how do you get, which ads do you start with, and what should the design look like? What should you test first and so we’re kind of taking a little bit of risk here, offering this because it is a lot of work to do. But in addition to kind of the matching our traditional matching ad budget, i think this one’s up to five thousand dollars we’ll match for new advertisers, we’re also going to do a complete ad set if you just go back to the previous slide of the popular ones that we did that we’ve just tested for you guys. So we’ll start with the 300 by 250, is how the process is going to work. It’s going to be the first twenty advertisers that go in and create an account. We’ll do the 300 by 250, you get one review of it, our professional designers will be working on that. We’ll try to come up with something based on what’s worked in the past with you, and then we’ll extend it out to the other sizes. You’ll get a copy of the ads loaded up into your account and also an original image that you can edit and play with, make variations of. So that’s actually a pretty good offer. We are going to limit it to whoever’s first. I think Kathleen is putting the link in the chat. You can also just go to perfectaudience.com/match and sign up and we’ll go ahead and get your accounts created today and then someone from our creative team with follow up and with some questions about kind of what you want designed. So yeah this is a good deal, kind of reward for being on this. Go ahead and check it out.
But on the optimization side, don’t get too stressed out if that’s not something you guys do. That’s kind of what we help you with right? Is so anybody that comes through this, we’re going to be monitoring and helping and applying our technology. So it’s not something that human is necessarily is going to have to optimize, but I appreciate everyone coming. Any other questions or do we have anything else we need to cover?
Yeah we have, unless you have anything you want to add Eric about the offer, we’ve got our next webinar coming up in two weeks, so that’ll be July 23rd. it’s going to be the same time, it’s going to be 2pm. So if, and we talked about this a little bit, we’ve touched on this. We’ve done some research with Ascend2 with Todd, so thanks Todd for that. And we’ve come up with the seven most important metrics that you should be using as a marketer in 202. Eric if you want to have anything you want to add? But if you want to register for that today you can go ahead and do that at perfectaudience.com/event and I will go ahead and throw that in the chat while Eric tells you a little bit more about this.
Yeah I think this is going to be an interesting one. We started talking about this weeks and weeks ago with the previous webinar and the subject of KPIs and how to measure return on your investment, and why is ROAs versus some other KPI important, and how do you communicate that upwards if you’re reporting up to someone or how do you, if you’re an entrepreneur and you’re focused on ROI for your business, what do you hone in on. And so we, it’r surprising that there’s been so much feedback that we’ve gotten from that. And we’ve got some pretty in-depth KPIs, little some charts and things that you can take aware that will help you be able to build your own flash reports and help you to know what to zero in on, to be able to get the most out of your campaigns. So it’s gunna be fun and it’ll be something that we can all sort of take and work with when we’re optimizing our campaigns. And we wanna, we’re trying to make successful business decisions.
Yup yeah that’s right. So the link is in the chat if you want to sign up for that you can go ahead, or you can at any time visit our website perfectaudience.com and you’ll find it there. But with that, I think that’s t for today. We’ve had some great questions and engagement. So thank you guys for joining. Thanks for being here with us. Thank you Gaba and Todd and Jalali and Eric for all of your knowledge and expertise. And we will see you guys next time. Thank you everybody. Peace.