Surfside PPC Podcast

What's up everyone? Welcome to Surfside PPC podcast episode number 17. So first to get started, we are going to start running some campaigns in here and follow along on the podcast with some of the campaigns that I'm running for Surfside PPC. Some of that will also be premium content, but obviously if we'll release a lot of that for free directly through this podcast. So today's topic is going to be AB testing. So really just one question and really just want to cover Google Ads AB testing. So you can AB test in Google Ads, meta ads. There's really only so much you can AB test and I would say my thinking on AB testing has changed a lot over the years with you know more AI and quicker learning through Google ads and Google being able to figure out what is actually driving results much quicker. So let's get into the video. I'm going to put in the question now and you know after the after the question I will answer it and go through all the different things related to AB testing or split testing in Google Ads. What are the best ways to split test in Google Ads? Can I run a true AB split test effectively? What are things that can test to optimize my performance. Okay, we have our question and essentially it's just what you know how do we split test effectively to improve our results and ultimately what we're trying to do is look at some of the different key elements in here and you can almost go through some of these and say okay we have ads we can split test between ads now there is I'm going to use the word split testing a lot here and I would say it's not necessarily a split test what I'm talking about I will give you examples of split testing but ultimately the way that I view Google at this point is that Google wants us as advertisers and you know businesses to basically give Google a lot of assets to work with. Assets are going to be your headlines, your description lines, your images, your site links, your callouts, structured snippets. Basically, anything that's useful you should have as an asset. You can also use headline assets, description line assets. You can do new headline description line assets that actually match ongoing promotions you're running. You can use promotions. You can import prices for the actual, you know, product, services, that you have connect your Google business profile or Google business profiles. So all of this helps Google really build your advertisements with this set of assets. So I actually don't view it as much as you know straight up AB testing and I don't think that's how you should view it either. I think in your assets performance here what Google's looking for and I don't have anything in this last 30 days so there's not going to be anything showing but in your performance you'll actually have a breakdown of all of these different types of assets that we have and you know my thought process is if we have zero YouTube videos compared to the advertiser who has 50 YouTube videos and it's a big difference obviously but that one with 50 YouTube videos might be saying yeah we're running ton of video ads three or four of them are doing really well this other one who has one or two or zero is getting nothing in terms of video so just a whole channel in terms of YouTube Google display network video that's not being used as well and with more audience data in Google ads I do think kind of understanding the whole funnel is really important and that's why getting too caught up in AB testing and looking at direct performance from one to the other isn't really the right way to look at it. But what I prepared is this this document of some of the different things that we can actually split test. So this is specific for my episode 17. Pretty cool. I could put together these quick HTML pages and start uploading them, you know, based on the podcast content and everything like that. So Sside PPC podcast episode number 17, some of the different things we can split test. So starting with ad copy, this to me the best way to do this isn't to say, you know, let's run one ad with 15 headlines and, you know, 15 or four description lines or, you know, in a performance max. Let's just make sure we fill everything out. I think where you're much better off is if I'm running ads for Google Ads Consulting and we go into that ad group. What makes a lot more sense is me running multiple different ads throughout this, putting my pricing in there, putting different ad copy in there. I think this was also, this will go into our next thing. This is also testing landing pages. So, a lot of my testing actually happens within the ad. in the ad groups by using a couple different landing pages and using different ad copy in in the ads. You can run three responsive search ads per ad group. So, if you do have really tightly themed ad groups and you're running different ads throughout those really tightly themed ad groups and then you you're also testing a couple different landing pages throughout them, basically you're giving Google different things to test as far as where traffic's going, what people are searching. I guess not really what people are searching in this case, but what people are seeing in terms of the advertisement. So, after the search, you know, that whole experience is really what we're trying to optimize. To me, it's not saying, "Let's do a direct AB test." You can do, you know, one ad, duplicate it, change the final URL for that second ad. And I've done that before, and that's that's really a true AB test for landing pages as long as they're getting the same amount of clicks, impressions, and you have a good enough set of data to do that. It is much easier to AB test in large accounts that are spending a lot, and it's a lot more difficult to AB test in accounts that just don't spend nearly as much. So, ultimately, what we want to do is look at Can we create more sets of headlines? Can we pin specific headlines within our ads? So, if we go in here, can we pin specific headlines within our ads? Do uh dynamic keyword insertion ads work a little bit better? Can we use location insertion? Can we use a countdown? So, maybe I have some limited time promotion. Those are the types of things. And you can see I have pinned headlines in here, which I do sometimes just to make sure I my it fits exactly what people are searching. But ultimately, this these are the different types of things that you can try to test. is just different headlines. Scroll down here. Different description lines. Upload as many images as possible. I'm not following my own advice. It should have 20 there. Business name, business logo, get as many site links as possible. Now, this was a pretty quick campaign I built. I'm actually going to create a couple new campaigns. And that that I'm going to go through on like a whole video series. So, that's that's what my plan is to do that on a whole video series to to make it almost like a a free miniourse I'm going to release through Google or through YouTube, excuse me. But for this one for split testing. It's really, you know, creating more ads is a good thing. Using additional landing pages is a good thing as well. If you have two pages on your website that can fit for an individual service, like if you're, you know, for local service companies, sometimes your homepage can convert pretty well and then you have a specific page for this is the actual service we're providing in this city. And that's actually also a really important thing. So, those are going to be the two things that really you can test the most. Landing pages, I think, also goes along with offers. So, with landing pages, It's really can we do two sets of offers? One with a promotion, one with just, hey, give us a call today and see if there's any difference with those. And really, you're just looking at conversion rates for landing pages. On the ad side, you're looking at click-through rates. You're looking at which headlines, description lines actually lead to conversion. So, you're looking at really improving conversion rates and click-through rates with your ad copy and your landing pages. And as we scroll down here, next is going to be campaign level settings. So, bidding strategies can always be a big test. Maximize clicks in the long run is not going to perform better than target CPA or target return on ad spend. In the short term, it can help you gather data and target some of the right keywords, but really with campaign level settings, you want to get to a smart bidding strategy of maximize conversions, target CPA, maximize conversion value, target return on ad spend. You can test different bid types. It is getting harder to test bids at times because it's sometimes you raise your target CPA by 20% and all you see is your cost per conversion goes up and there's really no added volume coming in. Sometimes you lower it by 20% and then you see volume drop. Cost per conversion might go down a little bit, but then the client's like, "Why are we getting less leads?" So, you're really trying to balance, you know, what the right target CPA setting is. That's why it's good to actually have a good communication setting with your client where they say, "Hey, this is what I'm willing to pay per conversion. This is really what I would like to pay per conversion." The other thing is testing your match types. So, as far as keyword targeting, broadphrase, and exact, I do think keyword targeting, I mentioned this in an earlier podcast, it's I mean, it's a slowly dying, a slow death really. We're going to have keyword targeting, but it's going to turn into using more broad match and then just making sure you have audience data and conversion data. So, Google can ultimately find new customers and find, you know, existing customers that might be repeat ad rotation settings, not something I change too often. Budget allocation is important. If you're running two, three, or four different campaigns, where does that money go? What is actually driving the results in the account? And those are all really important things to to understand the success in your account. Test who should see the ads. So, this is one of those where audience, again, I'm not testing as much as an individual audience as much as on building these big audience signals and basically saying what is working the best. I think the most important thing is making sure that you upload customer match segments, customer lists, which is your customer match segments, but you know it could be email lists, it could be you know connecting a YouTube channel, it could be connecting the website, making sure that every single last piece of these are the people that are visiting my website and interacting with my content, making sure that they're all seeing your advertising is really important. So testing the channel mix inside Google. I will say I'm running a lot more performance max than I used to. And I, you know, you wonder what is going to happen with search, performance max, and different campaign types in Google Ads longterm with AI. I think search, you're always going to have search, and that's going to turn into more search with AI max. I do have a really good strategy for that. And then search with performance max. And my plan is to essentially try to build my business around some of these different strategies. And that's been part of the reason why I switched to Shopify, just because it does take into account, you know, people purchasing on the website and you know some of that information really really useful makes it a lot easier that way. So testing your campaign types testing you know we went over budget allocation standard shopping versus performance max. I will say I'm seeing really good performance out of performance max recently. So it is one of those campaigns that's improving with time and it's one that I'm running in a lot more accounts because it drives a lot of clicks. Search can sometimes feel very limited and you know it's like one of those where if you're running ads for a roofer and they're like hey we're you know we're trying get a couple jobs. You know, it's not like they need 10,000 jobs unless it's a massive roofing company. But a roofing company in a smaller market is, you know, you're trying to just keep your your crews busy with a job every, you know, every time a job ends, you got a new job. If you have multiple crews, obviously you have multiple crews running. But sometimes on the search side, it's like, okay, I'm spending $50 per click. I'm only getting, you know, 20 clicks a month and or 30 clicks a month. And it's like it still costs me a couple thousand dollar. And it's, you know, sometimes you're like, I feel like I can get more out. of my local market than just trying to go after these really expensive keywords. So testing performance max alongside search. I almost wonder if we're going to be running multiple search campaigns, multiple performance max campaigns. It's been kind of a little bit more of my strategy recently, but it's it's an ongoing see what works and and keep going towards it. So testing supporting elements. This to me the assets site links call out structured snippets call assets just get them all up there. Put do a lot of them. There's no downside to running multiple structured snippets running 30 40 call outs running 10, 15, 20 site links if they make sense. So there's no problem with running supporting elements and Google is also doing a better job of actually taking those supporting elements and making sure that they match the actual advertisement itself. Just the improvement of AI very quickly test where and when ads should appear, geo targeting radius for specific locations, ad scheduling, day parting. So this is one you could always test and just see, okay, how do we work on different devices? Are there days that are performing horribly? Are there locations dragging down our budget? Again, smart bidding does fix a lot of this, but it's one of those where testing a lot, understanding what's working does help you with really getting the most out of your campaigns. So, starting with the highest leverage test, high-spend campaigns first, high intent services first, pages with enough conversion volume, tests with one variable clearly isolated. I don't do that nearly as much as I used to. It's just the changes within Google are I don't want to say throw a bunch of things in the wall and see what sticks, but really make smart decisions in terms of what ads we're creating, creating different headlines, description lines, testing different offers, testing different landing page designs, those things are going to help improve everything. So, how to test well one variable time, real control, give it enough data, match metrics to the test uh to the type of test, check every combination, document what you learn. So, practical framework, test the ad message and call to action, test landing page friction and trust elements, bidding, match types, audience layers when you're in here in the Google Ads campaign. So, your ads are going to be a big portion of where you're actually testing. And then the other thing is just going to be okay under for insights and reports. Let's see when and where ads showed how they perform on devices. Where ad showed isn't so basically for match locations. Maybe we'll have it here. Yeah. So we could see match locations to say okay where what cities are we showing in there's not enough data for this to make a real determination. But you know if you learn over time hey this is oh wow Lakewood Ranch is driving really good results for us. You know if that's the case then you know that's something that you could potentially go after. So we'll come back to all campaigns over here and yeah. So so essentially most important thing is when you're testing, it's more about feeding in really good conversion data and first-party audience data into Google Ads and then using the right targeting and just ultimately as you get conversion data in, testing all of these different variables throughout the process. And that's just a matter of analyzing results, seeing what's working and you know using smart bidding strategies to ultimately automate some of this in terms of you know there's the manual portion of Google ads with changing bids and doing all that on a granular level is done and that's a good thing. So you're able to kind of do more within an account in terms of, you know, what you're targeting, what you're doing. And I do think where Google might shift is to more of a creative, you know, being able to create videos, images, you know, even put out text ads, offers, and really line those things up as quickly as possible and as effectively as possible for your business so that you're actually staying on target with like a content calendar and some of those types of seasonality that you're running. So, if you have any questions about this, please leave in the comments section. This will conclude Surfside PPC podcast episode 17. Best way to support the podcast, share, like, subscribe, leave a comment, leave a review. All those things would be really helpful. But I appreciate all the support. Trying to grow this and get some more listeners on the podcast, regrow the YouTube channel. So, we'll we'll be active and and pretty busy with creating a lot of content. So, again, thank you for listening and start putting up some more assets and testing every single last possible thing you can in Google Ads to get the best results.