The RIGHT Way to Test and Optimize Responsive Search Ads: Step-by-Step Tutorial

Hey, this is awesome. From Grow My Ads. In my last video, I taught you how to create responsive search ads. In today's video, I'm going to walk you through how to actually test responsive search ads. So there's a lot of different ways of going about testing different ads. I'm going to give you four top ways of testing responsive search ads and I'm also going to give you my favorite way of doing it. So let's go ahead and dive in. Okay, so the easiest way to start with ad testing is actually just going into your current responsive search ads you have running and cleaning out the low quality descriptions or headlines. So I'm inside of an ad group and I have the ads up. If you hit View asset details, this will take you to an actual responsive ad report and here they will give you performance labels and in this case most of them are labeled good. Thanks Google. A lot of good information I can do with that. But in most cases you will see several that are tagged low. So in this I've got for this ad a lot of descriptions and headlines added. But right here, this one is tagged low. So the easiest way of continuing to just improve your ads and test ads is go in and clean out all the ad copy that has a performance label of low. So in this case I would come in here, this conquer your awkward space headline has been tagged low. So I would go in, remove that headline and actually create a new headline to give Google for further testing of the ad copy combinations. Now one note. Make sure for this headline I would actually have an ad testing sheet where I have here are headlines that we've tested here. Here are headlines that are doing well are labeled you know best. So this is a description here. Get swatches. Try risk free trial flexible for giving family friendly sectionals. This has a best performance label. I will have a sheet with all of that that I've tested with description and headlines. So something like this conquer your awkward space which has been tagged low. I will put that in my sheet. Hey, this has been tested. It did get a low rating that way. I know don't add conquer your adwords space to any future ad test. And this is feedback I give back to the client to make sure don't be using this ad copy in any other marketing channels whether that's, you know, Facebook or TikTok. So here's a headline that just did not do well on Google. All right, testing strategy number two is simply adding another responsive search ad into the ad group. So here I'm inside of an ad Group, it only has one ad running. I could simply go in, create a new responsive search ad which again I have a full YouTube video on creating responsive search ads that I will have in the description below. I go in, I can create a brand new responsive search ad and then I can test that against this current ad here. Now a few things and I talk about this. In that other YouTube video on creating responsive search ads, you have the ability of adding 15 headlines and four descriptions. I hammer on my last video about making sure each of those headlines are unique and impactful. A lot of people will add similar phrases of words such as save 30% off or 30% off, right? And they will take two headlines saying the same thing. That doesn't make sense. That's just, you're just wasting your time and the testing of the machine learning for these ad combinations. So if you are running a current responsive search ad, what I recommend is testing themes. So if I were to build another responsive search ad to be running to inside of this ad group, I would have a theme that I'm trying to actually test against. So I'm doing big tests, not some small one headline test. I'll talk about how I go about doing small headline or call to action tests a little bit later in this video. So what types of themes could you be testing? Well, price, selection of product features and benefits. And in some cases I may have one ad that's calling out certain features and benefits and then another ad that's calling out another set of features and benefits. I want to see which set of features and benefits do better or have better performance for us in our ad testing you also could do comparison if, if your keywords are based on maybe you know like competitor or alternative type terms, you could have comparison in your ad copy and testing that against. Again another theme in there. Another theme that you can test that I like doing is pinning headlines versus non pinning. And if again my, in my last video I talk about ad strength and how it's totally a nonsense metric that Google created, it does not impact the performance. So pin and non pin is something that I like to test as well. I'll have a full built out responsive search ad with 15 headlines, four descriptions. Then in my next ad I'll just pin the top headlines and the top descriptions and run that against a full built out ad with all of the assets and see which one does better. In some cases my pinned ads actually perform significantly better than a full built out responsive search ad. The issue with running two ads inside of one ad group is you have no control of how Google is actually going to divide the impressions. So the other way of going about testing ad copy is and controlling how impressions get split between the ads is through a campaign experiment. So I'm at the campaign level of or actually the account level of an account. And I'm going to go down to this section called Experiments. And here I'm actually going to create a new experiment. Now they give you several options and they actually do have one called Optimize Text Ads. I'm holding that off for last, so don't worry about that one yet. But what we're going to do is actually a custom experiment and then in canned campaign type is search and then you name the experiment. So I'm going to do deepsea Features one. So this tells me what I'm actually testing. It's a, a feature set and then I'm clicking the Deep Seat campaign and then it will add this to the experiment name here. So Deep Seat Features one. So I'm going to hit save and continue. Now I'm essentially inside of this campaign. So if I click here, this is exactly how the Deep Seat campaign is. I'm building a experiment shell of it. And so I'm going to go in and I've got an ad running in here. And this campaign is pretty targeted towards just this one specific set of sectional sofa terms. So here I have an ad here, right? I'm actually going to pause that ad. So I went and I paused the ad that is running in my original campaign and I've created a new ad that is now in my experiment campaign. So you can see draft change here, check mark. So this is the ad. This is the only ad that will be running inside of here. Then all I have to do is hit schedule. It allows you to select up to two goals. So so here I actually want my conversions to increase and my CPA no significant change. So I want more conversions for the same cost per acquisition on this. Then this is the nice part about doing the experiment. Experiment split. Select a portion of your original campaign's traffic and budget for this. So 50% is trial, 50% is base. And then you also have the option of doing search based or cookie based. Search based usually will go quicker. So here's this little snippet. With search based experiment splits, you can assign users randomly. You may get statistically significant results faster than with a cookie based split. I've never actually had any major issues clicking either of these. So it defaults to cookie base. So I normally just leave it at Cookie base. So go ahead and just leave it at cookie based as well. When you're building this, then start date, you can't start it the day of of building this, so it has to be the next day when you build it. And then you set the duration of time you're willing to allow this experiment to run. You obviously can shut this thing off whenever you want. But here I just want to set to 30 days and then this Enable Sync, this is a very nice feature. If I make changes to the original campaign, it will automatically make those changes to this campaign. So if I went in and made a bid change to that original campaign, well that would screw up the entire experiment of ad testing. Then potentially if I'm running one different bid strategy now compared to this campaign that I'm trying to just test an ad difference. So it would make this Enable Sync feature allows for it to make that same bid change to your experiment campaign. So here I have this enabled on. And again, here's their information. Google's changes made to your base campaign will automatically sync to the trial campaign so you don't have to make changes manually. So nice feature. Then boom, you hit Create Experiment. Then you can go into this while this is running. Obviously this experiment has no data to it yet, but this will actually give you full reporting on the data between the original campaign or the base campaign and then this experiment campaign or the trial campaign is what they call it. Also at the account level you will now see your experiment and it is labeled. So again it's, it's the Deepsea campaign that I'm testing the 5050 split with. But it gives the experiment name here. So I named it Deep Seat Features one. So you see this here, the icon will be different as well. It's more of this sort of like lab testing icon, which is kind of cool. And then you can also see data at the account level looking at your campaigns without needing to go inside of that experiment report. If, if you just want to look at it from this level, if you want to pause, you can just come here and actually just pause it out. Or you run, you go to the experiments, you go into the experiment you're running and then here it says cancel because this is, hasn't went live yet. But if it, if it were live, it would have the option of just pausing the experiment out completely. Okay. My number one way of testing ads is actually through a feature called Ad Variations. So I'm inside of an ad group and I'm going to go to the blue icon to create a New ad. Except now I'm going to go to this button here for ad variation. An ad variation allows you to use rules to test. You could do this in multiple campaigns or one campaign and you can test different headlines, call to actions, final URLs and it does it in a very clean manner. So in this example I am going to only focus on this deep seat campaign. And now I need to target an ad. So this is going to be the targeting feature of what I'm looking for. So I am looking for any ad that contains conquer your awkward spaces and let me fix that. So remember earlier in the video, conquer your awkward spaces was a low performance headline in the performance labels of the responsive search ad reporting tool. So I want to find any ad that is using this headline. So that's the sort of targeting feature here. And that's my base campaign or base ads that I'll be testing against my new ones. Now choose the type of variation you wish to apply to your ads. So here I'm actually going to do the same thing and I'm probably. I knew I was going to spell that again incorrectly. Maybe after this video I'm just going to work on spelling awkward correctly when typing it out. So I'm looking for choose the type of variation you wish to apply to your ad. So I'm finding the text conquer your awkward spaces and headlines and and now I'm going to replace it with a new headline. Okay, so now I'm going to have this text conquer your awkward spaces replaced with perfect for small spaces. This is what I'm testing against. I'm going to hit continue. This is what I like. So in small spaces is the name here. Start date, duration. I'm just going to do 30 days. And then here's the nice feature, experiment split. Split. Experiment split is the percentage of your campaign's budget that's allocated to the variation and percentage of auctions your variation is eligible. So here I just want a 50% split between those two. Then I'm going to hit create variation. Okay, now at the experiment tab, add variations. You see my variation small spaces here. I don't have enough data here yet, but once those ads start running against each other, you would see here's the modified ad that would be the, the small spaces headline that I'm I'm testing. And then the original ad, that would be the ad that has conquer your awkward spaces headline. Then it would give me data for whatever time frame I'm looking at. Since it started. It would give me a click impression. Click through rate all the data that you would ever need to look at. And then you can see now the differences between testing those headlines out. So with ad variations you have find and replace. Here you can find text and replace it with any other text that includes headlines and descriptions. So here if you just want to test a specific headline like I just showed you in the ad variation I built, or you can do that for paths, URLs and descriptions. You also have the ability to just update URLs. This is perfect for landing page tests. This gives you a nice clean way of testing different landing pages. So you can just swap out the URLs and a nice 50, 50% split with an ad variation. And you also can update text. So here you're just kind of editing, editing, removing or pinning text with inside of your ad. So here's the ad remove pen actions that you can do. You can do that for headlines and you can do that for descriptions. So this gives you the ability to test. So simply if I were to test what would pinning a headline due to my ad. Well, you could actually come in here, hit pin and then pin this to position one. You could also pin it to position two or three. You could also do multiple headlines being pinned and test that against the current ad for your ad test. There you go. You've got the top four ways of testing responsive search ads. Make sure to also check out our other videos related to responsive search ads. Those will be in the description below. We have how to create responsive search ads, how to create search campaigns which you would need in order to even get to the ad creation part. We also have videos on ad copy and creating ad copy that's going to to get you higher click through rates. All of that in the description below. Make sure to check those videos out and I will see you on the next video.