How To Test Winning Product Images

Are your product images killing your Google performance? 

In this episode, host Jeremy breaks down how to test product images to boost CTR and conversions. Learn two powerful testing methods, when image optimisation actually matters, and which tweaks can make your ads stand out. 

If your Shopping ads aren’t getting clicks, this could be the game-changer you need!

Get your free 30 minute strategy session with Jeremy here: https://www.younganddigital.marketing/

Scale your store with 1:1 coaching: https://www.younganddigital.marketing/1-2-1-coaching

Transcript

Welcome back, guys, to Google Ads unleashed. Hope. Everyone is doing fabulously this week. So today we are going to be talking about an amazingly interesting topic. But the weird thing is, there's actually not that much content about it out there, and that is how to effectively test product imagery on Google Shopping. So as everyone hopefully knows who is listening to the podcast, there are several elements, and I've done loads of episodes already on this topic, there are loads of elements which you can improve with your product feeds, right? So the obvious ones, which arguably is the most important part, is improving your product titles so that you appear for the right kind of searches. There's absolutely no point. I always like to say there's no point having IKEA titles, right? So something like table Edwin or something like that. Like, no one's going to search for that. People are going to search for, you know, oak desk size, one meter times whatever, right? So that's, that's the way people search. So that is how your product titles should be. So arguably, that's the most important part. Then, of course, price is massively important. You have to have a good price, or your product page needs to be absolutely insane to justify the experience. Otherwise you might be outgunned by competition. Don't forget, it is a price comparison engine. Then the feed quality needs to be good. You know, all of the sort of data that is within your product, feed, color, whatever, etc, etc, all of the things that are sort of standard to every product feed, and something that a lot of people really, actually know but don't sort of think about, is product image. Now there's a couple of things I need to preface before Do you even say that word like that. But anyway, before I actually carry on with this, because, first of all, I would skip this episode or not, like sort of be too concerned if you have 1000s and 1000s of products, right? Because the reality is, you won't have time to optimize your images, right? There is absolutely nothing you can do to optimize your imagery to such a degree that it is some sort of shows some sort of efficacy or effectiveness. Okay? Why is that? Because you've got a million other things to do which are bigger levers than trying to optimize your images when you have 10,000 products, right? I would probably say this is more for, let's say you have a one product store, several product stores, you've got a very set set of competitors, okay, where you can sort of compare what they're doing, what you're doing, etc, etc. And you may, let's say, got a certain number of products right where, over time, you've excluded from your ads because, because they've not been profitable. But then you want to dive deep into the data. Why have they not been profitable? Let's say, for instance, conversion rate was bad. Then probably the product pages is something is wrong with that, or the offer. But if the click through rates bad, but the titles are great and everything, then maybe your image is bad. So that is where you want to swap it around. Okay, so how I'm going to go about this episode is first tell you sort of the methodology of swapping images and when you should be doing it right, what the whole point is, and when you might want to consider it then, like I said, the methodology, like, how do you actually go about it? Because it's not as straightforward as you think. Okay, and then I'm going to tell you about sort of the best sort of tools and other things, how you can test imagery, what type of stuff you can test, okay, so let's talk first about how you will want to test this. Okay, so the reality is that it's extremely difficult to test the efficacy of changing your product image and what kind of impact this has on Google Shopping. I'll be honest, a lot of people are using this and want to be on like social media and so on, and say, oh, you know, we change the product imagery, and we change the results and so on. To be honest, images not the biggest lever, right? So this is like the sort of stuff you want to get into when you want to really deep dive into what's going on with your ads on a very granular level, okay? And the only time you really. Want to be sort of test in an image is, well, like I said, when you have a low click through rate, or when there's so many other things that you've tried in an auction that you will want to then stand out. And the best way I've found to go about this, there's no tool, right? You think maybe some sort of tool out there which does some sort of product imagery testing, right? There isn't so, by the way, call to action. This is an opportunity for every advertiser. If you can find a tool to do AB testing and Happy Days, go for it. I think there's a market for that. But the reality is, the best way to go about is with a manual feed, okay? So actually, there's two ways, right? So the best way is with a manual feed. And you will want to do the following. Most of you will probably have already a just a box standard API or XML feed, okay? And through that, it pulls all of the product data from Shopify or whatever, and you've got your fabulous little listings in Merchant Center, and that's it. Okay, what you can now do is manually duplicate in a Google sheet or in a CSV file the product that you want to test. Okay? So the way you go about this is just to create a box signed a manual feed, right? Put in the exact same title, the exact same landing page, the exact same feed, everything in that regard. But the only thing that you swap out is your product image. The only thing that you cannot use reuse is your G tin, right? So I'll tell you about the limitation of this strategy and how you then go about it is make sure that this second feed then gets approved. Okay? So you then have another product with with a different product image, and then what you do is upload that into your Google Shopping and into performance Max campaigns, right? Make sure that the performance max or shopping campaigns then use a separate or use a functionality to use all products from all feeds, right? So that you know limited to a feed, otherwise you won't find the product. This is something you'll come across anyway when you're trying to test it, and then you just let it battle out, because over time, you will actually see on an item ID level, what product sells more and what product actually generates better. Click through, rates, okay, a lower CPCs. The drawback with that is, if you have reviews implemented for your product feed, and most of you hopefully do, because it's the best way to have social proof on your ads, then Sort of, yeah, you will not be able to do that with your new photo. Okay? So it already has an inherent disadvantage. So to make things fair, what I would do in this period is to implement a discount code in Merchant Center. Why is that? Because a discount code overrides your product reviews, and when it overrides your product reviews, all products are equal. So it doesn't matter if the new duplicated product with a different photo doesn't have reviews, because both all listings have the same discount code under the ad. So it might sort of make your ads overall a little bit less effective, but that way you can very quickly see what kind of product works. So we've done that exact thing with one of our clients. He's very, very keen on dominating brand search at the moment, so it's very specific case where basically what we've done is is already running with a Google Sheet. Anyway, we have duplicated the Google Sheet. We've changed out the product imageries, and we run the exact same products. More all of the products duplicate with different versions. And after three months of running it, we actually finding out if there's some sort of patterns that are disrupting, sort of the, you know,
yeah, if there's any pattern that we can find that is producing any better results, right? So that's how we gone about it. So that's the first way of going about it. So you make manual duplication of the product you want to test, swap out the image, let them run parallel. Battle out. That's the end of it. I would even suggest sometimes, if it's like a one product, one campaign kind of setup is to just make a duplicate shopping campaign, right? You can do that as well, just as fine. The second way of going about is a little bit more hard work. Okay, so this is asynchronous testing. So what you do is you want to define a really specific time frame. Let's say you start on the first of January. You got your product image. You look at the one that you already have. You gather data for one month, let's say, and you've got a pretty solid idea what kind of click through, rate it generates. Then on the first of February, you change it to new. Photo and just observe the impact, right? Does it have a better click through rate as low as CPCs, all that sort of stuff. And then on the first of March, again, a different time frame, you then swap back to the old one and see if it if the effect changes again, right? So if in that time you seen uplift and clicks and whatever. Clearly the product image is better. And if it isn't after that time, then, of course, yeah, it's then there's no point in changing, right? So this, this is the way, the two ways I would go about it. When would I do this? I would do this when I've run out of all options, when all titles are great, when the product feed is fully optimized, when you have reviews on there, when there's pretty much everything that you've done, and you're starting to be looking especially with one product stores or maybe a few product stores, and you're starting to realize that potentially, this might be one angle to stand out on the SERP. Okay, maybe I can now finish off with what kind of stuff I would test if I were to test that. So generally speaking, there's a few tests that you can run. Most of you guys will have some sort of just normal product shot or product photos right for your for your Google ads. So you might have things such as, you know, like a white background or something, what Google loves? Google loves a white and plain background, right? Sometimes you might even have a floating image, like a PNG, right? So a clear, first test that you can do is to add us add a splash of color. That is something I would test, right? So add a splash of color, for instance, some fancy patterns, or some sort of colorful background. That is something how you can stand out in the SERPs, so that you don't even have to do much with that, right? So that is the first one that I would recommend. The second thing that I always would test, if you want to try and stand out, is to test a product photo, versus, like a lifestyle kind of photo. Okay, so you might have a photo with someone actually using the product, or someone with a nice lady in the photo, something I are, right? Just, you know, use your brain, something that could show the product in action, show the product how it works. You can even get creative with this sort of stuff. For instance, we've got one of our one client, right? He sells shoes that make you taller. Basically, as a man, you just be a little bit taller. So what we've got is, like, a little sort of photo, in the photo, in the product image, and it just shows, like the man being a little bit taller, so like a before and after picture, right? So that kind of stuff can work really well. If you very cheeky. You can even test something like a little bit of a promotional overlay, or some something like that. You might get disapproved with Google, but it's worth, worth a crack, if you want, right? Sometimes you can get really sneaky with it, and you can also get away with it. So this is something that I might test as well. The third thing that I would sort of traditionally test would be, then, more or less. How should I say, like a season, seasonal thing, right? So something that we've sort of found can work quite well is if you swap out photos based on seasonality, like for Christmas or something like that. Okay, so that your products look a little bit more Christmassy, and that usually improves performance as well. Now, how would you sort of make those changes? So first of all, either you, of course, do the manual labor right and just edit the photos, you can use product studio within Merchant Center to do like fat variations of your product, sometimes in Shopify as well. So this is something that I would go about it. Or sometimes you can even go to something like, if you use channel, or if you use data feed, watch a similar channel, but it's very, probably very good example you can actually create it. Let's say you're already using channel, right, and you've got an XML feed of all of your products. So this is the sort of stuff that you can do in bulk then, which is one of the few How should I say exceptions. But what you can do is just generate a big old XML and just do an overlay over every single kind of product, or kind of a change to every single product. You can do this in bulk. It's a bit more complicated. But if you want, if you can't visualize what I'm thinking about, you've probably seen it multiple times without realizing big companies do this all the time on Facebook, right? So you've seen dynamic product ads on Facebook, which are nothing else than a big feed injected into meta ad. Ads, and then they run ads off the off the back of that. And you've probably seen it's the same product photos like on the site, but they may have a frame around them or some sort of promotion in the photo. You can do those things and in bulk with a channel or a different other tool. And this is then how you can very quickly create duplicate feeds and run them off. Yeah. So that's how we go about testing product photos. If this is something that you know what? This is something I never thought of, something I want to do, just get in touch. Guys. There's good news. We are expanding as a team. So we have another team member joining the march with another team member joining in April. So there are slots for the agency. So if you are interested in mastering your Google ads and get into the next level, simply get in touch with me. It's young digital dot marketing. You can book a call with me, or you can simply head to LinkedIn. Jeremy Young Google ads. You will find me and let's talk. Maybe I can help, maybe I can't. The least I can do is give you a little bit of advice, and you go away with a good, fuzzy feeling. We both will if you want to do Google ads yourself. We also do one to one coaching, so get in touch. There's some slots now going in q2 our coaching is coming to an end now. In March, new slots coming in q2 and I'm super excited to get that started. This has been Jeremy Young, your personal Google Ads expert, and I wish you a happy and successful, productive week ahead.