Google Responsive Search Ads: How to Automate Ad Copy A/B Testing
No matter how much data you have about your clients or customers, you can always use more. How do they like to receive information? When do they peruse the web? When do they make decisions? What subject lines appeal to them? Are they more emotional or analytical in nature? Do they respond better to humor or statistics? There’s just so much you want to know, and this is especially true with responsive search ads.
On a more general level, of course, you probably already know much of this about your audience. But to what extent? If you’re like most companies, you could stand to get significantly more granular. For instance, if you send out an email with two different subject lines with a one-word difference, can you predict which one they’ll choose? Probably not. Yet it could be the difference between getting a click and not, making a sale and not, gaining a client for life…and not.
In a nutshell, A/B testing lets you find the optimal approach in any situation so you can replicate those results next time and refine them for ever-greater conversion. When you do this over and over again, for every campaign—while simultaneously optimizing bidding and analyzing your results—the outcomes are pretty amazing.
Yet sadly, according to a survey of 800 digital marketers, less than 40 percent of companies use A/B testing in their campaigns.
Well, we should clarify: it’s sad for them. Really, it’s a positive for you. It means your competition is seeking “good enough” results, making it easier for you to seek the absolute best ones and rise to the top of your field.
But running A/B tests is hard, not to mention time-consuming. When you include writing new ad copy, loading those ads, optimizing them and analyzing them, there’s a lot more to do. Even writing copy can take its toll. Once you write several headline and body copy versions, then you have to mix and match them to create multiple combinations, load them in, etc.
If you’re getting a headache right about now, take a deep breath. Now, what if we told you that this approach won’t be necessary for much longer?
Enter Google’s new responsive search ads, an answer to the prayers of any dedicated online advertiser. Responsive search ads are Google Ads newest ad type, launched alongside their latest overhaul of the interface. Meant to help further automate A/B testing and PPC ad optimization, responsive search ads are adaptive, delivering a more relevant message to searchers based on their queries.
With this ad type, you provide multiple headlines and descriptions when you create the ad. Then Google Ads does the heavy lifting, adjusting your ad’s content to match search queries, testing different combinations of text and discovering which is the most relevant for your audience.
Let’s talk about what these ads are and what they can do for your business. We’ll also take a look at how you can use them to automate ad copy creation and A/B testing, as well as several best-practices tips for integrating these into your marketing routine.
So if you’re ready to save time, learn more about your audience, outstrip the competition and increase your conversions, get ready to take notes. Responsive search ads are about to be yours.
Features and benefits of responsive search ads
In the past, Google suggested advertisers create multiple expanded text ads for each ad group to help advertisers test and identify the most effective ad copy for your marketing goals. It was time consuming, but worthwhile if you wanted to get more reach for your ads and understand how your audience reacts to different messages. Now, responsive dynamic ads are positioned to make this “best practice” obsolete by allowing you to test ad copy variations all in one place.
While they’re similar to expanded text ads, responsive ads have some key features that set them apart:
- They can show up to three headlines instead of two
- They can show up to two 90-character description fields (vs. one 80-character field for expanded text ads)
- You can include up to 15 headlines and four descriptions to combine for a single search ad
- Most importantly, Google Ads automatically rotates combinations based on search queries
By providing a series of headlines and description options to Google Ads, users save time A/B testing different ad element combinations. This also has a secondary effect of giving you more opportunities to compete in auctions as your multiple headlines and descriptions can be relevant to more search queries. Over time, as Google Ads shows your different ad text combinations, it will identify and automatically prioritize the best performing variants for you.
If those benefits aren’t enough to convince you to try responsive search ads, just remember they’re allowed more real estate in SERPs than any other ad-type. What’s not to love?
Like with every ad type, there are of course some limitations when using responsive search ads. Advertisers who are already less inclined to give over full optimization to the Google Ads machine won’t be chomping at the bit to try out this ad type. It appears that responsive search ads also don’t fully support ad customizers, a feature many advertisers love to use to manually tailor their ad message to user search queries. But this might change in the long run.
With all that in mind, it’s still clear that responsive search ads can almost certainly help increase your ad group performance overall…if you take full advantage of the benefits.
7 tips for creating your first responsive search ads
Ready to start realizing the many benefits of responsive search ads for yourself? Here are seven tips you should follow to ensure your responsive search ads meet qualifications, help you automate your ad text optimization, and more.
1. Use a lot of headlines
When creating your first responsive search ad, you’ll need to provide a minimum of three headlines and two descriptions to rotate. But Google Ads recommends you provide at least five different headlines for your responsive search ads to increase the chances that your ad shows. The more options you provide, the more opportunities there are to appear for search queries and optimize your ads for the most relevant message. Using as many as eight or 10 headlines would be ideal to get the most benefit out of this ad type.
2. Avoid redundancy
No matter how many headlines you include, make sure the variants are sufficiently different from each other. If you use too many of the same or similar phrases in your headlines, the system will have more trouble generating ad combinations.
Here’s an example of a poorly optimized responsive search ad because of redundancy issues:
Similar headlines like “Fashionable Women’s Shoes” and “Trendsetting Women’s shoes” make it hard to generate ad combinations and ultimately limit the benefits of this ad type to reach a larger audience with your diverse ad descriptions.
Try varying lengths of the headlines you create and add at least two distinct descriptions. It is possible for responsive search ads to show up to two descriptions at a time, making it even more important to avoid redundancy.
3. Ensure all possible combinations make sense
It’s possible for your headlines and descriptions to appear in any order, so you need to make sure all combinations make sense when viewed together. Google Ads recommends writing your first three headlines assuming they would appear together in your ad. Does the message make sense/avoid redundancy?
Here’s an example of a responsive search ad that’s well optimized:
All headlines and descriptions are unique and make sense no matter how you pair them up.
4. Highlight features and benefits
Your responsive ads shouldn’t be about using different words/phrasing to deliver the same message for your headlines and ad description. Instead, focus on illustrating different features and benefits of your product/service to see which are the most effective at driving clicks.
For example, let’s say you’re advertising international health insurance. “Free quote,” “global coverage” and “Plans start at $199/month” are all examples of benefits you can test as part of your description variations.
5. Limit keyword insertion
Google Ads recommends you include a target keyword in two of your headlines, but also that you have at least three more that don’t include any keywords. This has to do with (you guessed it) redundancy issues. Including keyword insertion in too many headlines can lead to redundant text in the ad.
Here’s an example of a poorly optimized responsive search ad because of this problem:
6. Pin important information
There may be some important information that you want to appear in every ad text combination. In order to include this text while avoiding redundancy, you’ll want to pin it to the ad. You can pin text at Headline position 1 or 2, or Description position 1. Make sure the text you pin is less than 80 characters long.
For example, if you need to include a disclaimer in all your ads, just write it in one of your descriptions then pin it to Description position 1. This will ensure that every ad includes the disclaimer in that part of the description.
That said, you should only pin information that you really need to appear on every ad—otherwise, it limits the number of headlines or descriptions that can appear for a search query. For that reason, Google Ads doesn’t recommend pinning for most advertisers.
7. Monitor performance
Google Ads may use automation to help advertisers create, show, and optimize their ads, but that’s no reason to not monitor performance metrics yourself. On the Ads & Extensions page, you can see performance metrics for each of your responsive search ads, including all the standard stats you receive for other ad types.
These statistics are performance totals of all the combinations of headlines and descriptions for that particular responsive search ad. Unfortunately, there’s currently no way for advertisers to see how individual ad text variants perform against each other within the ad. They’ll have to trust that Google Ads is doing a good job of optimizing them.
One thing you may want to look at is how your responsive ads compare to your regular text ads. Assuming you’re fully utilizing the features of responsive search ads, Google Ads should help you create a more visible and effective ad based on query relevance and audience behavior.
Machine learning and the future of advertising
Google Ads has changed a lot in recent years, and continues to roll out new features and ad types that rely heavily on machine learning and automation for optimization. Responsive search ads are just the latest addition that require users to relinquish more control in order to benefit from insights and optimization capabilities.
There will always be traditional advertisers among us who love nothing more than to analyze performance and manually tweak ads themselves to perfection. There’s no denying the power of the human touch to create a highly targeted ad that speaks to audiences on a granular level.
But the truth is most businesses today don’t have the time or resources to manually target and optimize their ads at scale. And as more big data insights are available to help improve ad targeting in real-time, it would be unwise to overlook this resource and risk falling behind the competition as an advertiser.
There are already a number of bid management tools and predictive advertising technologies that make it simple to synthesize and automatically derive insights from consumer data to optimize your bids and minimize wasted ad spend. Advertisers who embrace the power of machine learning and automation are already benefiting from these technologies to outbid and outperform their competitors.
Google Ads’ machine learning and automation features are no different, and they are coming at the perfect time for most busy advertisers. Responsive search ads are just an example of the direction things are heading for ad creation, optimization, and bid targeting. Advertisers who choose to embrace and fully utilize these technologies are the ones who will best illustrate their benefits early on. In the long run, automation and machine learning are the way of the future in advertising.