Presented by Consumer Acquisition
We’ve all seen big changes in user acquisition (UA) advertising in the last few years. Artificial intelligence (AI) is automating more and more of media buying, as the best practices outlined in Facebook’s Structure for Scale framework demonstrate.
As a result, the AI optimizing ad platforms has gotten better and better and it’s leveled the playing field for advertisers big and small. Now, creative is the key differentiator driving UA.
And, while quality and volume of creative are key factors in success, we’ve discovered that they’re not the only factors that drive UA success.
Monitoring creative trends and doing in-depth competitive analysis are a must within any UA or creative team today.
But, cutting edge advertisers don’t stop there. They also incorporate player profiles and user motivations into their creative strategy.
Player profiles are actually an old idea in game design, but they’re a very new idea in advertising. As of late last year, only about 10% of the UA teams we reached out to were using player profiles, and they tended to be the most elite advertisers.
Understanding player profiles and emotional hooks may change how you see ads forever. It’s a wickedly-effective framework for motivating users, but also a great lens to use as you analyze competitors’ creative.
And all of this — from creative trends research and competitive analysis to deep dives into player motivations — is further fueled by the power of creative testing.
Through extensive research, we’ve developed a new approach to creative testing that solves the adage of “why the control always seems to win,” which we’ll reveal further in this article.
Why app advertising is all about creative
Creative is any advertiser’s best opportunity for a competitive advantage in social advertising, and it has been since 2019.
- Expensive, third-party automated adtech platforms aren’t as much of a competitive advantage as they used to be. Facebook and Google’s platforms have integrated the best features of those tools and offered them as standard features to campaign dashboards. All that fancy adtech is now available in your ad accounts, for free. However, if you are spending over $ 250,000 per month or running multiple titles, an inexpensive ad management platform like AdRules can greatly simplify your workflow and increase productivity.
- If you’re spending less than $ 250,000 per month, you may not need expensive, high-end ad jockeys. Automation features have taken a lot of the controls away from UA managers and ad accounts with smaller ad spends have become easier to manage. However, if you are spending over $ 250,000 per month or across several networks, you may find it helpful to look for inefficiencies in automated solutions. There are still many levers where a rockstar UA manager will outperform an algorithm.
Those two factors alone leave creative as the most critical competitive advantage. Because you can now run powerful campaigns without adtech, and lower-spending advertisers can actually get good results, two key competitive advantages are gone. Creative is now critical.
The data backs this up. After managing over $ 3 billion in ad spend, we’ve seen creative prove itself over and over as the single-best driver of ROAS (return on ad spend).
Across our portfolio of accounts, we’ve found that 75% of the ROAS gains we’ve generated for clients can be directly attributed to creative optimization. Compare that to 15% of the ROAS gained from audience expansion, and 10% of the ROAS gained from media buying tactics and optimization.
Audience expansion and media buying tactics are still fantastic tools for the Rockstar UA manager, but creative wins hands down. Creative is king.
But not all creative is created equal.
Just pumping out more and more creative variations from winning ads won’t give you the advantage you need. You need high-performance new concepts and control-beating fresh ideas. Those can be hard to find. Only 1-3 ad concepts out of 20 ever beat the control.
To know which creative assets and elements deliver big results, you need an efficient, accurate testing system. Without the efficiency, it’s easy to blow your budget on testing variations that end up generating a disproportionate percent of non-converting spend.
This can be more of a challenge than it might seem. Beating the control is surprisingly hard…but more about that in a moment.
Creative tunnel vision and how to overcome it
Even with a world-class testing methodology and creative optimization process, your creative team needs fresh ideas. They have to be aware of all the trends in ad creative that your competitors know about and are already taking advantage of.
To increase the success rate of creative testing while eliminating creative tunnel vision, we reviewed thousands of Facebook and Google ads to create a “visual taxonomy” of creative trends. We use those trends to generate a list of fresh creative concepts informed by competitive ads, player motivations, and advertiser performance.
As Picasso said, “Good artists borrow, great artists steal!” Once you identify your competitors’ best ads, you have an endless supply of “tested” concepts. However, we have found that analyzing competitive creative, as challenging as it is, is not a UA or Design function. It appears to be a cross-disciplinary exercise.
At this particular snapshot in time, in early June 2020, we’re seeing trends like the following drive ad performance:
- Connect the Dots
- Noob vs Pro
- Hidden Object
Again: These trends come and go faster than fashion, and you need to be in touch with the heartbeat of online advertising across thousands of advertisers to even recognize the trends in the massive number of ads launched every day. Then, once you’ve recognized them, you need to adapt these creative trends so you can use them in your own ads. Get it all right, then test it with a super-efficient creative testing system, and you win the prize: New ad creative that beats the control.
It’s all a little bit like baseball…you’re only as good as your last hit.
Trying to keep up with trends that evolve so rapidly is hard. Trying to do it while you maintain ROAS is even harder. And expecting the same team of people to do it over and over again, week after week is damn near impossible. Eventually, almost covertly, creative teams learn what works and what fails and then “play it safe” with their creative. They’ll default back to their greatest (and safest) hits of the past and become just a little bit too cautious about testing any out of the box ideas that might not work. We call this “creative tunnel vision.”
Unfortunately, while many out of the box ideas don’t work, some will. And one or two of those ads — those unlikely, out-of-the-blue winners — could have delivered you a 10-40% leap in ROAS.
But if your team never develops those out of the box ideas, they’ll never test them. You’ll never know how much revenue you’ve missed.
This is why an outside team can be so helpful. And why competitive analysis should be a core skill of somebody in every creative team.
Competitive analysis has two key benefits:
- It lets you peer into an almost endless library of tested concepts (tested with your competitors’ ad budgets, not yours).
- It increases your success rate for your new concepts, so your entire testing program involves a lot less risk.
Think of all this shared knowledge as the “wisdom of creative crowds.” Fortunately, we have ample resources to make competitive analysis efficient. Creative research tools like AdRules+SensorTower are great, as are SocialPeta and AppAnnie and many others. There’s even Facebook’s ad library, which is completely free, though it only lets you see which ads your competitors are running, not how those ads have actually performed.
If that’s not enough, you can also check out our resource of over 500,000 competitive video ads from Facebook and Google AC advertisers. Just remember to check competitive benchmarks data so you can recognize what’s breakout performance and what’s just average for that genre.
Creative trends & recommendations
So creative trends change fast — but they also vary dramatically across genres. While it’s good to know broad creative trends, you’ll also want to hone in on your particular niche and overlay all this with actual performance data. We want to chase trends that boost ROAS, not just chase every trend that comes along.
To get you started with this, we’ve developed seven genre-specific creative trend reports. You can find individual creative trend reports, such as for Social Casino Games.
Player profiles: Market segmentation & user motivations
Monitoring creative trends and doing in-depth competitive analysis are great. But cutting-edge advertisers don’t stop there. They also incorporate player profiles and user motivations into their creative strategy.
Player profiles are actually an old idea in game design, but they’re a very new idea for game advertising. The way most UA teams are using them is a hybrid of principles from game theory and a concept known as “horizontal segmentation” borrowed from the consumer goods industry.
If you can identify certain customer preference clusters within your client base, and if you can then develop products that are designed specifically to fulfill those preferences, you can align your messaging with their motivations for playing your game or trying your service.
Here’s what a fully developed player profile might look like if it was used within a creative brief. This is all fake data, but it gives you an idea of how powerful this approach to creative development can be:
Player profiles are also a demonstration of another megatrend in UA advertising that we’ve been seeing for a while: The merging of data science and UA teams.
This merger may become more pronounced as UA advertising automation continues to “streamline” teams, and as ad creative and data science teams start to share ideas and assets with each other even more than they do now.
Once you’ve got your app’s player profiles dialed in, then you can incorporate these six gaming emotional hooks into your creative so the strategy is complete.
Understanding player profiles and emotional hooks may change how you see ads forever. It’s a great lens to use as you analyze competitors’ creative.
Creative testing: Why the control is so hard to beat
So, you know creative is the best path to high ROAS. And you know competitive analysis, creative trends, and player personas and motivations should all be shaping your creative strategy. You know all this requires a lot of testing — and efficient testing — to surface the high-performance creative you need to keep your campaigns alive.
But Facebook’s testing algorithm is more sophisticated than it might first appear to be.
We’ve gotten to know it well after performing more than 10,000 A/B and multivariate tests on Facebook and its competitor, Google. We produce more than 100,000 video and image ads every year.
We test a lot of creative, and we test in a unique way. When we test, our goal is to compare how new concepts will perform versus the winning video (control) to see if the challenger can outperform the champion. Why? Because if you can’t outperform the best ad in a portfolio, you will lose money running the second or third place ads. A new ad either needs to beat the best-performing ad (the control), or you haven’t really gained anything.
Keep in mind that we focus on these verticals: gaming, ecommerce, entertainment, automotive, D2C, financial services, and lead generation. So what we have learned is specific to those niches.
Our testing process has been architected to save both time and money by:
- Killing losing creatives quickly
- Significantly reducing non-converting spend
While our process can generate both false negatives and false positives (as almost any test can), it is designed to deliver maximum efficiency in exchange for those risks. We typically allow our tests to run between 2-7 days to provide enough time to gather data without requiring the capital and time required to reach statistical significance. We also always run our tests using our AdRules software via the Facebook API.
To be clear, our process is not the Facebook best practice of running a split test and allowing the algorithm to reach statistical significance, which then moves the ad set out of the learning phase and into the optimized phase.
The insights we’ve drawn are specific to these scenarios we outline here and are not a representation of how all testing on Facebook’s platform operates. In some cases, it is valuable to have old creative retain its learning history to seamlessly A/B test without obstructing ad delivery.
If you’d like to learn more about our creative testing system, and what we’ve learned about Facebook’s testing algorithm, check out our explainer video or read our white paper on the subject.
Here’s a summary of what our tests have shown:
- During our testing, we noticed that Facebook maintains creative history for the current video winner during an A/B test. This makes the control video very difficult to beat when using our IPM testing methodology.
- Ads that are visually similar to the control appear to be grouped together in terms of how the algorithm treats them. They are also automatically given a higher IPM (impressions per thousand) value, even at the very start of a test.
- Any new creative that is visually very different from the control does not share this creative history. This “very different creative” will start the test with a lower IPM. As a consequence of this, any creative that is visually very different from the control will require more time and money to “warm up” its creative history. It will also require more time and money to get out of the learning phase and into the optimized phase before it ever has a chance of competing against the control.
Here’s what this all looks like in terms of the performance of different ads:
If the split-testing did not maintain creative history, the starting value for IPMs would be much closer. But as you can see in the chart above, this is not what our tests show is happening.
This feature causes test results like the one below, where we have set up a brand new test with an existing control, and right at the start, the control is granted an IPM around 8-9, but all the other (visually very different) ads get IPMs around 0-2.
This discovery forced us to retest a lot of creative and re-think our testing methodologies. The good news is that after re-testing a lot of our old creative, 95% of the newly retested creative that had previously failed to beat the control, actually improved its performance to a 85% failure rate (as opposed to the 95% failure rate all that creative had had before). This all resulted in a 29% lift in ROAS thanks to our new approach to testing, which takes into account the “creative history” factor in Facebook’s testing algorithm.
It took us a lot of work… but ultimately, we got a 29% lift in ROAS.
So how can you work with creative history being maintained? Watch our video for the step by step process, but basically, don’t use a control video as the benchmark for testing new creative. Don’t put your control inside the split-test of your new creative concepts. Instead, run all your new creative concepts against each other — without the control. Then let these “challenger” videos build up enough learning to effectively compete with the control “champion video.”
Conclusion: What this means for you
Once you understand how the social advertising platforms are changing, and you understand the power of testing, it’s clear: CREATIVE IS KING. Creative trends allow you to effectively steal ideas from competitors. Player profiles extend the audience and performance of your ads, while enhanced creative testing saves time and money. That’s the state of creative strategy and optimization right now. Hopefully, you read this article quickly, because it all changes so fast that a new trend or a new feature may have surfaced even while you read this.
Brian Bowman is CEO of ConsumerAcquisition.com.
Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. Content produced by our editorial team is never influenced by advertisers or sponsors in any way. For more information, contact email@example.com.