End cards: why you might be throwing away 30% extra conversions by not optimizing your end cards

End cards are the best-kept secret in mobile advertising, says Bigabid’s Avi Cohen. The reason isn’t that no one knows about them: we all encounter them pretty much every day if we play games and use apps. Rather, the reason is that most marketers don’t take end cards into account when doing creative optimization.

Big mistake.

Because the right end card combination with the right creative can massively boost performance.

Hit play, then keep scrolling:

End cards can boost ad performance by 30%

Cohen ran a test for an ad by using 2 different end cards. 

One was related to the video in the ad itself: similar looking, just with a CTA and a logo. The other was completely unrelated and generic. Common sense tells you the first ad is going to do better, but common sense would be wrong.

Instead, the generic-looking end card worked better. 

Marketers testing creative see this sort of violation of common sense all the time. Sometimes the world just doesn’t make sense, which is why marketers have learned to be data driven and not just trust their gut all the time.

In another test in a highly regulated vertical, an end card change resulted in a 20% improvement in CPI. ROI in some of those tests achieve a 50% jump as well.

One takeaway: often elements of a game or app do well in an end card. Think about using characters from your game or celebrities, influencers, spokespeople, or characters from a brand’s marketing in the end card.

“We saw success with characters almost everywhere,” Cohen says. “Because of the emotional connection … it’s always easier to relate to even a cartoon character than just their playable.”

Another tip: direct people to what you want them to do. For example, you could use a finger pointing at the install button. It all helps, Cohen says. Psychological tricks and gimmicks make a difference.

But end cards are hard to test …

The problem is that testing end cards isn’t easy. There are different standards and processes for how different networks use end cards, but that’s just the beginning. 

The real problem is the additional layer of complexity involved in testing end cards.

“It’s very hard to test [an end card] because it’s part of the video, but it’s actually a separate component,” Cohen says. “So it’s very hard to test it alone without the video and to [isolate] performance only for the end card, and not for the video.”

That’s the challenge, right?

End cards’ performance is always related to the ad that they’re attached to. If the video or playable is super awesome and incredibly successful, maybe it drags a mediocre or even bad end card’s performance up. Or if the end card is the really impressively and highly performant piece, maybe it uplevels the entire ad’s results.

Alternatively, of course, maybe both of them suck royally.

So what do you do?

You have to test combinations. Only then can you isolate performance benefits and detriments of the individual components. The challenge, of course, is that this significantly adds to testing complexity: testing with multiple variations of videos and end cards until you find the best performing version of the entire ad.

How much of your budget should you spend on testing?

“We usually use 30% of the budget [for testing],” Cohen says. “Only after 30% we are confident enough to go full on for the rest of the 70%.”

Even then, of course, there’s always a small fraction that continues to be used for testing, he says.

Making great creative: break every rule

In some sense, it’s comforting to know that there are no rules. There is no physics formula handed down by Albert Einstein that tells you exactly how to create the perfect ad. If there was, the machines could make the perfect ad every single time.

But they can’t. Even humans can’t.

Which is why injecting some element of randomness — or even insanity — is an important part of maximizing advertising ROAS. Cohen’s team dedicates 10% of their time to think outside the box. Wildly outside the box.

Think:

• Getting your kid to make an ad
• Asking a random person off the street what you should do
• Making an (almost totally) empty ad
• Telling people to NOT click
• Telling them this app probably isn’t for them
• Putting a random turtle in your car ad
• Slurping raw eggs in a UGC bit

The problem is that sometimes when we try to be smart or we try to do what we think people want, we get boring.

Predictable.

Bland.

Blah.

Break out by doing crazy stuff. One time Cohen did that, the result was an ad that looked like it had been created by a 5-year-old kid. 

The result?

“It was a great success,” Cohen says.

Much more in the full Growth Masterminds

Subscribe to Growth Masterminds wherever you find your podcasts. And connect to our YouTube channel to get recommendations right in your feed.

What you’ll get in this episode alone:

• 00:00 Introduction to Effective End Cards
• 02:03 Diving into End Card Formats
• 03:02 The Psychology Behind End Cards
• 03:36 The Evolution of Mobile Ad Lengths
• 06:09 Testing and Best Practices
• 09:03 The Importance of End Cards
• 13:40 Real-World Examples and Success Stories
• 21:54 Encouraging Creativity and Breaking Rules
• 24:34 Final Thoughts and Key Takeaways