We sat down with ironSource Art Director Achva Kahana to learn how she and her team create experiences users enjoy, and design mobile video ads people click on.

Let’s start by tackling app trailers - can you give us an overview on the different types of app trailers there are? When is it better to use one type over the other?

Sure. There are two main types of app trailers: gameplay/features and storytelling. 

In a gameplay/features trailer, we highlight key features, letting the viewer know exactly what to expect if they download the app. It’s very straightforward.

In a storytelling trailer, it’s more about telling a story without giving the game away. The purpose is to intrigue a potential user, create hype, and drive brand awareness. Think of the Game of War ad with Kate Upton that was featured in the 2015 Super Bowl. It’s a stunning trailer, but it doesn’t have much to do with the actual Game of War game.

In a best case scenario, you’d combine gameplay/feature and storytelling. The Clash of Clans: Revenge ad with Liam Neeson is a perfect example. It starts with some gameplay in the beginning, and then seamlessly moves into a shot on Liam Neeson in a cafe playing Clash of Clans whispering, “I can’t wait to destroy your village. I will have my revenge.” That’s great storytelling.

As for when one or the other type of app trailer works best - if your app isn’t too well known yet, you should probably stick with gameplay/features option. The user needs to understand what the app is if they want to download it. Otherwise, they’ll just download something they weren’t expecting and either never use it or immediately delete it.

Big name apps, on the other hand, can rely on a storytelling trailer to build brand awareness, bring in organic app searches and downloads.

Can you walk us through your creative process? How do you make the magic happen?

First, the client tells us who their end user is. Basically, who are they trying to target? They’ll give us demographic information like age, gender, location, and interests. They might tell us that their targeting teenage girls in India, or they might say they’re after geeks who love sports.

Then we get together as a studio and spit out any idea that comes to mind. It might be a cliche, but there are really no bad ideas at this point in the flow. We think about what makes the app unique. Does it tie into a bigger cultural trend? Do the features solve some pain point?

We try to distill the essence of the app’s value proposition, and then tie that into situations from daily life which users can relate to.  

You guys do this all the time, you’re probably experts, but how would a beginner approach brainstorming? For example, a developer who’s creating his own video app ad.

In many ways if you’re a developer trying to build a video ad for your app you’re in an even stronger position, because you know your app best. You just have to find the part of it that will come across best in 15 seconds of video. There’s probably some part of your app that you’re really proud of. I recommend watching how other people interact and play with your app. It’ll help you find out which part of your app is the most engaging. Watching them play could give you an idea for a snippet of gameplay to include. It’ll get you inspired.

Great. What else should advertisers keep in mind during the brainstorming phase? What’s the one critical point they should never forget?

With each idea, make sure you also have a “value proposition.” What are your viewers supposed to get out of the ad? There should be one crystal clear message you are trying to get across. When you’re playing with 15-30 seconds, one message is all you’ve got. Everything in the video ad should be designed to communicate and support that message.

Okay, what’s the process look like once you’re done brainstorming?

We widen the circle, start asking around, getting feedback on each idea, and seeing which idea gets the best responses. We go back and forth on this until we have two or three really strong ideas.

Then we move onto the storyboard. It’s hard to stay on track without one.

Is there a set structure you generally stick to?

It sounds obvious, but it’s  always good to have a beginning, middle, and end. Just like any good story. Sometimes with short app video ads it’s tempting to abandon that basic structure, thinking there isn’t the time or space to make it work, but there’s a reason stories evolved that way - it just works.

What we tend to do is start with the app icon, move onto some gameplay or top features (with descriptive, fun copy), and end with a card that shows the app icon, a strong CTA, and the App Store or Play Store icon. This is the simplest storyboard structure.

It’s worth remembering that the top features or gameplay you include in the “middle” section should highlight the concept and value proposition you defined in the brainstorming phase.

Do you ever wander out of this general structure?

For big name apps like Facebook or Angry Birds, we like to get a bit more creative, and really dig deep into the “middle” section. Instead of just showing off features, we’ll focus more time on explaining the meaning and importance behind them -- we’ll go deeper into the why.

Can you share an example of what you mean?

In an ad we recently designed for Facebook, we focused on creating family moments. That’s the why - meaning, unique family moments are why sharing statuses and photos with friends is exciting. There’s one frame in the ad where we have a baby smiling through the Facebook camera lens. The previous frame was a screenshot of a Facebook status saying, “meet our new baby.” It’s important to remember simplicity is key - that’s two quick frames which really drive home the concept of family memories shared through a social platform.

How do you keep audience and targeting in mind when you’re creating the trailers?

Besides getting information about the general audience (age, demographic, interests) like I mentioned before, it’s super important to know geos. Localization is critical. It completely influences the way we approach the video.

If the video is being served in the US, we’ll be a bit more hands off. If the video is going to emerging markets, where users are less familiar with well-known apps, we’ll explain the top features a bit more explicitly.

We also need to make sure we’re using local names and taking local behaviors into consideration. For instance, we once made an ad for a ride sharing app in India. Because it’s more common for Indians to pay in cash, we made sure to create an animation that highlights the cash feature.

Obviously, the end goal is for the user to download at the end.  How do you make sure that the call to action isn’t too abrupt or annoying?

A/B testing is a must. There isn’t a one-size-fits-all solution, so we never give the client just one ad. We give them three or four different versions, each with a different CTA. The end card could have just a download button and nothing else. Or an image with text. Or a Google Play badge. Or a more colorful frame. It’s up to the advertiser to run the ads and see which one does best. Once they have a winner, they’ll only run the top-performing ad.

Where do you see app trailers headed in the future?

(AK) In the future, I think we’ll see app videos become way more engaging. It’s pretty common for users to download apps they don’t like because they see an ad that didn’t tell the whole story. But if ads could give a sneak peak and really pull the user in, we’d see less of that. I think the next step is interactivity. It’s kind of like what we’re seeing with playable ads right now.

Let's put these tips to good use

Unify your app's marketing with Luna