Melissa Lertsmitivanta (Marketing Director at Realtor.com) discusses creative testing, letting creatives do the targeting and more with Dyan Khor (Growth at Cash App - Mobile Banking App), Olivia Smith (Growth Marketing Manager at HopSkipDrive - Rides for Kids) and Rhiannon Price (Director, Creative Marketing at JamCity - Mobile Games).
HopSkipDrive uses the same brand feeling and emotional value across channels but what might differ is the goal of each campaign. Example: Twitter for retargeting/re-engagement and Facebook for conversion.
Try to take a portfolio-based approach to creative testing and understand that you're adding something new to the mix. Some channels do not have a "clean" way to A/B test (e.g. TikTok) but even if there is it might not give you results of what will happen "live".
HopSkipDrive really builds creatives off of surveys they do with their customers. You can A/B test all day but you really want to first understand the value you're creating for all verticals. Example for drivers: either emotionally-driven or money-motivated and they use different creatives for each bucket.
Money-motivated user → more text-oriented creatives because these users don't really identify with the emotional imagery: it's more about presenting the options they have to make money. Emotionally driven → pictures showcasing some drivers to illustrate the look and feel of helping kids getting to school.
Initially JamCity was looking at traditional demographics (gender, age, etc.) but now they look more at the motivations. Mobile games are broadly appealing and the obvious motivations are boredom and wanting to relax. But there are also much deeper motivations why people play the game. They've identified much more specific segments that they are then targeting through creatives (not through targeting on Facebook) and even product.
You can initially start with a creative in a broad targeting to see what kind of users it brings then tweak it in that specific direction. Doing this several times you end up with a creative portfolio so that on Facebook the creatives can find their subset of users (e.g. influencer-like for younger audiences, more straightforward for older audiences).
JamCity has a fairly large consumer insights team doing quantitive and qualitative which brings a huge amount of data that they can base tests on for both creatives and product. They leverage that to improve the ads themselves but also the early retention.
JamCity has an internal team and several offices but also a huge amount of vendors (about 15) that have different specialties (UA, general marketing, production companies, high-end CG, etc.) which helps keep things diverse and creatives fresh. They try out a lot of different vendors, sometimes even outside of gaming/entertainment.
There are huge opportunities on TikTok to leverage great creators that are doing innovated things. The platform encourages finding ideas that are convincing but also "game" the algorithm. Influencer content is translatable to other platforms that have vertical video (Snapchat, Instagram) so TikTok can be a good testing ground.
For JamCity playables haven't worked out as well as everyone had hoped but HTML-interactive ads on Google UAC are really impactful. Make sure you have enough inventory for that kind of assets.
HopSkipDrive uses the same brand feeling and emotional value across channels but what might differ is the goal of each campaign. Example: Twitter for retargeting/re-engagement and Facebook for conversion.
Try to take a portfolio-based approach to creative testing and understand that you're adding something new to the mix. Some channels do not have a "clean" way to A/B test (e.g. TikTok) but even if there is it might not give you results of what will happen "live".
HopSkipDrive really builds creatives off of surveys they do with their customers. You can A/B test all day but you really want to first understand the value you're creating for all verticals. Example for drivers: either emotionally-driven or money-motivated and they use different creatives for each bucket.
Money-motivated user → more text-oriented creatives because these users don't really identify with the emotional imagery: it's more about presenting the options they have to make money. Emotionally driven → pictures showcasing some drivers to illustrate the look and feel of helping kids getting to school.
Initially JamCity was looking at traditional demographics (gender, age, etc.) but now they look more at the motivations. Mobile games are broadly appealing and the obvious motivations are boredom and wanting to relax. But there are also much deeper motivations why people play the game. They've identified much more specific segments that they are then targeting through creatives (not through targeting on Facebook) and even product.
You can initially start with a creative in a broad targeting to see what kind of users it brings then tweak it in that specific direction. Doing this several times you end up with a creative portfolio so that on Facebook the creatives can find their subset of users (e.g. influencer-like for younger audiences, more straightforward for older audiences).
JamCity has a fairly large consumer insights team doing quantitive and qualitative which brings a huge amount of data that they can base tests on for both creatives and product. They leverage that to improve the ads themselves but also the early retention.
JamCity has an internal team and several offices but also a huge amount of vendors (about 15) that have different specialties (UA, general marketing, production companies, high-end CG, etc.) which helps keep things diverse and creatives fresh. They try out a lot of different vendors, sometimes even outside of gaming/entertainment.
There are huge opportunities on TikTok to leverage great creators that are doing innovated things. The platform encourages finding ideas that are convincing but also "game" the algorithm. Influencer content is translatable to other platforms that have vertical video (Snapchat, Instagram) so TikTok can be a good testing ground.
For JamCity playables haven't worked out as well as everyone had hoped but HTML-interactive ads on Google UAC are really impactful. Make sure you have enough inventory for that kind of assets.
HopSkipDrive uses the same brand feeling and emotional value across channels but what might differ is the goal of each campaign. Example: Twitter for retargeting/re-engagement and Facebook for conversion.
Try to take a portfolio-based approach to creative testing and understand that you're adding something new to the mix. Some channels do not have a "clean" way to A/B test (e.g. TikTok) but even if there is it might not give you results of what will happen "live".
HopSkipDrive really builds creatives off of surveys they do with their customers. You can A/B test all day but you really want to first understand the value you're creating for all verticals. Example for drivers: either emotionally-driven or money-motivated and they use different creatives for each bucket.
Money-motivated user → more text-oriented creatives because these users don't really identify with the emotional imagery: it's more about presenting the options they have to make money. Emotionally driven → pictures showcasing some drivers to illustrate the look and feel of helping kids getting to school.
Initially JamCity was looking at traditional demographics (gender, age, etc.) but now they look more at the motivations. Mobile games are broadly appealing and the obvious motivations are boredom and wanting to relax. But there are also much deeper motivations why people play the game. They've identified much more specific segments that they are then targeting through creatives (not through targeting on Facebook) and even product.
You can initially start with a creative in a broad targeting to see what kind of users it brings then tweak it in that specific direction. Doing this several times you end up with a creative portfolio so that on Facebook the creatives can find their subset of users (e.g. influencer-like for younger audiences, more straightforward for older audiences).
JamCity has a fairly large consumer insights team doing quantitive and qualitative which brings a huge amount of data that they can base tests on for both creatives and product. They leverage that to improve the ads themselves but also the early retention.
JamCity has an internal team and several offices but also a huge amount of vendors (about 15) that have different specialties (UA, general marketing, production companies, high-end CG, etc.) which helps keep things diverse and creatives fresh. They try out a lot of different vendors, sometimes even outside of gaming/entertainment.
There are huge opportunities on TikTok to leverage great creators that are doing innovated things. The platform encourages finding ideas that are convincing but also "game" the algorithm. Influencer content is translatable to other platforms that have vertical video (Snapchat, Instagram) so TikTok can be a good testing ground.
For JamCity playables haven't worked out as well as everyone had hoped but HTML-interactive ads on Google UAC are really impactful. Make sure you have enough inventory for that kind of assets.
Notes for this resource are currently being transferred and will be available soon.
Dyan (Cash App)
New to UA and still testing a lot of things. Really built a financial brand that users love. Homepage is bright with colors and animation and they continue that in UA for creatives (3D animation, textures, etc.).
Channels: snapchat, tiktok, reddit, and expanding into larger channels as well.
Olivia (HopSkipDrive)
UA for many verticals. Biggest channel is Facebook: all placements. Running about 10-20 different campaigns at the time, top of funnel and bottom of funnel.
Have B2B and B2C and also have restrictions due to kid focus so need to make sure they're creative with their creatives (so the creatives do the targeting).
[💎 @03:14:32] HopSkipDrive uses the same brand feeling and emotional value across channels but what might differ is the goal of each campaign. Example: Twitter for retargeting/re-engagement and Facebook for conversion.
Rhiannon (JamCity)
Dominantly Facebook. Demo is more older which make Snapchat/Tiktok less relevant. Also use rewarded videos.
Different strategies around different networks but they really test and explore variety on Facebook.
Rhiannon (JamCity)Rhiannon (JamCity)
Always testing and do not assume A/B testing is the way to go. Test different geos, markets, audiences, MAI/AEO/VO, etc.
Dyan (Cash App)
[💎 @03:17:22] Try to take a portfolio-based approach to creative testing and understand that you're adding something new to the mix. Some channels do not have a "clean" way to A/B test (e.g. TikTok) but even if there is it will not give you results of what will happen "live".
Olivia (HopSkipDrive)
[💎 @03:18:00] HopSkipDrive really builds creatives off surveys they do with their customers. You can A/B test all day but you really want to first understand the value you're creating for all verticals. Example for drivers: either emotionally-driven or money-motivated and they use different creatives for each bucket.
Olivia (HopSkipDrive)
[💎 @03:19:32] For money-motivated users they tend to serve creatives that are more text-oriented because these users don't really identify with the emotional imagery: it's more about presenting the options they have to make money. For the emotionally driven they use pictures showcasing some drivers to illustrate the look and feel of helping kids getting to school.
Rhiannon (JamCity)
[💎 @03:21:15] Initially JamCity was looking at traditional demographics (gender, age, etc.) but now they look more at the motivations. Mobile games are broadly appealing and the obvious motivations are boredom and wanting to relax. But there are also much deeper motivations why people play the game. They've identified much more specific segments that they are then targeting through creatives (not through targeting on Facebook) and even product.
Dyan (Cash App)
[💎 @03:22:25] (Hopper and) Cash App have a pretty broad user base. You can initially start with a creative in a broad targeting to see what kind of users it brings then tweak it in that specific direction. Doing this several times you end up with a creative portfolio so that on Facebook the creatives can find their subset of users (e.g. influencer-like for younger audiences, more straightforward for older audiences).
Olivia (HopSkipDrive)
The motivational driver can be a bigger part to find subsets of users rather than trying to "find them" through targeting.
Rhiannon (JamCity)
[💎 @03:23:57] JamCity has a fairly large consumer insights team doing quantitive and qualitative which brings a huge amount of data that they can base tests on for both creatives and product. They leverage that to improve the ads themselves but also the early retention.
Olivia (HopSkipDrive)
BI team has developed models that show ideal motivational drivers: the profile they fit, the channels they are coming from. It helps them shape their creative.
Dyan (Cash App)
Not the case yet. They're still defining the user demo but already finding different behaviors.
Olivia (HopSkipDrive)
Rely on internal. Work with a designer that also designed the brand guidelines. They have also used agencies before as well.
Rhiannon (JamCity)
[💎 @03:26:40] JamCity has an internal team and several offices but also a huge amount of vendors (about 15) that have different specialties (UA, general marketing, production companies, high-end CG, etc.) which helps keep things diverse and creatives fresh. They try out a lot of different vendors, sometimes even outside of gaming/entertainment.
Dyan (Cash App)
Still discovering the production process but big focus on strong brand look and feel so working with internal team on improving performance. There is a need for speed that could make them consider outsourcing but also quality.
Dyan (Cash App)
[💎 @03:28:59] There are huge opportunities on TikTok to leverage great creators that are doing innovated things. The platform encourages finding ideas that are convincing but also "game" the algorithm. Influencer content is translatable to other platforms that have vertical video (Snapchat, Instagram) so TikTok can be a good testing ground.
Olivia (HopSkipDrive)
One of the top performing campaigns was a hybrid between UGC and testimonial content. The campaign was about "Meet X" featuring a driver, with soft CTAs. Better than a br
Rhiannon (JamCity)
A year ago she would have said playable but it didn't work out as well as they had hoped.
[💎 @03:30:47] For JamCity playables haven't worked out as well as everyone had hoped but HTML-interactive ads on Google UAC are really impactful. Make sure you have enough inventory for that kind of assets.
Dyan (Cash App)
New to UA and still testing a lot of things. Really built a financial brand that users love. Homepage is bright with colors and animation and they continue that in UA for creatives (3D animation, textures, etc.).
Channels: snapchat, tiktok, reddit, and expanding into larger channels as well.
Olivia (HopSkipDrive)
UA for many verticals. Biggest channel is Facebook: all placements. Running about 10-20 different campaigns at the time, top of funnel and bottom of funnel.
Have B2B and B2C and also have restrictions due to kid focus so need to make sure they're creative with their creatives (so the creatives do the targeting).
[💎 @03:14:32] HopSkipDrive uses the same brand feeling and emotional value across channels but what might differ is the goal of each campaign. Example: Twitter for retargeting/re-engagement and Facebook for conversion.
Rhiannon (JamCity)
Dominantly Facebook. Demo is more older which make Snapchat/Tiktok less relevant. Also use rewarded videos.
Different strategies around different networks but they really test and explore variety on Facebook.
Rhiannon (JamCity)Rhiannon (JamCity)
Always testing and do not assume A/B testing is the way to go. Test different geos, markets, audiences, MAI/AEO/VO, etc.
Dyan (Cash App)
[💎 @03:17:22] Try to take a portfolio-based approach to creative testing and understand that you're adding something new to the mix. Some channels do not have a "clean" way to A/B test (e.g. TikTok) but even if there is it will not give you results of what will happen "live".
Olivia (HopSkipDrive)
[💎 @03:18:00] HopSkipDrive really builds creatives off surveys they do with their customers. You can A/B test all day but you really want to first understand the value you're creating for all verticals. Example for drivers: either emotionally-driven or money-motivated and they use different creatives for each bucket.
Olivia (HopSkipDrive)
[💎 @03:19:32] For money-motivated users they tend to serve creatives that are more text-oriented because these users don't really identify with the emotional imagery: it's more about presenting the options they have to make money. For the emotionally driven they use pictures showcasing some drivers to illustrate the look and feel of helping kids getting to school.
Rhiannon (JamCity)
[💎 @03:21:15] Initially JamCity was looking at traditional demographics (gender, age, etc.) but now they look more at the motivations. Mobile games are broadly appealing and the obvious motivations are boredom and wanting to relax. But there are also much deeper motivations why people play the game. They've identified much more specific segments that they are then targeting through creatives (not through targeting on Facebook) and even product.
Dyan (Cash App)
[💎 @03:22:25] (Hopper and) Cash App have a pretty broad user base. You can initially start with a creative in a broad targeting to see what kind of users it brings then tweak it in that specific direction. Doing this several times you end up with a creative portfolio so that on Facebook the creatives can find their subset of users (e.g. influencer-like for younger audiences, more straightforward for older audiences).
Olivia (HopSkipDrive)
The motivational driver can be a bigger part to find subsets of users rather than trying to "find them" through targeting.
Rhiannon (JamCity)
[💎 @03:23:57] JamCity has a fairly large consumer insights team doing quantitive and qualitative which brings a huge amount of data that they can base tests on for both creatives and product. They leverage that to improve the ads themselves but also the early retention.
Olivia (HopSkipDrive)
BI team has developed models that show ideal motivational drivers: the profile they fit, the channels they are coming from. It helps them shape their creative.
Dyan (Cash App)
Not the case yet. They're still defining the user demo but already finding different behaviors.
Olivia (HopSkipDrive)
Rely on internal. Work with a designer that also designed the brand guidelines. They have also used agencies before as well.
Rhiannon (JamCity)
[💎 @03:26:40] JamCity has an internal team and several offices but also a huge amount of vendors (about 15) that have different specialties (UA, general marketing, production companies, high-end CG, etc.) which helps keep things diverse and creatives fresh. They try out a lot of different vendors, sometimes even outside of gaming/entertainment.
Dyan (Cash App)
Still discovering the production process but big focus on strong brand look and feel so working with internal team on improving performance. There is a need for speed that could make them consider outsourcing but also quality.
Dyan (Cash App)
[💎 @03:28:59] There are huge opportunities on TikTok to leverage great creators that are doing innovated things. The platform encourages finding ideas that are convincing but also "game" the algorithm. Influencer content is translatable to other platforms that have vertical video (Snapchat, Instagram) so TikTok can be a good testing ground.
Olivia (HopSkipDrive)
One of the top performing campaigns was a hybrid between UGC and testimonial content. The campaign was about "Meet X" featuring a driver, with soft CTAs. Better than a br
Rhiannon (JamCity)
A year ago she would have said playable but it didn't work out as well as they had hoped.
[💎 @03:30:47] For JamCity playables haven't worked out as well as everyone had hoped but HTML-interactive ads on Google UAC are really impactful. Make sure you have enough inventory for that kind of assets.
Dyan (Cash App)
New to UA and still testing a lot of things. Really built a financial brand that users love. Homepage is bright with colors and animation and they continue that in UA for creatives (3D animation, textures, etc.).
Channels: snapchat, tiktok, reddit, and expanding into larger channels as well.
Olivia (HopSkipDrive)
UA for many verticals. Biggest channel is Facebook: all placements. Running about 10-20 different campaigns at the time, top of funnel and bottom of funnel.
Have B2B and B2C and also have restrictions due to kid focus so need to make sure they're creative with their creatives (so the creatives do the targeting).
[💎 @03:14:32] HopSkipDrive uses the same brand feeling and emotional value across channels but what might differ is the goal of each campaign. Example: Twitter for retargeting/re-engagement and Facebook for conversion.
Rhiannon (JamCity)
Dominantly Facebook. Demo is more older which make Snapchat/Tiktok less relevant. Also use rewarded videos.
Different strategies around different networks but they really test and explore variety on Facebook.
Rhiannon (JamCity)Rhiannon (JamCity)
Always testing and do not assume A/B testing is the way to go. Test different geos, markets, audiences, MAI/AEO/VO, etc.
Dyan (Cash App)
[💎 @03:17:22] Try to take a portfolio-based approach to creative testing and understand that you're adding something new to the mix. Some channels do not have a "clean" way to A/B test (e.g. TikTok) but even if there is it will not give you results of what will happen "live".
Olivia (HopSkipDrive)
[💎 @03:18:00] HopSkipDrive really builds creatives off surveys they do with their customers. You can A/B test all day but you really want to first understand the value you're creating for all verticals. Example for drivers: either emotionally-driven or money-motivated and they use different creatives for each bucket.
Olivia (HopSkipDrive)
[💎 @03:19:32] For money-motivated users they tend to serve creatives that are more text-oriented because these users don't really identify with the emotional imagery: it's more about presenting the options they have to make money. For the emotionally driven they use pictures showcasing some drivers to illustrate the look and feel of helping kids getting to school.
Rhiannon (JamCity)
[💎 @03:21:15] Initially JamCity was looking at traditional demographics (gender, age, etc.) but now they look more at the motivations. Mobile games are broadly appealing and the obvious motivations are boredom and wanting to relax. But there are also much deeper motivations why people play the game. They've identified much more specific segments that they are then targeting through creatives (not through targeting on Facebook) and even product.
Dyan (Cash App)
[💎 @03:22:25] (Hopper and) Cash App have a pretty broad user base. You can initially start with a creative in a broad targeting to see what kind of users it brings then tweak it in that specific direction. Doing this several times you end up with a creative portfolio so that on Facebook the creatives can find their subset of users (e.g. influencer-like for younger audiences, more straightforward for older audiences).
Olivia (HopSkipDrive)
The motivational driver can be a bigger part to find subsets of users rather than trying to "find them" through targeting.
Rhiannon (JamCity)
[💎 @03:23:57] JamCity has a fairly large consumer insights team doing quantitive and qualitative which brings a huge amount of data that they can base tests on for both creatives and product. They leverage that to improve the ads themselves but also the early retention.
Olivia (HopSkipDrive)
BI team has developed models that show ideal motivational drivers: the profile they fit, the channels they are coming from. It helps them shape their creative.
Dyan (Cash App)
Not the case yet. They're still defining the user demo but already finding different behaviors.
Olivia (HopSkipDrive)
Rely on internal. Work with a designer that also designed the brand guidelines. They have also used agencies before as well.
Rhiannon (JamCity)
[💎 @03:26:40] JamCity has an internal team and several offices but also a huge amount of vendors (about 15) that have different specialties (UA, general marketing, production companies, high-end CG, etc.) which helps keep things diverse and creatives fresh. They try out a lot of different vendors, sometimes even outside of gaming/entertainment.
Dyan (Cash App)
Still discovering the production process but big focus on strong brand look and feel so working with internal team on improving performance. There is a need for speed that could make them consider outsourcing but also quality.
Dyan (Cash App)
[💎 @03:28:59] There are huge opportunities on TikTok to leverage great creators that are doing innovated things. The platform encourages finding ideas that are convincing but also "game" the algorithm. Influencer content is translatable to other platforms that have vertical video (Snapchat, Instagram) so TikTok can be a good testing ground.
Olivia (HopSkipDrive)
One of the top performing campaigns was a hybrid between UGC and testimonial content. The campaign was about "Meet X" featuring a driver, with soft CTAs. Better than a br
Rhiannon (JamCity)
A year ago she would have said playable but it didn't work out as well as they had hoped.
[💎 @03:30:47] For JamCity playables haven't worked out as well as everyone had hoped but HTML-interactive ads on Google UAC are really impactful. Make sure you have enough inventory for that kind of assets.
