Session 2 of the “Secrets of Game Growth” event by Mopub. Eric Seufert (Analyst and Strategy Consultant at Heracles Media, Founder at MobileDevMemo) moderates a panel about UA and creatives with Mary Kim (Head of Growth at Game Hive), Linus Otten (Senior Online Marketing Manager at Bytro and co-host of the Marketing Rockers podcast) and Michael O’Brien (VP of Growth at Headlight).
TV/OTT might have been disregarded in the past, but now ATT is opening the door to this kind of channel.
With one of their newer titles, Game Hive has started looking at performance on an eROAS basis: attributing all organic uplift to paid efforts. eROAS = (Net IAP Revenue from both Inorganic & Organic cohorts + Ad Revenue from both Inorganic & Organic cohorts) / (UA Spend on the Inorganic cohorts).
Ad creatives that incorporate the influencer (especially big ones) perform very well, which means influencer marketing helps with the channel itself but also supports UA.
Game Hive’s creative director analyzed all the patterns of all the creative concepts they’ve tested so far (including top and low performers), added that on to a matrix and tagged creatives to find which next creative concept would be the next high-impact creative.
The big catch is a playbook created by Facebook advocating for motivation-based creatives, rather than pushing out creatives to the same audience of players. It helps figure out who really are you players and what can motivate them to try your game.
Headlight looks for high leverage, big incremental wins (often radical shifts in concepts or positioning) combined with low production costs. They often start with static images and invest more effort if they see good results.
In midcore games, highly profitable creatives often have lower IPM or CTR: they perform from a ROAS perspective because they resonate with the core audience. Similarly, the higher CTR creatives typically end up leading to lower in-game engagement. Don’t make the mistake of looking just at IPM.
Finding creative winners is more about finding ideas introducing big differences and iterating in big ways rather than small incremental changes.
Bytro has had a specific messaging UA (war inside the US) that has been performing very well but initially that messaging was not really well captured in the game itself. Once they started using the UA messaging in the tutorial, they saw an improvement in all the KPIs.
Fake ads and clickbait ads do perform well sometimes (e.g. spelling mistake in copy). However you should not go only in that direction and at least have a balance. Fake ads will not go away anytime soon.
TV/OTT might have been disregarded in the past, but now ATT is opening the door to this kind of channel.
With one of their newer titles, Game Hive has started looking at performance on an eROAS basis: attributing all organic uplift to paid efforts. eROAS = (Net IAP Revenue from both Inorganic & Organic cohorts + Ad Revenue from both Inorganic & Organic cohorts) / (UA Spend on the Inorganic cohorts).
Ad creatives that incorporate the influencer (especially big ones) perform very well, which means influencer marketing helps with the channel itself but also supports UA.
Game Hive’s creative director analyzed all the patterns of all the creative concepts they’ve tested so far (including top and low performers), added that on to a matrix and tagged creatives to find which next creative concept would be the next high-impact creative.
The big catch is a playbook created by Facebook advocating for motivation-based creatives, rather than pushing out creatives to the same audience of players. It helps figure out who really are you players and what can motivate them to try your game.
Headlight looks for high leverage, big incremental wins (often radical shifts in concepts or positioning) combined with low production costs. They often start with static images and invest more effort if they see good results.
In midcore games, highly profitable creatives often have lower IPM or CTR: they perform from a ROAS perspective because they resonate with the core audience. Similarly, the higher CTR creatives typically end up leading to lower in-game engagement. Don’t make the mistake of looking just at IPM.
Finding creative winners is more about finding ideas introducing big differences and iterating in big ways rather than small incremental changes.
Bytro has had a specific messaging UA (war inside the US) that has been performing very well but initially that messaging was not really well captured in the game itself. Once they started using the UA messaging in the tutorial, they saw an improvement in all the KPIs.
Fake ads and clickbait ads do perform well sometimes (e.g. spelling mistake in copy). However you should not go only in that direction and at least have a balance. Fake ads will not go away anytime soon.
TV/OTT might have been disregarded in the past, but now ATT is opening the door to this kind of channel.
With one of their newer titles, Game Hive has started looking at performance on an eROAS basis: attributing all organic uplift to paid efforts. eROAS = (Net IAP Revenue from both Inorganic & Organic cohorts + Ad Revenue from both Inorganic & Organic cohorts) / (UA Spend on the Inorganic cohorts).
Ad creatives that incorporate the influencer (especially big ones) perform very well, which means influencer marketing helps with the channel itself but also supports UA.
Game Hive’s creative director analyzed all the patterns of all the creative concepts they’ve tested so far (including top and low performers), added that on to a matrix and tagged creatives to find which next creative concept would be the next high-impact creative.
The big catch is a playbook created by Facebook advocating for motivation-based creatives, rather than pushing out creatives to the same audience of players. It helps figure out who really are you players and what can motivate them to try your game.
Headlight looks for high leverage, big incremental wins (often radical shifts in concepts or positioning) combined with low production costs. They often start with static images and invest more effort if they see good results.
In midcore games, highly profitable creatives often have lower IPM or CTR: they perform from a ROAS perspective because they resonate with the core audience. Similarly, the higher CTR creatives typically end up leading to lower in-game engagement. Don’t make the mistake of looking just at IPM.
Finding creative winners is more about finding ideas introducing big differences and iterating in big ways rather than small incremental changes.
Bytro has had a specific messaging UA (war inside the US) that has been performing very well but initially that messaging was not really well captured in the game itself. Once they started using the UA messaging in the tutorial, they saw an improvement in all the KPIs.
Fake ads and clickbait ads do perform well sometimes (e.g. spelling mistake in copy). However you should not go only in that direction and at least have a balance. Fake ads will not go away anytime soon.
Notes for this resource are currently being transferred and will be available soon.
Linus
It will be harder to target users than before.
At Bytro (mostly war strategy games) they’ve invested more in influencer marketing since last year. Still direct response but a bit more: they look for influencers that are very well suited for their topics (e.g. WW3 game for which they look for influencers that talk about WW3 hypotheticals).
They also expect, and are starting to see, a positive branding effect from influencer videos as well as new users coming from older videos.
Michael
Challenges on direct response because of ATT are not a nail in the coffin: there are still many ways to reach users. But it did push everybody to get up to speed and build data models to measure incrementality.
[💎@07:40] TV/OTT might have been disregarded in the past, but now ATT is opening the door to this kind of channel.
Mary
[💎@09:00] With one of their newer titles, Game Hive has started looking at performance on an eROAS basis: attributing all organic uplift to paid efforts. eROAS = (Net IAP Revenue from both Inorganic & Organic cohorts + Ad Revenue from both Inorganic & Organic cohorts) / (UA Spend on the Inorganic cohorts).
They’ve found great success with influencer marketing as well.
[💎@09:50] Ad creatives that incorporate the influencer (especially big ones) perform very well, which means influencer marketing helps with the channel itself but also supports UA.
Eric
Media buying (e.g. TV for MZ) is not just about driving installs, it’s also about increasing conversion (CTR, CVR).
Mary
Still pumping out a lot of creative concepts.
[💎@12:02] Game Hive’s creative director analyzed all the patterns of all the creative concepts they’ve tested so far (including top and low performers), added that on to a matrix and tagged creatives to find which next creative concept would be the next high-impact creative.
[💎@12:36] The big catch is a playbook created by Facebook advocating for motivation-based creatives, rather than pushing out creatives to the same audience of players. It helps figure out who really are you players and what can motivate them to try your game.
Michael
iOS 14 means there are constraints as to how much you can test at once.
There are tradeoffs (big and small) in terms of what you’re exploring creatively (e.g. is it a new concept, just a new variant, just changing colors, etc.) as well as trade offs on the production costs (effort required).
[💎@14:18] Headlight looks for high leverage, big incremental wins (often radical shifts in concepts or positioning) combined with low production costs. They often start with static images and invest more effort if they see good results.
Eric
A lot of studios make the mistake of only optimizing for IPM and not wanting to commit to getting a ROAS measure.
Michael
[💎@15:49] In midcore games, highly profitable creatives often have lower IPM or CTR: they perform from a ROAS perspective because they resonate with the core audience. Similarly, the higher CTR creatives typically end up leading to lower in-game engagement. Don’t make the mistake of looking just at IPM.
Linus
[💎@16:52] Finding creative winners is more about finding ideas introducing big differences and iterating in big ways rather than small incremental changes.
Dedicate each week time to find ideas and plan what they can test next.
Michael
Marketing and product typically look at cohorts and metrics differently. There is usually a long process to align on what performance looks like and identifying where feature changes are supposed to impact performance. This is especially true when it relates to changes impacting late monetization and how that ties in with new user acquisition.
Mary
Still a small team so it is easier to collaborate. It’s a balance between the product team wanting to release more content to increase LTV and the UA team wanting more things to grow the user base.
“You can’t scale up if the early-game metrics look bad”
Alignment is also important for potential new events, what you could push in UA, etc.
Linus
Understanding of the data set on both teams is very important so both teams understand where the game is at.
They are now thinking about making their game more influencer-friendly: slow games that take weeks to complete don’t make good candidates for influencer streaming, so they’re thinking about adapting the game to the channel.
[💎@21:31] Bytro has had a specific messaging UA (war inside the US) that has been performing very well but initially that messaging was not really well captured in the game itself. Once they started using the UA messaging in the tutorial, they saw an improvement in all the KPIs.
Michael
Marketing teams and agencies should not accept superficial targets/metrics. Similarly, product teams often do not look at the changes in the acquisition sources.
Michael
A lot of people want to have the stance that “real” works better but we will most likely see fake ads continue and sometimes succeed (e.g. tower strategy games).
Mary
Even if you don’t like them, you have to admit that sometimes things like spelling mistakes work well.
[💎@24:41] Fake ads and clickbait ads do perform well sometimes (e.g. spelling mistake in copy). However you should not go only in that direction and at least have a balance.
"Fake ads will not go away anytime soon."
Linus
Have seen weaker performance when moving away from accurately representing the game.
With iOS 14.5 you have to be even more precise with your creatives because that’s also how you target users (e.g. On Facebook AAA).
Linus
It will be harder to target users than before.
At Bytro (mostly war strategy games) they’ve invested more in influencer marketing since last year. Still direct response but a bit more: they look for influencers that are very well suited for their topics (e.g. WW3 game for which they look for influencers that talk about WW3 hypotheticals).
They also expect, and are starting to see, a positive branding effect from influencer videos as well as new users coming from older videos.
Michael
Challenges on direct response because of ATT are not a nail in the coffin: there are still many ways to reach users. But it did push everybody to get up to speed and build data models to measure incrementality.
[💎@07:40] TV/OTT might have been disregarded in the past, but now ATT is opening the door to this kind of channel.
Mary
[💎@09:00] With one of their newer titles, Game Hive has started looking at performance on an eROAS basis: attributing all organic uplift to paid efforts. eROAS = (Net IAP Revenue from both Inorganic & Organic cohorts + Ad Revenue from both Inorganic & Organic cohorts) / (UA Spend on the Inorganic cohorts).
They’ve found great success with influencer marketing as well.
[💎@09:50] Ad creatives that incorporate the influencer (especially big ones) perform very well, which means influencer marketing helps with the channel itself but also supports UA.
Eric
Media buying (e.g. TV for MZ) is not just about driving installs, it’s also about increasing conversion (CTR, CVR).
Mary
Still pumping out a lot of creative concepts.
[💎@12:02] Game Hive’s creative director analyzed all the patterns of all the creative concepts they’ve tested so far (including top and low performers), added that on to a matrix and tagged creatives to find which next creative concept would be the next high-impact creative.
[💎@12:36] The big catch is a playbook created by Facebook advocating for motivation-based creatives, rather than pushing out creatives to the same audience of players. It helps figure out who really are you players and what can motivate them to try your game.
Michael
iOS 14 means there are constraints as to how much you can test at once.
There are tradeoffs (big and small) in terms of what you’re exploring creatively (e.g. is it a new concept, just a new variant, just changing colors, etc.) as well as trade offs on the production costs (effort required).
[💎@14:18] Headlight looks for high leverage, big incremental wins (often radical shifts in concepts or positioning) combined with low production costs. They often start with static images and invest more effort if they see good results.
Eric
A lot of studios make the mistake of only optimizing for IPM and not wanting to commit to getting a ROAS measure.
Michael
[💎@15:49] In midcore games, highly profitable creatives often have lower IPM or CTR: they perform from a ROAS perspective because they resonate with the core audience. Similarly, the higher CTR creatives typically end up leading to lower in-game engagement. Don’t make the mistake of looking just at IPM.
Linus
[💎@16:52] Finding creative winners is more about finding ideas introducing big differences and iterating in big ways rather than small incremental changes.
Dedicate each week time to find ideas and plan what they can test next.
Michael
Marketing and product typically look at cohorts and metrics differently. There is usually a long process to align on what performance looks like and identifying where feature changes are supposed to impact performance. This is especially true when it relates to changes impacting late monetization and how that ties in with new user acquisition.
Mary
Still a small team so it is easier to collaborate. It’s a balance between the product team wanting to release more content to increase LTV and the UA team wanting more things to grow the user base.
“You can’t scale up if the early-game metrics look bad”
Alignment is also important for potential new events, what you could push in UA, etc.
Linus
Understanding of the data set on both teams is very important so both teams understand where the game is at.
They are now thinking about making their game more influencer-friendly: slow games that take weeks to complete don’t make good candidates for influencer streaming, so they’re thinking about adapting the game to the channel.
[💎@21:31] Bytro has had a specific messaging UA (war inside the US) that has been performing very well but initially that messaging was not really well captured in the game itself. Once they started using the UA messaging in the tutorial, they saw an improvement in all the KPIs.
Michael
Marketing teams and agencies should not accept superficial targets/metrics. Similarly, product teams often do not look at the changes in the acquisition sources.
Michael
A lot of people want to have the stance that “real” works better but we will most likely see fake ads continue and sometimes succeed (e.g. tower strategy games).
Mary
Even if you don’t like them, you have to admit that sometimes things like spelling mistakes work well.
[💎@24:41] Fake ads and clickbait ads do perform well sometimes (e.g. spelling mistake in copy). However you should not go only in that direction and at least have a balance.
"Fake ads will not go away anytime soon."
Linus
Have seen weaker performance when moving away from accurately representing the game.
With iOS 14.5 you have to be even more precise with your creatives because that’s also how you target users (e.g. On Facebook AAA).
Linus
It will be harder to target users than before.
At Bytro (mostly war strategy games) they’ve invested more in influencer marketing since last year. Still direct response but a bit more: they look for influencers that are very well suited for their topics (e.g. WW3 game for which they look for influencers that talk about WW3 hypotheticals).
They also expect, and are starting to see, a positive branding effect from influencer videos as well as new users coming from older videos.
Michael
Challenges on direct response because of ATT are not a nail in the coffin: there are still many ways to reach users. But it did push everybody to get up to speed and build data models to measure incrementality.
[💎@07:40] TV/OTT might have been disregarded in the past, but now ATT is opening the door to this kind of channel.
Mary
[💎@09:00] With one of their newer titles, Game Hive has started looking at performance on an eROAS basis: attributing all organic uplift to paid efforts. eROAS = (Net IAP Revenue from both Inorganic & Organic cohorts + Ad Revenue from both Inorganic & Organic cohorts) / (UA Spend on the Inorganic cohorts).
They’ve found great success with influencer marketing as well.
[💎@09:50] Ad creatives that incorporate the influencer (especially big ones) perform very well, which means influencer marketing helps with the channel itself but also supports UA.
Eric
Media buying (e.g. TV for MZ) is not just about driving installs, it’s also about increasing conversion (CTR, CVR).
Mary
Still pumping out a lot of creative concepts.
[💎@12:02] Game Hive’s creative director analyzed all the patterns of all the creative concepts they’ve tested so far (including top and low performers), added that on to a matrix and tagged creatives to find which next creative concept would be the next high-impact creative.
[💎@12:36] The big catch is a playbook created by Facebook advocating for motivation-based creatives, rather than pushing out creatives to the same audience of players. It helps figure out who really are you players and what can motivate them to try your game.
Michael
iOS 14 means there are constraints as to how much you can test at once.
There are tradeoffs (big and small) in terms of what you’re exploring creatively (e.g. is it a new concept, just a new variant, just changing colors, etc.) as well as trade offs on the production costs (effort required).
[💎@14:18] Headlight looks for high leverage, big incremental wins (often radical shifts in concepts or positioning) combined with low production costs. They often start with static images and invest more effort if they see good results.
Eric
A lot of studios make the mistake of only optimizing for IPM and not wanting to commit to getting a ROAS measure.
Michael
[💎@15:49] In midcore games, highly profitable creatives often have lower IPM or CTR: they perform from a ROAS perspective because they resonate with the core audience. Similarly, the higher CTR creatives typically end up leading to lower in-game engagement. Don’t make the mistake of looking just at IPM.
Linus
[💎@16:52] Finding creative winners is more about finding ideas introducing big differences and iterating in big ways rather than small incremental changes.
Dedicate each week time to find ideas and plan what they can test next.
Michael
Marketing and product typically look at cohorts and metrics differently. There is usually a long process to align on what performance looks like and identifying where feature changes are supposed to impact performance. This is especially true when it relates to changes impacting late monetization and how that ties in with new user acquisition.
Mary
Still a small team so it is easier to collaborate. It’s a balance between the product team wanting to release more content to increase LTV and the UA team wanting more things to grow the user base.
“You can’t scale up if the early-game metrics look bad”
Alignment is also important for potential new events, what you could push in UA, etc.
Linus
Understanding of the data set on both teams is very important so both teams understand where the game is at.
They are now thinking about making their game more influencer-friendly: slow games that take weeks to complete don’t make good candidates for influencer streaming, so they’re thinking about adapting the game to the channel.
[💎@21:31] Bytro has had a specific messaging UA (war inside the US) that has been performing very well but initially that messaging was not really well captured in the game itself. Once they started using the UA messaging in the tutorial, they saw an improvement in all the KPIs.
Michael
Marketing teams and agencies should not accept superficial targets/metrics. Similarly, product teams often do not look at the changes in the acquisition sources.
Michael
A lot of people want to have the stance that “real” works better but we will most likely see fake ads continue and sometimes succeed (e.g. tower strategy games).
Mary
Even if you don’t like them, you have to admit that sometimes things like spelling mistakes work well.
[💎@24:41] Fake ads and clickbait ads do perform well sometimes (e.g. spelling mistake in copy). However you should not go only in that direction and at least have a balance.
"Fake ads will not go away anytime soon."
Linus
Have seen weaker performance when moving away from accurately representing the game.
With iOS 14.5 you have to be even more precise with your creatives because that’s also how you target users (e.g. On Facebook AAA).
