Doing more with less: A conversation on automation in performance marketing   


This article is part of a Gaming Insights series paid for by Facebook. 


We spoke with mobile gaming marketers from Tilting Point, Tripledot Studios, and etermax to better understand artificial intelligence (AI) and machine learning. Our discussion with these industry leaders dives into how each has leveraged automation for their user acquisition and marketing strategies, what success looks like (and if they’ve experienced challenges), and how they’ve used Facebook automated solutions into the work they do.

Karis Ng (Facebook): Let’s kick things off at a high level.  You’re each intricately involved in performance marketing for your games on a global scale. With digital marketing seemingly becoming more sophisticated, what does that look like for your day-to-day digital marketing and campaign management? 

Jan Sawicki-Hughes (Tripledot): Like everything at Tripledot, we always take a data-driven approach. It is crucial that we make our decisions based on what the data is telling us. We constantly implement new and more sophisticated automation to ensure we are adapting and apply changes quickly. And timing is everything. We’ve found that optimizing through automation is the most efficient way of managing hundreds of campaigns, thousands of creatives and millions of impressions. Automating these relatively simple day-to-day tasks allows us to focus on more impactful decision making.

Yi Gong (Tilting Point): Every day begins with analysis on campaign performance and user behavior in order to identify opportunities to scale UA. We leverage this analysis to design tests around audience targeting, creative variations, optimization type, etc to help us find the right strategy for a global games market. This obviously requires collaboration with other internal teams, so we spend a lot of time working with data science, creative, and product to make sure we are prioritizing things correctly.

Nahir Garcia (etermax): It is particularly important to maintain a clear vision on how to attract quality users to our games at scale. We do this by staying up to date on measurement and acquisition tools, and, above all, how to be more efficient by being creative in the communication of our ads.

KN: While you’ve all achieved growth and success, sustaining high performance at scale does carry with it some challenges. Can you expand on what “scale” looks like for your games/business?   

NG: After we launch a successful game, it’s important that we not lose engagement, nor the quality of the users we acquire. Connecting and entertaining people around the world allows us to foster knowledge of these new titles. Yet, one of our biggest challenges is building awareness of our titles to other active etermax gaming communities in a way where we don’t become our own competitor.

Focusing on casual games with simple mechanics allows us to transform and capitalize on cross-marketing. The etermax gaming ecosystem allows our users to play their friends on trivia, then take time to play a crossword puzzle or a word search game.

JSH: Each of our games has a certain return threshold required for continuous investment. This investment is not just about marketing but about how our colleagues across the company are working to achieve that return. Marketing spend exists in a multidimensional price/cost/volume world, so scale for me is finding the sweet spot for that app of all of these dimensions to maximum overall return.

Saying that, I love seeing our apps in the top charts!

YG: Since we work with so many different genres, this kind of depends from game to game. But generally, a game is considered “at scale” when we are able to spend in the mid-high six figures with a ROAS positive projection across multiple channels. But that level is different based on the games addressable audience, LTV, and retention rate. Some of our very niche games may be “at scale” while only acquiring a few thousand new users per month, while others require hundreds of thousands, or even millions, of net new users every month to be successful at scale.

KN: I’d like to dig into machine learning and artificial intelligence now. What does automation mean to each of you for your user acquisition and marketing strategies?

NG: Campaign automation is key to being able to work on a large-scale user acquisition strategy. We consider it essential in our long-term strategy and we firmly believe that as long as we can re-allocate more and more of our budget in campaigns that work with artificial intelligence, we will definitely be more efficient in what we do.

JSH: As an overarching strategy we try not to repeat routine tasks.  We seek opportunities to automate every part of the business we can, focusing on the most impactful areas of lifetime value — acquisition, retention, and monetization.

YG: Automation means letting the computers do the busy work that takes humans much longer to do, which helps our UA team manage their campaigns in a more efficient and scalable way. We build automation tools to execute daily campaign optimizations like campaign creation, bid, and budget updates. It saves a lot of time when we have tons of campaigns to work with. These tools can help with creative testing as well, where the system identifies the best creatives so we can build variants of it.

KN: With performance marketing often being as much science as it can be art, where does the automation factor in here in 2020? 

NG: This year accelerated many changes at the corporate level, at the process level and in our ways of working.  Considering this, we believe that automation does not have to behave differently: incorporating it into our regular operation, today more than ever, is a daily job where we think every day how to get the most out of it.

On the other hand, we are certain that it has to be accompanied with a creative strategy that guarantees to get the most out of networks like Facebook. We really believe that this point is not minor and the key is that automation is in tune with an excellent creative strategy.

JSH: We are fortunate to be in the gaming sector, which is one of the industries that has seen an uplift in unit economics. Disruption creates both risk and opportunity if handled appropriately. What 2020 has shown is that performance marketing teams need to be effective managers of change. This requires close collaboration with all areas of the business (development, data science, creative, monetisation etc.) to minimize risk and maximize any opportunity impacted by these disruptions. Automation is part of this. Automation is not “set and forget” — it is dynamic and flexible based on the most timely data.

YG: The areas where we leverage automation the most are probably campaign optimization and creative production. These were some of the most time consuming parts of managing campaigns a couple years ago, so we decided to invest in proprietary tools to help us in this process.  Recently we developed a proprietary tool we call Pixel to help automate creative analysis and production. In addition to analyzing which elements of a creative are delivering the best results, the tool also automatically resizes creative concepts to fit any ad placement. This has significantly shortened the feedback loop to our creative team and allowed us to hit the ground running across all new platforms we are testing.

KN: Believe it’s safe to assume you have each leveraged app install ads with Facebook for some time now. How have Automated App Ads, which prioritizes machine learning and automated systems for app installs ads, played a part in fueling growth and performance for your games? 

NG: Automated App Ads campaigns have saved our team considerable time towards monitoring these types of campaigns. We were able to keep creatives that performed well in terms of volume for longer periods of time.  It has also greatly simplified the overall campaign structure.

JSH: Facebook Automated App Ads have been a huge success for us. We have significantly increased our scale as a direct result of these campaigns. For us, the key difference has been creative testing which we spend a lot of time on, with quite volatile results. Facebook has allowed us to upload all our best performing creative and to select the right creative to meet our goals.

Being in a very competitive space, it is necessary to continuously search for new avenues of growth. With Automated App Ads, we understood the value of the campaigns very shortly after launch. We saw very promising results and acted rapidly.

YG: The product beta became available shortly after we launched one of our biggest mobile titles, Spongebob: Krusty Cook-Off, and proved to be pretty successful right off  the bat. We tested install, purchase, and value optimizations with the product.  The install campaigns yielded much lower CPIs compared to our standard app install campaigns which helped us push to the top rankings in the App Store in the first week after launch.

KN: Any key takeaway on the product so far worth sharing?                                                                

JSH: I think, one of the most crucial takeaways is that the importance of the right message and the quality of your marketing creatives is unparalleled. No amount of automation and algorithm sophistication will allow you to achieve great results without a clear and relevant marketing creative strategy. Once you have found the right creatives, it is paramount to keep searching for the next “big hit”, as it is an ongoing process and you must avoid creative fatigue.

YG: This has made campaign creation and management much easier for us. Allowing the algorithms to find the best users, we don’t have to put as much thought into defining audiences that might not translate across different countries or cultures. It is also more flexible in terms of creative testing, using up to 50 creatives at a time which is much higher than the typically recommended number of creative assets. This also allows us to use more ‘risky’ creatives that we might not prioritize otherwise.

KN: Before we let you three go, any final advice you can impart on fellow app and performance marketers looking to weave automated systems and machine learning into their marketing?

JSH: Do it! Start with simple high impact rules and grow from there.

NG: The automation of marketing campaigns, whether with first or third-party tools, requires algorithm training time and prior strategy creation.  But it is undoubtedly time very well invested that will allow you to enhance the performance of any team in the future. Our recommendation is always to start that path as soon as possible, to make effective use of time and obtain more chances to lower acquisition costs.

YG: Automation via machine learning and AI can have an enormous value on UA, if you have the data needed to feed algorithms. Tools need to be tested a lot before they become effective, and all machine learning tools are built on previous failures. So before investing in building tools for yourself, analyze where you may benefit most from automation. In addition, automation tools need to be updated with market changes. Building automation is a continuous process once you start, so consider the pros and cons of maintaining these tools before investing in building something yourself.

Automated App Ads is available for the app install objective through Ads Manager. Visit our website to learn more and get started.


BIOS

Karis Ng, Vertical Solutions Marketing Manager at Facebook gaming

Karis focuses on empowering gaming businesses to grow with Facebook Gaming solutions. Karis has been in the advertising and marketing space for nearly a decade, holding different positions from product marketer, brand manager, and strategist across gaming and technology companies.  

Yi Gong, Senior Growth Marketing Manager at Tilting Point

Tilting Point is a leading mobile game publisher in the global market where we power up mobile titles in a scalable way. As a senior UA manager here, Yi oversees the growth strategy across the channels for games like Spongebob: Krusty Cook-Off, Star Trek Timelines and TerraGenesis. In addition, she manages multiple analytics projects to help grow user LTV, retention and monetizations.

Jan Sawicki-Hughes, Growth Marketing Manager at Tripledot Studios 

Jan is a Growth Marketing Manager at Tripledot Studios. He looks after their User Acquisition efforts and has seen tremendous growth on their titles such as Solitaire, Woodoku and Blackjack. Tripledot Studios is an independent mobile games studio that takes a data-driven approach in all areas of the business. With a proven performance marketing strategy, they drive growth and deliver games that everyone can enjoy.

Nahir Garcia, Head of User Acquisition at etermax

Nahir joined etermax in 2016 as Head of User Acquisition. She was in charge of the acquisition strategy for various games in more than 30 markets in the United States, Latin America, Asia and Europe, leading a team of more than 15 people. Nahir has 15 years of experience in the technology industry and during her career held different positions related to online marketing, mobile advertising, SEM, Social Media Marketing and ASO.


VB Lab Insights content is created in collaboration with 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 sales@venturebeat.com.



Source link