With 3rd-party tracking cookies on the way out, new solutions are needed, and Deep Learning could be the key.
Advertisers and advertising agencies are facing a very different world in the wake of Google’s decision to retire 3rd-party tracking cookies. A cookieless future makes delivering personalized ads significantly more challenging, with over 47% of industry professionals citing finding a new solution in a post cookie world as very important. One of the most promising ad-tech solutions is the next evolution of machine learning: Deep Learning
What Does The End of 3rd-Party Tracking Cookies Mean For Advertisers?
The end of 3rd-party tracking cookies has been on the horizon for some time. Very real concerns regarding user privacy meant that many major browsers, such as Mozilla Firefox, had already made the decision to drop 3rd-party cookies without providing an alternative targeting method for advertisers, post cookies. Google’s decision to sunset these tracking cookies was the final nail in the coffin.
The loss of the rich-data these cookies provided presented an enormous challenge for advertisers. This is why Google took it upon themselves to build an anonymized ad-tech solution to 3rd-party tracking cookies called the Google Privacy Sandbox which would address a cookieless future.
We have discussed how the Google Privacy Sandbox works in the past. RTB House has been intimately involved in the project from day-one and was the first Demand-Side Platform to successfully use the system to globally buy real advertising impressions.
The challenge with a cookieless future and these new approaches is that traditional machine learning systems struggle with these newer, more complicated, anonymized data-sets. Fortunately, RTB House has already built the ultimate tool for campaigns dedicated to every stage of the funnel in the cookieless world: Deep Learning.
Deep Learning Is Uniquely Well Suited To Interpreting Anonymized Data
Deep Learning is an evolution of machine learning technology. The word “Deep” refers to the number of layers through which data is processed. Each layer is able to extract progressively more complex features from the raw input.
In the context of targeted advertising, lower layers might identify a user-group, while a higher layer may attempt to identify a specific interest within that group. This enables RTB House’s
Deep Learning algorithms to effectively determine what position a user is within the sales funnel, and also select the most appropriate content for that specific user.
The best thing about Deep Learning? The more often you use it, the better it becomes. It is a step above standard machine learning as Deep Learning algorithms are able to identify patterns without human intervention, which makes for much more efficient self-learning.
Like other machine learning algorithms Deep Learning improves based on the data it’s fed with. Where Deep Learning really stands out as an ad-tech is its ability to work with large, complicated datasets, even unstructured datasets, without requiring a human to define what patterns it needs to look for.
This approach has yielded real results. RTB House’s Deep Learning algorithms, as per RTB House campaign insights September 2021, have been able to drive up to 47% higher video completion rate and up to 33% better viewability compared to global market benchmarks. The solution has also been the recipient of numerous high-quality industry awards and certificates.
RTB House Deep Learning Is Already Being Used In Real World Campaigns
RTB House has been using Deep Learning actively since 2017 and is currently using Deep Learning algorithms in 100% of its campaigns. This has proven to be effective and RTB House has been able to drive significantly improved results for clients over the past 4 years.
Doubling OLX’s Incremental Conversions
One high-profile example is the campaign with OLX, a global classified company that makes it easy to buy and sell a variety of products, with over 330 million visitors to their websites every month. RTB House was able to build a personalized recommendation engine that helped to double OLX’s incremental conversions.
The solution enabled OLX to reach users that they would not have been able to reach organically, and RTB House’s Deep Learning powered retargeting solution has since become an important strategic tool for the OLX team.
Improving Trivago’s Retargeting Outcomes
Trivago is a leading hotel search website that compares rates for over 1.8 billion properties across approximately 190 countries. Trivago used RTB House’s Deep Learning technology in order to build a personalized recommendation solution for its users looking for accommodation, with efficiency and scalability at the core of the product’s design. All while maintaining user privacy and data protection.
The solution quickly helped to improve retargeting outcomes. Ads also fit each user’s preferences better and helped to improve budget efficiency significantly. Trivago and RTB House now cooperate in over 30 markets.
Enticing Lapsed Users Back To Lazada
Lazada is the number-one online shopping and selling platform in South East Asia. They turned to RTB-House as a partner to help them make their app channel even more efficient, with a particular focus on re-engaging lapsed users and increasing overall platform retention.
RTB House used Deep Learning to segment users based on when they last accessed the app. They were then able to set custom strategies for each segment and run category campaigns where only offers on sale were shown. This approach yielded much stronger reach, and increased the number of daily app users, with cost per visit decreasing by 60%.
Implementing Deep Learning Will Help To Future Proof Your Company
It’s clear that Deep Learning is already more effective than other existing solutions. With the removal of 3rd-party cookies on the horizon, it is essential that you begin to prepare for that post cookie shift as quickly as possible. This will enable you to optimize your processes, and ensure that you are not left bug-fixing a campaign after 3rd-party cookies have expired.
RTB House has already begun this effort. Our Deep Learning algorithms have proven their worth during performance campaigns, and we are already implementing them into other parts of the sales funnel. We are actively using Deep Learning in awareness and branding campaigns, which require a very different approach. Our algorithms can be specifically tailored to maximize any combination of viewability, VCR, reach, or CTR, to maximize the value of awareness and branding campaigns.
Don’t find yourself left behind. Contact RTB House today and discover how Deep Learning can help you get the most out of your marketing campaigns.
If you have any questions or comments regarding the article, please let us know.