Case Study: Leveraging Metadata Optimization in Connected TV Advertising

Introduction:

In the face of evolving privacy regulations and the diminishing effectiveness of traditional audience targeting, Rain the Growth Agency has been leaning into future-forward privacy by design targeting strategies. Utilizing machine learning (ML), we have developed custom algorithms to optimize programmatic and digital video advertising, with a particular focus on Connected TV (CTV).

Objective:

For our client, a well-known technology platform and a very privacy-conscious brand, our primary objective was to drive account creations as measured by iSpot through CTV advertising, while navigating the challenges posed by privacy concerns and shifting targeting capabilities. Rain aimed to develop a solution that leveraged metadata optimization to deliver measurable results while ensuring compliance with privacy regulations.

Background on Metadata:

In programmatic advertising, metadata refers to the information exchanged in real-time auctions for ad impressions. This data includes a wide range of information about each impression, such as device type, location, content context, and user demographics. By analyzing metadata, advertisers can gain valuable insights into the characteristics of each impression and make informed decisions about ad placement.

Approach:

Rain’s custom algorithm solution focuses on processing and analyzing metadata to identify key drivers of performance without relying on personally identifiable information (PII) or traditional audience segments. Our algorithms create tailored buying strategies for each client and campaign based on these insights, offering transparency and control over the optimization process while maintaining strict privacy compliance.

Implementation:

In the case of our client’s CTV campaign, Rain’s custom algorithms processed granular impression-based log files to identify specific metadata signals correlated with conversions. By prioritizing ad opportunities based on these insights, we optimized the bidding strategy to maximize performance. The algorithms continuously refined their approach through real-time feedback, adapting to changing market conditions.

During the course of data analysis and optimization, we discovered that, in descending order, ZIP code, time of day, genre and channel are the primary factors of conversion in connected TV. Once the algorithms understood which variables within those fields were over indexed for conversions, the algorithms optimized pre-bid to the right mix of metadata signals, i.e. prioritizing impressions within:

• Urban zip codes
• Delivery during primetime evening hours
• Documentary & reality TV genres

Results:

The impact of Rain’s metadata optimization approach on client’s CTV campaign was significant:

• Lower Cost of Acquisition: Rain achieved a notable 47% reduction in Customer acquisition cost (CAC) compared to historical performance.
• Cost Savings: Our algorithms delivered a 16% reduction in CPMs compared to the average campaign CPM, resulting in improved cost-effectiveness and increased reach.

Conclusion:

Use of metadata optimization represents a pragmatic solution to the challenges facing CTV advertising in an evolving privacy landscape. By prioritizing data-driven insights and privacy compliance, our approach offers a reliable alternative for advertisers seeking effective targeting strategies without compromising consumer privacy.

Industry Ramifications:

This case study highlights the potential of metadata optimization to transform advertising in the face of privacy concerns and shifting targeting capabilities. As marketers adapt to changing regulations and seek alternative targeting methods, Rain’s approach provides a model for achieving results while upholding the highest standards of privacy and compliance.

This article is featured in Media Impact Report No. 54. View the full report here.