It's All in the Technology & Algorithms


The history of digital advertising has demonstrated that technology is at the heart of it all. With the consolidation of buy-side and sell-side platforms, digital media runs at an exponentially larger scale, increasingly relying on algorithmic optimization. Over the last three years, Turn has invested heavily in our data-driven audience targeting approach, building complex optimization algorithms as well as the best infrastructure in digital advertising to run these algorithms in unprecedented speed and scale.

Our algorithm does three things in real-time:

1.       Intelligently explore inventories for each campaign along behavioral, contextual and social dimensions.

2.       Predict the precise values of each ad impression for each advertiser.

3.       Bid aggressively based on the dynamics across each inventory, auction market, publisher or type of page, etc.

Predictive audience models are an important part of Turn’s algorithmic approach to data-driven advertising. Turn’s models track every single user response (actions) across websites and every real-time-bidding (RTB) inventory source. To influence  each user interaction for each campaign, we statistically analyze different input variables such as time, location, browser ID, web page attributes, social media elements and user interactions with the brand itself. Because best-in-class algorithms require constant innovation, Turn continuously conducts A/B testing of new predictive models. As a result, in a recent performance competition for a top consumer electronics advertiser, Turn’s predictive audience approach netted 4x higher reach and 10% better performance than a competing DSP.

Predictive models are built to be highly sensitive to the statistical correlations between the input variables and the target variables (user engagement actions). Sometimes, the target is a composite variable that summarizes different types of user engagement actions.  Turn’s ROI Optimizer is designed to allow optimization to follow complex target variables precisely and automatically throughout the campaign cycle. Recently, one of the largest agency holding groups took advantage of the ROI Optimizer to achieve a 300% ROI boost on a complex campaign with multi-event plus shopping cart value optimization.

Last but not least, as a leader in RTB in both scale and sophistication, and as an algorithm driven bidder, Turn’s customers should expect successful campaigns to take full advantages of RTB inventory as well as traditional media inventory. This is accomplished by leveraging the same stack of sophisticated optimization and ad selection in RTB as well as in the non-RTB media buying environment. In addition, bids/costs in RTB become random variables that dynamically reflect the demand and supply of trading. Our RTB optimization algorithm acts like a seasoned trader on a stock exchange, it varies the bids minute-by-minute across different markets on different audience groups to achieve the best cost and reach for each advertiser. Leveraging the power of our RTB optimization, a top telecom provider recently launched and rebranded its high-speed ISP product line. It’s RTB portion out-performed the non-RTB audience targeting by 44%, and out-performance run-of-network by 562%.

In today’s online advertising, you can only rely on true technology and deep science to build full-scale, sustainable success. As the saying goes in Silicon Valley, it’s the algorithm, stupid. The Turn Platform will continue to innovate and integrate the best of breed algorithms and technology for our clients.