One of the things we, as mobile evangelists, have always struggled to understand is how marketers can hold our industry accountable to standards set forth by digital; a medium that's been around for well over 15 years and has just recently begun to standardize and stabilize. It seems unfair and, at times down right idiotic, but we've come a long way in the last 5 years and we continue to make big strides in the marketplace. A great example of where mobile has begun making strides is in the programmatic buying category.
If you haven't been paying close attention recently, the world of mobile Real Time Bidding (RTB) has all but exploded in the industry. Out of the 50 or so networks I've met with in the past month and a half, at least 10 have had some sort of RTB opportunity to present, which is completely took me off guard. Now, if you've never done any work in the RTB space, these guys quite literally speak a different language and it can sometimes be enough to make your head spin. Between the tracking, data mining, third-party data layering and otherwise, it can get rather intimidating.
For those of you who need a quick once around the block, here's how it works:
A user visits a page on a publisher’s site or opens an application on their mobile device. Simultaneously, the publisher/developer sends out a “bid request” to thousands of potential advertisers that says, “We’ve got this user who is 25, Japanese, female, based in Los Angeles and recently searched for a Toyota Tundra within our site/app. How much do you bid to be the only ad on this page?” In around 100 milliseconds the site/app receives bids from dozens of advertisers, which it then analyses to figure out the highest bidder and corresponding ad. The winner is notified by the publisher/developer and allowed to place its ad on the page.
This entire series of back and forth takes place in 300-500 milliseconds, causing no visible delay to the user, and is repeated for every ad slot on a page. This ads up to billions of ad requests and terabytes of data for these companies to process on a daily basis; all wrapped up in a nice little bow for the advertiser on the back-end.
An emerging industry leader I've been following has taken things to an all new level. AdTheorent, Inc., an intelligent RTB-enabled mobile ad network has recently released the industry's first real-time learning and predictive modeling platform. The AdTheorent Real-Time Learning Machine™ (RTLM) learns in real time, generates data-driven predictive models "on the fly" and predicts faster than any other data mining technology. They've essentially taken a process that, in the past, has taken humans days (sometimes weeks) to accomplish and automated it to yield tangible results for mobile advertisers. According to AdTheorent, during the 2012 holiday shopping season, the AdTheorent RTLM delivered an average engagement level of 200-300% above industry benchmarks for leading retail brands.
This data mining technology was developed by AdTheorent's Chief Data Scientist, Dr. Saed Sayad and currently the application analyzes 50,000 bid requests per second on a single server. It filters out bids with a low probability of click, conversion or awareness lift and "learns" from incoming bid requests in order to build and modify predictive models on the fly.
What does this mean for brand advertisers?
"The value of data in media is immense and intelligent technology companies are now using that data across mobile devices. This is enabling real-time learning and prediction, which is presenting a great opportunity to move marketers toward data-driven results." These new developments offer real promise in enhanced performance and brand lift; a combination that will spark interest across the mobile marketplace."
- Sal Candela, Mobile Director at PHD
By applying such models in-flight, AdTheorent's RTLM can match each mobile advertisement with the optimum mobile impression. As a result, the system filters-out undesirable targets, allowing participating advertisers to enjoy improved engagement levels such as: uplift in click through rates (CTR), increased awareness and an overall reduction in cost per acquisition (CPA). In some campaigns the uplift has been as high as 500% and, in a recent CPA engagement, the technology was able to run 2000 model variations in under five seconds to extract the most efficient model for prediction of conversion events.
It's important to note that the RTLM isn't just about segmentation and more efficient impressions on the front end. The new platform leverages AdTheorent's recently introduced Traktion™ product, which affords mobile advertisers a seamless way to track the ever-illusive post-click behavior of consumers across the entire spectrum of mobile channels. The power to learn in real time, model "on the fly" and predict faster than any other data mining technology helps to distinguish AdTheorent from other mobile ad networks.
"RTB companies face a big data problem in modeling terabytes worth of data every month. RTLM is the only solution that can scale to meet the big data demands to produce result driven models in the RTB world. Our RTLM system is designed for mobile advertisers, and we anticipate a greater willingness from advertisers seeking to deploy real-time bidding in the coming year."
- Anthony Lacovone, CEO at AdTheorent
Below is an excerpt from AdTheorent's recent press release discussing exactly how their Real-Time Learning Machine works:
The AdTheorent RTLM leverages technology that processes Big Data for real-time analysis and scoring, based on criteria including advertiser's demographic data, geographic data, publishers' data and other information. The RTLM enables variables to be added or removed from the analysis as data evolves so that, for example, if one demographic data point were removed (e.g., "women age 25-49"), the RTLM would almost instantly re-calibrate the predictive model without that data.
Conventional data mining algorithms operate in a batch mode, where having all of the relevant data at once is a requirement. Due to large increases in the rate of generation of data, the quantity of data and the number of attributes (variables) to be processed, the data situation is, increasingly, now beyond the capabilities of conventional data mining methods. AdTheorent's RTLM uses a three-stage data analysis approach that refines billions of bid requests, removes extraneous data, and delivers the most accurate and efficient targeting for advertisers' mobile campaigns. The RTLM provides the only viable predictive modeling platform to process Big Data with zero-latency. The power and efficiency of AdTheorent's RTLM allows one data scientist to perform the work of 10.
AdTheorent's RTLM is unprecedented in mobile advertising, featuring:
- Incremental learning (Learn): immediately updating a model with each new observation without the necessity of pooling new data with old data
- Decremental learning (Forget): immediately updating a model by excluding observations identified as adversely affecting model performance without forming a new dataset omitting this data and returning to the model formulation step
- Attribute addition (Grow): Adding a new attribute (variable) on the fly, without the necessity of pooling new data with old data
- Attribute deletion (Shrink): immediately discontinuing use of an attribute identified as adversely affecting model performance
- Scenario testing: rapid formulation and testing of multiple and diverse models to optimize prediction
- Real Time operation: Instantaneous data exploration, modeling and model evaluation
- In-Line operation: processing that can be carried out in-situ (e.g.: in a mobile device, in a satellite, etc.)
- Distributed processing: separately processing distributed data or segments of large data (that may be located in diverse geographic locations) and re-combining the results to obtain a single model
- Parallel processing: carrying out parallel processing extremely rapidly from multiple conventional processing units (multi-threads, multi-processors or a specialized chip).
"We did not invent a new algorithm; we merely use data analysis in a different way. Traditional predictive modeling is a much more time-intensive endeavor, and is nowhere near as flexible as the approach that we have created. We can test 100 different predictive models in one second and then remove 80-90 percent of those noisy or correlated variables," said Dr. Sayad. "The real-time learning that is deployed in our system creates massive spikes on top of an existing lift in positive targeting, and this is the breakthrough that has produced the success that we are seeing in campaigns that have used this system."
About Dr. Saed Sayad
A pioneering researcher in real-time data mining and big data analysis, Dr. Sayad has designed, developed and deployed many business and scientific applications of predictive modeling. The author of ‘An Introduction to Data Mining,' Dr. Sayad teaches a popular graduate course in data mining at the University of Toronto, where is an adjunct professor.
AdTheorent is the world's first intelligent Real Time Bidding (RTB)-enabled mobile ad network, powered by a platform built from the ground up to address the specific needs of the mobile advertising ecosystem. AdTheorent's RTm™ Platform integrates its 20,000+ mobile inventory sources and analyzes hundreds of thousands of potential impressions per second based on highly enriched demographic information, behavioral factors, location data, device features, as well as other advertiser-specified targeting criteria. Using predictive modeling to identify impressions with a higher propensity for conversion and awareness lift, the AdTheorent RTm Platform places bids in real-time within the pricing parameters established by the advertiser. The result to brands and marketers is higher conversion rates at a significantly lower cost -- The Intelligent Impression™. For more information visit:www.adtheorent.com
Image Credit: Smaato
Dan Wittmers is the Mobile Manager for Initiative – NY and the Founder of the Mobile Leaders Alliance. He has a natural understanding of the entire mobile ecosystem, and during his tenure, has had the opportunity to work with Fortune 500 brands and agencies across North America. Educated in media, messaging, development, SaaS tools and predictive analytics, he is an emerging thought leader in the mobile industry.