RTB House, a globally operating Polish company specializing in innovative retargeting technology, has implemented a new recommendation mechanism. Thanks to deep learning (currently the most promising field of artificial intelligence research), products displayed on advertising banners will be matched more accurately, which in turn will increase campaign effectiveness by up to 50%.
One of the key indicators of the effectiveness of online advertising activities is CTR (the ratio of banner clicks to impressions), and its level depends, among other things, on the accurate selection of products presented on banners. The closer they are to the purchasing preferences of potential customers, the greater the chance of interest in the offer. In retargeting campaigns, what is shown on banners is decided by algorithms, while their effectiveness depends on the level of advancement of the technology used.
RTB House has developed an innovative recommendation mechanism based on deep learning, i.e. methods inspired by the biological activity of neurons in the human brain. This makes it possible to collect even more accurate information about user behavior and, consequently, to ultra-precisely match the products promoted on banners to the interests of a specific consumer. The entire process takes place in a fully automated manner, without human intervention.
Bartłomiej Romański, Chief Technology Officer at RTB House, emphasizes that the once futuristic concept of artificial intelligence is now essentially a permanent part of everyday life – Deep learning helps in the analysis of data that has so far been difficult to process and interpret, enabling the full use of the potential of big data. Increasing precision in the selection of products in creatives means that we can personalize ads even better. As a result, campaigns are simply more effective. This in turn brings the effect of a higher return on investment and confidence that advertising budgets are being used in the best possible way – he adds.
Operations related to data processing based on deep learning algorithms require enormous computing power, and the implementation of innovative methods is possible only thanks to technologically advanced solutions. RTB House used NVIDIA's deep learning software for this purpose.
Serge Lemonde, NVIDIA Deep Learning Startups Business Director EMEA and India, explains that deep learning is a very important field of artificial intelligence that enables computers to learn independently from data. – Thanks to the speed, precision and large scale of analysis, virtually every industry already sees the benefits of implementing artificial intelligence. RTB House is bringing deep learning into the world of advertising, and by processing and interpreting additional data thanks to it, it makes activities even closer to the real needs of audiences – Lemonde concludes.
RTB House is one of the few companies in the world that has created its own technology for purchasing ads in the RTB (real-time bidding) model. The company currently handles more than 1,000 campaigns in more than 40 markets across Europe, the Middle East, Africa, Asia and the Pacific, and Latin America.
The company participates in the NVIDIA Inception program dedicated to companies that are revolutionizing the industry through advanced work on artificial intelligence issues. This is RTB House's third algorithm based on deep learning. Previously, RTB House introduced a model that increases the probability of conversion and estimates its value.
