RTB House, a globally operating Polish company specializing in innovative retargeting technology, has implemented new algorithms that allow for highly accurate estimation of the CTR rate.
This translates into the ability to more precisely predict whether a user will click on an ad, resulting in a higher return on investment for clients (ROI). RTB House was one of the first retargeters to implement deep learning in such an advanced way – the most promising field of research into artificial intelligence.
CTR, the rate determining the ratio between the number of clicks and impressions of an online ad, is one of the most popular measures of an effective advertising campaign. That is why technologies that make it possible to increase CTR and reach potentially the best customers while maintaining the same budget enjoy great popularity. Algorithms play a key role in this process, enabling the prediction of users' purchasing behavior based on their previous activity on a given e-commerce site and recommending the best offers to them.
Deep learning, an innovative approach to data processing inspired by the way neurons work in the human brain, makes it possible to obtain more reliable and richer information about the purchasing potential of different users, and this without human intervention.
The algorithm based on deep learning implemented by RTB House allows for a better estimation of the probability of users clicking on ads (CTR), which translates into an increase in the aforementioned rate by as much as 16.5 percent without the need to change the budget framework of campaigns. As a result, this leads to a higher return on investment for advertisers.
The module determining the value of the CTR click probability is the fourth significant deep learning implementation by RTB House. Thanks to the earlier implementation of deep learning, the modules estimating the probability of conversion and its value are able to increase the overall effectiveness of retargeting activities by as much as 29 percent. Moreover, recommendations based on deep learning can match the displayed products to the purchasing preferences of potential customers with up to 41 percent greater accuracy, compared to campaigns that had not previously used such methods.
Bartłomiej Romański, Chief Technology Officer at RTB House, emphasizes the importance of using deep learning in retargeting technology at every stage of the advertising process. – For more than a year and a half we have been working on the gradual modernization of our solutions. This has brought us to the point where we can already say that 100 percent of our algorithms are based on deep learning. This translates into measurable benefits for clients and an increase in the effectiveness of their online advertising activities. This is particularly visible in industries such as tourism, where a lot of data has to be analyzed and purchasing patterns are dynamic and difficult to predict. In such cases, self-learning algorithms can determine users' needs much more precisely and accurately. This is a definite improvement compared to other methods currently used in retargeting – Romański concludes.
RTB House is one of the few companies in the world that has managed to develop and implement its own technology for purchasing ads in the RTB model. The company operates globally and runs more than 1,000 unique campaigns for international brands in over 40 markets in Europe, South America, Asia and the Pacific, the Middle East and Africa.
The company's discoveries in the field of artificial intelligence were presented, among other places, during the international conference devoted to neural networks, the International Joint Conference on Neural Networks in Anchorage. Earlier, RTB House, as the only company representing Poland, was selected to lead a workshop session as part of the largest international industry conference devoted to the practical use of artificial intelligence – the International Conference on Machine Learning in New York. Its representatives also took part in the 31st edition of the AAAI Conference on Artificial Intelligence in San Francisco.
