In association with TAPTAP Digital.
Digital advertising, particularly programmatic, has revolutionised how ads are served and enabled the collection of data on a scale that would seem scarcely believable just a couple of decades ago. However activating this wealth of data, along with linking it to offline activity, has been an inefficient process up until now, restricted by low-quality traffic and fraudulent activity.
Artificial intelligence (AI) and machine learning (ML) capabilities are offering a promising solution to these issues. In this article for ExchangeWire, Álvaro Mayol Garrido, partner and chief product & technology officer at TAPTAP Digital discusses how AI can be used across the advertising ecosystem to enable more efficient media spending and to improve user experience.
The way in which digital media is bought and sold has changed dramatically in recent years. The programmatic advertising ecosystem has gone from being an experimental technology, to being the epicenter of all decisions that are made in advertising transactions.
Artificial intelligence and machine learning have begun appearing on the cover of almost all magazines, blogs, and technological articles, as the most fashionable concept in the digital world where technological evolution is exponential. More specifically, these new concepts represent a revolution in the present and near future in the worlds advertising and marketing. AI allows machines to perform tasks with a high procedural and computational load so that people can be free to provide maximum added value in matters such as creativity or strategic management.
All this intelligence is key due to the proliferation, which is increasingly established, of programmatic advertising. These new digital trends are placing the user as the axis of their actions, posing a new paradigm in the advertising market. To date, the great effect of branding and awareness offered by offline media hasn’t been interconnected, with the deep capacity for segmentation, interaction and measurement of digital media, especially within the mobile medium. Little by little, we’re seeing the great value that mobile-centric strategies offer in interaction with users in all phases of the purchase funnel, and through any media. These interactions are cross-media and cross-device, but are increasingly digital in nature. In turn, all this allows us to move from less effective mass-media communication (one-to-N), to a more personalised one that impacts the user at the most appropriate time (one-to-one), thus reaching communication or interactions that are more effective and relevant.
In an ecosystem where personalisation in communication begins to be key to optimising the return on investment in media, these techniques allow us to know the context and mindset of users more accurately, in order to increase relevance and usefulness of the message. We’ve evolved towards an expert system that, in real time, analyses the probability that a user converts, buys from or interacts with a brand based on a multitude of different data sources. Among these user data-points is their location, the context, their demographic profile, and digital and behavioural interests.
In Sonata (our proprietary DSP & DMP), thanks to having our own technology, we’ve evolved our processes of analysis and data processing, in order to build audience profiles and pre-bid systems with a greater degree of specialisation, depending on the concrete moment of the purchase process each user is in. We start from unsupervised models of machine learning (clustering) to later refine algorithms based on supervised machine learning models, where we include real data samples, as well as genuine advertising interactions for each vertical or category and, of course, all the brand’s research and knowledge of the user.
It’s important to know how to incorporate and weigh other quality sources of certified, guaranteed external data so that we can implement systems which allow us to identify low-quality, or fraudulent, inventory and advertising traffic. This is a key objective in minimising losses and optimising results in an environment where fraud is increasingly complex and difficult to detect. Big data analysis, AI and process automation allow us to refine and certify, with pinpoint accuracy, the signals and decisions we make when activating advertising and reaching consumers.
A key variable of immense value, both in carrying out the segmentation, activation and attribution of audiences at digital and offline levels, and in implementing mobile-centric strategies, is the location. To measure this, Sonata has an audit and classification system for the location signals available in the advertising ecosystem called LQI (Location Quality Index). Thanks to artificial intelligence and the daily analysis of billions of pieces of data on a global scale, the expert system is able to discard all data of fraudulent origin or of low quality, taking into account not only the user’s location, but also dozens of additional variables, such as advertising interactions, viewability, ad placement, origin and quality of the media, etc.
This has allowed us to continue innovating and developing different systems in the field of Geospatial Intelligence. This new evolution towards knowledge and the development of expert systems that make intelligent decisions, based on the study of different data variables in hyperlocal environments, allows us to help brands, and especially retailers, make the most optimal decisions while increasing the ROI of their investments in advertising or marketing. You can know, in real time, which are the most compatible zones or areas in which to carry out an activation or communication depending on audience, traffic, competitive analysis or the arrangement of offline media such as OOH.
Artificial Intelligence not only helps us in the engineering of intelligent systems focused on decision-making when buying or serving advertising, or for the creation of more complex and accurate audience profiles, but it’s also key to the optimisation of results and maximisation of the KPIs and ROI of each strategy. Through the execution of Machine Learning algorithms that analyse millions of pieces of historical data, our platform evaluates and executes decisions in real time that allow us to optimise different variables at the same time, acting at each moment based on probabilities and forecasts of success or conversion. All this, in order to provide the best possible results to advertisers, and to provide the greatest utility and relevance to consumers who receive the content, which is increasingly adapted.
Artificial intelligence is here to continue to revolutionise industries such as advertising, with the aim of optimising results and reducing costs, both on the supply side and on the demand side. But it not only gives us financial and statistical benefits, consumers will increasingly benefit and will receive more and more personal, relevant and useful messages accordingly.
In short, Sonata, like the market, is evolving along with programmatic advertising trends so it can implement solutions and strategies beyond the borders between digital and offline. But it’s not about completing or unifying a vision about the media, it’s about concentrating the media around their natural link, the user. Being able to identify this user as the fundamental axis of any conversion or sale process, we began to abandon the language of advertising to make way for models oriented to real business objectives.