The Power of AI in OOH Advertising

Gone are the days when processing vast amounts of data and employing sophisticated measurement capabilities were considered exceptional in advertising. Today, AI is at the forefront of innovation in OOH advertising, enabling micro-targeting, predictive modeling, and ultimately delivering more value for advertisers' money.

Josko Grljevic, Chief Operating Officer at Talon, a pioneering OOH agency, emphasizes the need for new and innovative approaches in leveraging AI to predict audience behaviors with unprecedented accuracy. The automation and simplification of OOH processes make it accessible and intuitive for advertisers, positioning it as a "brilliant broadcast medium."

Shifting from Demographics to Behaviors

In recent years, there has been a significant shift from planning campaigns based on demographics to understanding and targeting audience behaviors. Grljevic highlights the limitations of relying solely on demographics, citing the example of two individuals who, though demographically similar, may have vastly different interests. AI facilitates a more nuanced approach to audience targeting, ensuring campaigns resonate effectively.

Hyper-Targeting through Data Analysis

The COVID-19 pandemic served as a catalyst for OOH advertisers to delve deeper into data analysis. Talon capitalized on this opportunity by developing its data management platform, Ada. Through detailed analysis and hyper-targeting, Talon can rank the effectiveness of every OOH frame in the UK, providing advertisers with granular insights into audience behavior, dwell time, and proximity.

Independently verified outcomes showcase impressive results, with campaigns experiencing a 13x uplift and a remarkable 79% increase in purchasing intent. The granularity of data even challenges traditional assumptions, revealing, for instance, that major city-center shopping days may vary depending on the day of the week and time of year.

The Science of Prediction

Talon's commitment to innovation extends to predictive modeling. By building a time series predictive model, they aim to enhance their ability to predict audience behavior. Grljevic reveals that the model boasts a 90-92% success rate when predicting up to 90 days ahead, demonstrating its effectiveness in comparison to historical data.

Multi-Dimensional Modeling for the Future

Talon's journey with AI doesn't stop at predictive modeling; they are actively working on expanding the speed and capabilities of their AI learning. The goal is to eliminate manual and time-consuming tasks, allowing their talented team to focus on creativity and delivering outstanding outcomes for clients. The mathematical models involved are becoming increasingly sophisticated, drawing inspiration from elements of the natural world.

Automated, AI-powered models are proving to be up to ten times more effective than manually planned campaigns. The emphasis is on speed and simplicity, mirroring the user-friendly and agile nature of online advertising. The vision is to create a seamless user journey, enabling advertisers to effortlessly state their desired outcomes, target audience, and campaign timing, with the ability to automate the buying and transaction processes.