Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data

This thesis aims to identify how present day advertising agencies can compete in an increasingly disruptive technology-driven future. By investigating the early history of advertising and the subsequent modern evolution of advertising technology, it is determined that a new way of creative advertisi...

Full description

Saved in:
Bibliographic Details
Main Author: Phay, Han
Other Authors: Jesvin Yeo Puay Hwa
Format: Thesis-Master by Research
Language:English
Published: Nanyang Technological University 2020
Subjects:
Online Access:https://hdl.handle.net/10356/143187
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-143187
record_format dspace
spelling sg-ntu-dr.10356-1431872020-10-28T08:29:15Z Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data Phay, Han Jesvin Yeo Puay Hwa School of Art, Design and Media JesvinYeo@ntu.edu.sg Visual arts and music::Media This thesis aims to identify how present day advertising agencies can compete in an increasingly disruptive technology-driven future. By investigating the early history of advertising and the subsequent modern evolution of advertising technology, it is determined that a new way of creative advertising should evolve with the times. Through the integration of Artificial Intelligence and Big Data with current industry processes, a new kind of advertising agency will emerge that is run like a tech company. Using Big Data to inform creative design and Artificial Intelligence to identify patterns, predictive analysis can be applied to creative online banner ads for the purpose of increasing user engagement through higher click-through rates. This new methodology of data-driven design increases productivity for ad agencies, whilst increasing returns-on-ad-spend for clients. To support the hypothesis of design driven by data, a controlled experiment is carried out with digitally generated online banner ads. A total of 8 generated digital online banner ads were put through a 7 weeks adversarial A/B test on the Google Display Network, with one winning ad determined at the end of the period based on click-through % scoring. Performance data for the ads are presented, as well as a discussion on the implications of the findings, specifically on the positive potentials of data-driven design and the importance of ad agencies becoming like tech companies. The research is structured in 7 chapters: (1) A Historical Perspective on Modern Advertising, looking at the way creative advertising agencies have evolved throughout the earlier years, (2) Present Day Advertising Technology, continuing the narrative post-internet through exploration of the advertising technology industry in present day, (3) Consumer Psychology, specifically on how creative design attributes in ads, such as visuals, text and sound, can subconsciously have an effect on humans and are important to the cause of advertising, (4) Data-driven design, a new way of looking at the digital online ad creation process, integrating big data and artificial intelligence to predict the optimal creative attributes in digital ads, (5) Multivariate Testing for Creative Optimisation, a controlled experiment to determine that generated ads do not underperform current human designed ads, (6) Translating Research into Product, where the ad agency as tech company model is brought forth through the form of a startup, and lastly (7) Future Opportunities, that identifies other areas of emerging tech and the implications on future research. Master of Arts 2020-08-11T11:45:34Z 2020-08-11T11:45:34Z 2019 Thesis-Master by Research Phay, H. (2019). Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/143187 10.32657/10356/143187 en This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Visual arts and music::Media
spellingShingle Visual arts and music::Media
Phay, Han
Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data
description This thesis aims to identify how present day advertising agencies can compete in an increasingly disruptive technology-driven future. By investigating the early history of advertising and the subsequent modern evolution of advertising technology, it is determined that a new way of creative advertising should evolve with the times. Through the integration of Artificial Intelligence and Big Data with current industry processes, a new kind of advertising agency will emerge that is run like a tech company. Using Big Data to inform creative design and Artificial Intelligence to identify patterns, predictive analysis can be applied to creative online banner ads for the purpose of increasing user engagement through higher click-through rates. This new methodology of data-driven design increases productivity for ad agencies, whilst increasing returns-on-ad-spend for clients. To support the hypothesis of design driven by data, a controlled experiment is carried out with digitally generated online banner ads. A total of 8 generated digital online banner ads were put through a 7 weeks adversarial A/B test on the Google Display Network, with one winning ad determined at the end of the period based on click-through % scoring. Performance data for the ads are presented, as well as a discussion on the implications of the findings, specifically on the positive potentials of data-driven design and the importance of ad agencies becoming like tech companies. The research is structured in 7 chapters: (1) A Historical Perspective on Modern Advertising, looking at the way creative advertising agencies have evolved throughout the earlier years, (2) Present Day Advertising Technology, continuing the narrative post-internet through exploration of the advertising technology industry in present day, (3) Consumer Psychology, specifically on how creative design attributes in ads, such as visuals, text and sound, can subconsciously have an effect on humans and are important to the cause of advertising, (4) Data-driven design, a new way of looking at the digital online ad creation process, integrating big data and artificial intelligence to predict the optimal creative attributes in digital ads, (5) Multivariate Testing for Creative Optimisation, a controlled experiment to determine that generated ads do not underperform current human designed ads, (6) Translating Research into Product, where the ad agency as tech company model is brought forth through the form of a startup, and lastly (7) Future Opportunities, that identifies other areas of emerging tech and the implications on future research.
author2 Jesvin Yeo Puay Hwa
author_facet Jesvin Yeo Puay Hwa
Phay, Han
format Thesis-Master by Research
author Phay, Han
author_sort Phay, Han
title Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data
title_short Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data
title_full Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data
title_fullStr Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data
title_full_unstemmed Creative AI : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data
title_sort creative ai : a data-driven design approach for creative online ad optimisation using artificial intelligence and big data
publisher Nanyang Technological University
publishDate 2020
url https://hdl.handle.net/10356/143187
_version_ 1683494456013094912