Recent Projects
Mini Bio
My PhD research was in computer science and complexity science, although I also published in web-technologies (and even paleontology!) I have founded, run and sold a high-tech business based on my research and written two technical text-books published by Elsevier.
Predicting Viral Marketing Spread
December 2005 - October 2006
For this project I developed a system that allows advertising agencies and marketers to dry-run viral advertising campaigns. The software provides detailed best-case, worst-case and most-likely predictions and helps identify weaknesses in the advertising strategy.
An old advertising adage says "50% of my advertising budget is wasted - I just don't know which 50%." Despite being haled as a lost-cost alternative to conventional advertising channels, viral marketing is a very fickle technique. The cost of reaching someone in a successful campaign can be tiny, but even a high budget failed campaign can reach only a few tens of consumers. Tools for predicting the reach and repetitions (i.e. the number of people and how often they see an ad) for conventional advertising routes are relatively mature, but nothing existed for viral marketing.
I developed a system that allows my client to run viral marketing campaigns in simulated population, using real world data from their research and licensed demographic data. The simulation can be run many thousands of times, allowing the agency to form a picture of the range of possible outcomes and the critical factors to making the campaign a success.