Skip to Content

The New Science of Computational Advertising

The web has changed the nature of advertising. Now computer science is set to revolutionise it all over again.

There was a time, not so long ago, when planning an advertising campaign was a black art based on past experience and informed guesses. The pricing strategies for publishers was equally opaque and usually the result of considerable negotiation and horse-trading. 

All that changed in the late 90s with the advent of search-related advertising. All of a sudden, advertisers were able to target their messages with pinpoint accuracy to individuals with an interest in a specific topic.

Google is undisputed king of the online advertising world having pioneered search-related advertising, the auctioning of advertising slots by keyword to the highest bidder and the development of increasingly sophisticated computer-based tools that advertisers can use to get the biggest bang for their buck.

As a result, advertising is no longer a black art but an emerging science in its own right. Today, Shuai Yuan and buddies at University College London review the new field of computational advertising and discuss the research challenges that it faces.

Yuan and co point out that there are essentially four groups of stakeholders in this area: advertisers, online publishers, ad exchanges and web users. 

Many stakeholders are actively researching the future of computational advertising. Social networking companies, in particular, have bet the farm on being able to extract more “value” from their users. 

One problem here is that much of the data is proprietary–perhaps we’ll never know exactly how these organisations exploit their networks for profit.

On the other hand, individual web users are having an increasingly important influence on this debate and the way their information is exploited for profit. Many individuals are removing details and taking measures that prevent advertisers from identifying them, such as blocking the use of cookies. This is a running battle that is unlikely to be settled soon. 

Then there are the advertisers who want increasingly sophisticated tools to plan campaigns, to ensure value for money and to monitor their return on investment. 

For example, Yuan and co say there is increasing need for a futures market for advertising space that allows advertisers to take an option on advertising space several months in advance but without being obliged to buy the space  when the time comes.  

This has long been in print media where travel agents, for example, need to know they can advertise their products in the run up to the holiday season. 

Publishers ought to be willing since selling futures allows them to lock in a profit if they think prices are likely to fall. 

Such a market does not yet exist for online space meaning there is an opportunity here for somebody with the computational firepower willing to take on the considerable challenge.

Finally there are the publishers themselves who need more effective ways of measuring the effectiveness of their inventory so that advertisers can choose with greater precision where they can get the best return. 

Back in the heady days off 2006, the global market for online search advertising was worth some $10 billion. Since then the world has undergone a severe financial crisis and yet this market is expected to be worth over $70 billion by 2015. 

That’s a significant pot of money. And those with the computational fire power to best exploit this new world are likely to claim the majority of it. 

Ref: Internet Advertising: An Interplay among Advertisers, Online Publishers, Ad Exchanges and Web Users

Keep Reading

Most Popular

still from Embodied Intelligence video
still from Embodied Intelligence video

These weird virtual creatures evolve their bodies to solve problems

They show how intelligence and body plans are closely linked—and could unlock AI for robots.

conceptual illustration showing various women's faces being scanned
conceptual illustration showing various women's faces being scanned

A horrifying new AI app swaps women into porn videos with a click

Deepfake researchers have long feared the day this would arrive.

protein structures
protein structures

DeepMind says it will release the structure of every protein known to science

The company has already used its protein-folding AI, AlphaFold, to generate structures for the human proteome, as well as yeast, fruit flies, mice, and more.

Stay connected

Illustration by Rose WongIllustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.