typically sell a fraction of that attention
Historically, advertising sales featured straightforward allocation rules
and manual negotiations. But now
more aspects of advertising, including
its sale, delivery, and measurement, are
being automated. Web-search engines
such as Google and Yahoo! have led
the way, selling space beside particular
search queries in continuous dynamic
auctions worth billions of dollars annually.
is automation changing the advertising industry? More information can be
found in the Proceedings of the Workshop on Ad Auctions series.
Auctions and exchanges for all types
of online advertising—including banner and video ads—are commonplace
at present, and they are run by startups
and Internet giants alike. Advertisers
can buy not only space but also contextual events—such as clicks from a specific user on a specific property at a specific time—or, more generally, bundles
of contextual events. An ecosystem of
third-party agencies has grown to help
marketers manage their increasingly
complex ad campaigns.
The rapid emergence of new modes
for selling and delivering ads is fertile
ground for research, both from economic and computational perspectives.
25 Edelman et al.
15 and Varian31
model how advertisers bid in search-ad auctions. Essentially, the advertisers raise their bids until they reach a
point of indifference between staying
where they are and swapping with the
advertiser above them on the page. The
authors show that this bidding strategy
forms the basis of a symmetric Nash
equilibrium and, in a nice example
where theory aligns with practice, that
real bidding behavior is largely consistent with the model.
A number of questions drive research in ad auctions and exchanges.
What mechanisms increase advertiser
value or publisher revenue? What user
and content attributes contribute to
variation in advertiser value? How can
bids for different contingencies (
impressions, clicks, or conversions) be
integrated and optimized over time?
What constraints on supply and budget
make sense? How should advertisers
and publishers bid? How can publishers and advertisers incorporate learning and optimization (while trying to
balance exploration and exploitation)?
How do practical constraints such as
real-time delivery affect design? How
The eliciting and aggregation of information from diverse and frequently
self-interested sources is in general called “knowledge integration,”
with a particular case being “price
discovery”—a side effect of market-based resource allocation. The balance
point of supply and demand reveals the
negotiated value of the resource.
In some cases, the value revealed
in prices can rival or eclipse the value
of trade. For example, the price of an
asset that pays $1 if a category- 5 hurricane hits Florida in 2009 can be seen
as a probabilistic forecast of this catastrophic event. The value of an actual
and more accurate forecast could run
into the millions of dollars.
A “prediction market” is a market
designed primarily for price discovery
rather than resource allocation, and this
alternate focus leads to a different pri-oritization of design goals. For example,
the market operator may be happy to
pay for the information it seeks, instead
of enforcing neutral or positive revenue.
Trading is not the end goal but a means
to the end of acquiring complete, accurate, and timely information. For
example, the Iowa Electronic Market
forecasts the outcomes of political elections, and intrade.com predicts events
ranging from the outbreak of avian flu
to Osama bin Laden’s capture.
Liquidity and expressiveness play
important roles in prediction markets.
If a trader with information cannot reveal it to the market, either because illi-quidity prevents matching with another trader or because the market does
not support the way in which the trader
wants to express information, then the
mechanism may fail.
Designing prediction markets to improve liquidity and expressiveness poses substantial though not insuperable
Liquidity can be addressed through the use
of automated market makers that are
always willing to buy and sell at some
prices and that adjust prices dynamically to ensure a bound on their worst-case loss.
18, 26 Expressiveness is gained
at an often-severe computational cost,
thus placing a difficult computational
problem in the lap of the auctioneer
striving to bring matched traders together or of the market maker trying
to (implicitly) maintain an exponential
number of prices.
10 In some instances,
there is a reasonable and useful compromise between expressiveness and
A more direct means of obtaining
information is to pay an expert, though