Optimal budget allocation across search advertising markets

Yanwu Yang, Jie Zhang, Baiyu Liu, Daniel Zeng

Research output: Contribution to conferencePaperpeer-review

6 Scopus citations


How to rationally allocate the limited advertising budget is a critical issue in search auctions. However, due to the heterogeneousness of major search markets in terms of auction mechanisms, ranking algorithms and advertising structures, it is becoming increasingly difficult for an advertiser to manipulate advertising budget simultaneously across several search markets. In this paper, we establish a novel optimal budget allocation model across search advertising markets, under a finite time horizon. By considering distinctive features of search auctions, we introduce the quality score q and the dynamic advertising effort u to extend the advertising response function to fit budget decision scenarios. We also provide a feasible solution to our model and study some desirable properties: (a) the marginal return is non-increasing with respect to the advertising budget; (b) the optimal budget solution satisfies the condition that the advertising effort u is positively proportional to the product of the change of accumulated revenue in a market ∂V, the change of market share ∂ θ and the advertising elasticity α. Computational experiments are made to evaluate our model and identified properties. Experimental results show that the advertiser with increasing advertising elasticity is suggested to invest more budget in the late stages, but the advertiser with decreasing advertising elasticity should invest more budget in the initial stage, in order to maximize net profits.

Original languageEnglish (US)
Number of pages6
StatePublished - 2011
Event21st Workshop on Information Technologies and Systems, WITS 2011 - Shanghai, China
Duration: Dec 3 2011Dec 4 2011


Other21st Workshop on Information Technologies and Systems, WITS 2011

ASJC Scopus subject areas

  • Information Systems


Dive into the research topics of 'Optimal budget allocation across search advertising markets'. Together they form a unique fingerprint.

Cite this