We present a vertical introduction to campaign optimization; that is, the ability to predict the user response to an ad campaign without any users' profiles on average and for each exposed ad. In practice, we present an approach to build a polytomous model, multi response, composed by several hundred binary models using generalized linear models. The theory has been introduced twenty years ago and it has been applied in different fields since then. Here, we show how we optimize hundreds campaigns and how this large number of campaigns may overcome a few characteristic caveats of single campaign optimization. We discuss the problem and solution of training and calibration at scale. We present statistical performance as {\em coverage}, {\em precision} and {\em recall} used in classification. We present also a discussion about the potential performance as throughput: how many decisions can be done per second streaming the bid auctions also by using dedicated hardware.
[1]
C. Borror.
Generalized Linear Models and Extensions, Second Edition
,
2008
.
[2]
A. W. Kemp,et al.
Univariate Discrete Distributions
,
1993
.
[3]
J. Andel.
Sequential Analysis
,
2022,
The SAGE Encyclopedia of Research Design.
[4]
J. Bernoulli,et al.
Jacobi Bernoulli ... Ars conjectandi : opus posthumum : accedit Tractatus de seriebus infinitis, et Epistola Gallicè scripta de ludo pilae reticularis
,
1968
.
[5]
Ali Jalali,et al.
Real time bid optimization with smooth budget delivery in online advertising
,
2013,
ADKDD '13.
[6]
Wentong Li,et al.
Estimating conversion rate in display advertising from past erformance data
,
2012,
KDD.