Systems and methods for targeted advertising to specific consumers are disclosed. A system may include a memory storing instructions; and at least one processor configured to execute the instructions to: receive, over a network, consumer data from a client device; identify a plurality of client-provided consumers from the consumer data; obtain a plurality of unique consumer identifiers corresponding to the plurality of client-provided consumers; and identify at least one first overlapping unique consumer identifier by matching at least one of the plurality of client-provided consumers with at least one publisher-provided consumer provided by a first publisher device of a plurality of publisher devices, the first publisher device having a highest priority among the plurality of publisher devices.

Described are techniques for A/B testing including a computer-implemented method of identifying, in an A/B testing database, a set of feature values with a statistically significant difference in A/B testing outcomes above a threshold. The method further includes partitioning records in the A/B testing database into a plurality of population strata according to the set of feature values. The method further includes performing A/B testing, and identifying heterogeneous outcomes of the A/B testing for respective strata of the plurality of population strata.

Systems and methods are described herein for resolving advertisement placement conflicts. Specifically, a number of parameters may be entered into a system in order to distribute advertisements into advertisement slots. In many instances, a combination of these parameters causes a conflict in the system where all the parameters cannot be applied in order to place advertisements into advertisement slots. The conflict may be resolved by using an advertisement assignment model to determine which parameters may be relaxed in order to arrive at an optimal solution that violates a smallest number of parameters having the least priority. When such a solution is found, the advertisement assignment model may be modified and advertisements may be placed into advertisement slots based on the modified advertisement assignment model.

An influence system for predicting advertisement impact for advertising response selection. A treatment group of entities is selected which have received an advertising treatment of a campaign. A control group of entities is selected which excludes the treatment group entities is selected which have not received the advertising treatment of the campaign. An influence model is created for each campaign by comparing features of the treatment group converters to features of the control group converters. A campaign is selected for an opportunity to expose a specified entity to advertising based on the result of applying each respective campaign's influence model to features of the specified entity.

The present disclosure provides for management of impressions in advertisement campaigns. Impressions may be moved between different impression media based on performance metrics and historical data. Impression budgets may be modified in an active campaign based on forecasts determined using current performance data. Impression budgets may be reallocated between simultaneously active advertisement campaigns.

A data processing method includes: obtaining an advertisement state of each candidate advertisement corresponding to a current exposure request and an overall state of an advertising platform in response to the current exposure request; determining, by a classification network in a scoring model, probability of each candidate advertisement belonging to different reference advertisement types; determining, by a scoring network in the scoring model, a competition score of the candidate advertisement for the current exposure request according to the advertisement state corresponding to the candidate advertisement and the overall state based on the probability of the candidate advertisement belonging to different reference advertisement types, the scoring model including multiple scoring networks corresponding to different reference advertisement types; and determining a target advertisement exposed by the current exposure request according to the competition score of each candidate advertisement for the current exposure request.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating models. In some implementations, a system obtains data that comprises promotions and parameters for an opportunity. The system generates transformation spaces based on the promotions and the parameters, wherein each transformation space comprises states, each state is based on the parameters for a particular promotion. The system iterates over a number of iterations. For each transformation space, the system adjusts a state of the transformation space based on actions. The system generates a model by combining each adjusted state. The system generates an entropy for the model. The system compares the entropy to a threshold value, wherein the threshold value corresponds to one of the parameters. In response to determining that the entropy exceeds the threshold value, the system iterates. The system provides the generated model for output.

This paper presents a practical method for measuring the impact of multiple marketing events on sales, including marketing events that are not traditionally trackable. The technique infers which of several competing media events are likely to have caused a given conversion. The method is tested using hold-out sets, and also a live media experiment for determining whether the method can accurately predict television-generated web conversions.

In general, methods and system for modeling the incremental value of a response to different types of treatment are disclosed. Some examples include modeling discount sensitivity for specific guests. One aspect is a method for modeling incremental sales for a retail enterprise which includes generating a proxy campaign. In some embodiments, the proxy campaign is used to train an uplift model to predict an uplift score for each guest in response to a proposed campaign. In some embodiments, the uplift score is used to select guests for the proposed campaign.

A digital ad-buying platform uses counterfactual-based incrementality measurement by implementing randomization and/or a correction for auction win bias to avoid the need to identify counterfactual winner types in the control group. This approach can estimate impact at the individual consumer level. Confidence levels can be determined using Gibbs sampling in the context of causal analysis in the presence of non-compliance.