DIGITAL INCREMENTAL CONVERSION
A method may include obtaining observational data associated with a digital advertising campaign for multiple households. The method may also include estimating a likelihood that a household of the multiple households may be reached by the digital advertising campaign and a counterfactual conversion outcome for a reached household. The estimating may be performed as a function of the one or more covariates. The method may further include generating an estimate of incremental conversions attributable to the digital advertising campaign by combining differences between the observed conversion outcome of the reached households and the counterfactual conversion outcome for the reached households and a propensity-based correction derived from outcomes of the plurality of households based on the estimated likelihood that a household is reached. The method may also include outputting the estimate of incremental conversions.
1 . A method, comprising:
obtaining observational data associated with a digital advertising campaign for a plurality of households, the observational data comprising a reach indicator, one or more covariates, and an observed conversion outcome;
estimating a likelihood that a household of the plurality of households is reached by the digital advertising campaign and a counterfactual conversion outcome for a reached household, wherein:
the estimating is performed as a function of the one or more covariates; and
the counterfactual conversion outcome is estimated using data from unreached households of the plurality of households;
generating an estimate of incremental conversions attributable to the digital advertising campaign by combining:
differences between the observed conversion outcome of the reached households and the counterfactual conversion outcome for the reached households; and
a propensity-based correction derived from outcomes of the plurality of households based on the estimated likelihood that a household is reached; and
outputting the estimate of incremental conversions.
2 . The method of claim 1 , wherein the counterfactual conversion outcome represents an expected conversion outcome for the reached household if the reached household had not been reached by the digital advertising campaign.
3 . The method of claim 1 , wherein the reach indicator indicates whether a household was reached at least once by the digital advertising campaign and wherein the one or more covariates comprise household-level digital activity data measured independent of a first campaign exposure.
4 . The method of claim 1 , wherein the observational data is obtained from one or more digital platforms.
5 . The method of claim 1 , wherein the observed conversion outcome comprises at least one of a conversion count during a measurement window, purchased units during the measurement window, revenue during the measurement window, an online conversion, a location visit, and an in-store purchase.
6 . The method of claim 1 , wherein estimating the likelihood that the household is reached comprises training a propensity model using the observational data to predict reach status from the one or more covariates.
7 . The method of claim 1 , wherein estimating the counterfactual conversion outcome comprises training an outcome model using data from the unreached households to predict expected conversion outcomes as a function of the one or more covariates.
8 . The method of claim 7 , further comprising applying the outcome model to reached households to produce the counterfactual conversion outcome, for each of the reached households, representing the expected conversion outcome if not reached.
9 . The method of claim 1 , wherein the propensity-based correction is derived from differences between the observed conversion outcome of the unreached households and predicted conversion outcomes for the unreached households.
10 . The method of claim 9 , wherein deriving the propensity-based correction comprises weighting contributions of the unreached households based on the estimated likelihood of being reached so that the unreached households are reweighted to reflect a covariate distribution of the reached households.
11 . The method of claim 1 , wherein generating the estimate of incremental conversions comprises computing household-level incremental conversion contributions for the plurality of households and aggregating the household-level incremental conversion contributions.
12 . The method of claim 11 , further comprising estimating incremental conversions for a subset of households by aggregating the household-level incremental conversion contributions over the subset of households.
13 . The method of claim 11 , further comprising computing an uncertainty measure for the estimate of incremental conversions based on variability of the household-level incremental conversion contributions.
14 . The method of claim 1 , further comprising incorporating predetermined weights associated with the plurality of households, wherein generating the estimate of incremental conversions comprises weighting household-level incremental conversion contributions based on the predetermined weights.
15 . A system, comprising:
one or more non-transitory computer-readable storage media configured to store instructions; and
one or more processors communicatively coupled to the one or more non-transitory computer-readable storage media and configured to, in response to execution of the instructions, cause the system to perform operations, the operations comprising:
obtain observational data associated with a digital advertising campaign for a plurality of households, the observational data comprising a reach indicator, one or more covariates, and an observed conversion outcome;
estimate a likelihood that a household of the plurality of households is reached by the digital advertising campaign and a counterfactual conversion outcome for a reached household, wherein:
the estimate is performed as a function of the one or more covariates; and
the counterfactual conversion outcome is estimated using data from unreached households of the plurality of households;
generate an estimate of incremental conversions attributable to the digital advertising campaign by combining:
differences between the observed conversion outcome of the reached households and the counterfactual conversion outcome for the reached households; and
a propensity-based correction derived from outcomes of the plurality of households based on the estimated likelihood that a household is reached; and
output the estimate of incremental conversions.
16 . The system of claim 15 , wherein the counterfactual conversion outcome represents an expected conversion outcome for the reached household if the reached household had not been reached by the digital advertising campaign.
17 . The system of claim 15 , wherein the reach indicator indicates whether a household was reached at least once by the digital advertising campaign and wherein the one or more covariates comprise household-level digital activity data measured independent of a first campaign exposure.
18 . The system of claim 15 , wherein the observed conversion outcome comprises at least one of a conversion count during a measurement window, purchased units during the measurement window, revenue during the measurement window, an online conversion, a location visit, and an in-store purchase.
19 . The system of claim 15 , wherein estimating the likelihood that the household is reached comprises training a propensity model using the observational data to predict reach status from the one or more covariates.
20 . The system of claim 15 , wherein estimating the counterfactual conversion outcome comprises training an outcome model using data from the unreached households to predict expected conversion outcomes as a function of the one or more covariates.