Methods, systems, computer-readable media, and apparatuses for providing responsive advertisements are described herein. In some embodiments, a computing device may send a presentation including multiple advertisements to a user device. The computing device may receive an indication that an advertisement has been overlaid by a window and may, in response, determine to resize or relocation the advertisement. The computing device may determine impression credit for the advertisement based on a display duration for the advertisement prior to being overlaid by the window and a display duration for the resized or relocated advertisement while not being overlaid by any window.

Apparatuses, methods, and systems for determining winning arms of electronic testing. One method includes obtaining historical data values related to the A/B test of a user, storing the historical data values, determining a historical weight for the historical data values, receiving new data values from the plurality of computing devices collected based on recipient actions during execution of the A/B, constructing a Dirichlet distribution, inferring corresponding central tendencies of samplings of a metric distribution, wherein each central tendency of the corresponding central tendencies is determined by sampling the Dirichlet distribution, constructing an overall utility distribution for each arms of the A/B test by combining the central tendency of each sampling of the metric distribution with a corresponding sampling of a conversion probability distribution, determining a winning arm of the A/B testing by comparing the overall utility distribution of each arm with each other arm of the A/B test.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for managing application program interface calls.

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.

A method includes: receiving a first product value; determining a cannibalization value based on at least one of experimental data and non-experimental data; generating a second product value by adjusting the first product value using the cannibalization value; determining a return value based on return data; generating a third product value by adjusting the second product value using the return value; determining a lifetime customer value based on at least the cannibalization value; generating a fourth product value by adjusting the third product value using the lifetime customer value; determining an organic rank value based on organic placement data and value of organic placement data; generating a fifth product value by adjusting the fourth product value using the organic rank value; generating, based at least on the fifth product value, an actionable report descriptive of an efficacy rate of a target product in a computer-networked marketplace.