Predictive recommendation system using price boosting
In general, embodiments of the present invention provide systems, methods and computer readable media for ranking promotions selected for recommendation to consumers based on predictions of promotion performance and consumer behavior. In embodiments, a set of promotions to be recommended to a consumer can be sorted and/or ranked according to respective relevance scores representing a probability that the consumer's behavior in response to the promotion will match a ranking target. In embodiments, calculating scores is based on a relevance model (a predictive function) derived from one or more contextual data sources representing attributes of promotions and consumer behavior. In embodiments, an absolute relevance score represents an absolute prediction of a ranking target variable. In embodiments, absolute relevance may be used to determine personalized local merchant discovery frontiers; featured result set thresholding for impressions; and/or promotion notification triggers. In embodiments, predictive models based on gross revenue may be optimized using promotion category-dependent price boosting.
1. A computer-implemented method, comprising:
receiving a set of available promotions for a particular consumer associated with a computing device;
generating a relevance score for each available promotion of the set of available promotions, the relevance score representing an estimated gross revenue per impression for an available promotion, wherein the estimated gross revenue is associated with a promotion price for the available promotion;
ranking the set of available promotions based on respective relevance scores;
dynamically adjusting each relevance score to generate a respective adjusted relevance score for each available promotion of the set of available promotions, wherein the dynamically adjusting comprises:
receiving features representing the available promotion associated with the relevance score, wherein the features include the promotion price and a promotion category for the available promotion; and
in response to a determination that the promotion price satisfies a price threshold associated with the promotion category of the available promotion, assigning the price threshold as the promotion price for the available promotion;
re-ranking the set of available promotions based on the respective adjusted relevance scores to generate a re-ranked set of available promotions; and
presenting a subset comprising a top-N ranked promotions of the re-ranked set of available promotions to the computing device associated with the particular consumer.
2. The computer-implemented method of claim 1 , wherein the relevance score associated with the available promotion is generated by multiplying the promotion price by a conversion rate per impression for the available promotion.
3. The computer-implemented method of claim 1 , wherein
the respective adjusted relevance score represents an estimated minimum gross revenue per impression for the available promotion.
4. The computer-implemented method of claim 1 , further comprising: in circumstances where the promotion price is below the price threshold associated with the promotion category of the available promotion,
generating an exponential conversion rate multiplier based on the promotion price; and
adjusting the relevance score by multiplying the relevance score by the exponential conversion rate multiplier.
5. The computer-implemented method of claim 1 , wherein the price threshold associated with the promotion category of the available promotion is determined based on a combination of business strategies.
6. The computer-implemented method of claim 5 , wherein the combination of business strategies is associated with configurable parameters including at least one of a sweet spot price range for the promotion category, a lowest price point for the promotion category, a highest price point for the promotion category, a price exponent for the promotion category, or a threshold price point below which the exponential conversion rate multiplier is applied to a conversion rate per impression for the promotion category associated with the available promotion.
7. The computer-implemented method of claim 5 , wherein the combination of business strategies is associated with promotion data representing attributes of promotion performance, or user data representing attributes of consumer behavior.
8. The computer-implemented method of claim 1 , wherein dynamically adjusting each relevance score further comprises:
determining an active time session for the particular consumer; and
dynamically adjusting each relevance score in circumstances where the particular consumer is in a highly active time session.
9. A computer program product, encoded on a computer-readable medium, operable to cause a data processing apparatus to perform operations comprising:
receiving a set of available promotions for a particular consumer associated with a computing device;
generating a relevance score for each available promotion of the set of available promotions, the relevance score representing an estimated gross revenue per impression for an available promotion, wherein the estimated gross revenue is associated with a promotion price for the available promotion;
ranking the set of available promotions based on respective relevance scores;
dynamically adjusting each relevance score to generate a respective adjusted relevance score for each available promotion of the set of available promotions, wherein the dynamically adjusting comprises:
receiving features representing the available promotion associated with the relevance score, wherein the features include the promotion price and a promotion category for the available promotion; and
in response to a determination that the promotion price satisfies a price threshold associated with the promotion category of the available promotion, assigning the price threshold as the promotion price for the available promotion;
re-ranking the set of available promotions based on the respective adjusted relevance scores to generate a re-ranked set of available promotions; and
presenting a subset comprising a top-N ranked promotions of the re-ranked set of available promotions to the computing device associated with the particular consumer.
10. The computer program product of claim 9 , wherein the relevance score associated with the available promotion is generated by multiplying the promotion price by a conversion rate per impression for the available promotion.
11. The computer program product of claim 9 , wherein
the respective adjusted relevance score represents an estimated minimum gross revenue per impression for the available promotion.
12. The computer program product of claim 9 , wherein the operations further comprise: in circumstances where the promotion price is below the price threshold associated with the promotion category of the available promotion,
generating an exponential conversion rate multiplier based on the promotion price; and
adjusting the relevance score by multiplying the relevance score by the exponential conversion rate multiplier.
13. The computer program product of claim 9 , wherein the price threshold associated with the promotion category of the available promotion is determined based on a combination of business strategies.
14. The computer program product of claim 13 , wherein the combination of business strategies is associated with configurable parameters including at least one of a sweet spot price range for the promotion category, a lowest price point for the promotion category, a highest price point for the promotion category, a price exponent for the promotion category, or a threshold price point below which the exponential conversion rate multiplier is applied to a conversion rate per impression for the promotion category associated with the available promotion.
15. The computer program product of claim 13 , wherein the combination of business strategies is associated with promotion data representing attributes of promotion performance, or user data representing attributes of consumer behavior.
16. The computer program product of claim 9 , wherein dynamically adjusting each relevance score further comprises:
determining an active time session for the particular consumer; and
dynamically adjusting each relevance score in circumstances where the particular consumer is in a highly active time session.
17. A system comprising one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations, comprising:
receiving a set of available promotions for a particular consumer associated with a computing device;
generating a relevance score for each available promotion of the set of available promotions, the relevance score representing an estimated gross revenue per impression for an available promotion, wherein the estimated gross revenue is associated with a promotion price for the available promotion;
ranking the set of available promotions based on respective relevance scores;
dynamically adjusting each relevance score to generate a respective adjusted relevance score for each available promotion of the set of available promotions, wherein the dynamically adjusting:
receiving features representing the available promotion associated with the relevance score, wherein the features include the promotion price and a promotion category for the available promotion; and
in response to a determination that the promotion price satisfies a price threshold associated with the promotion category of the available promotion, assigning the price threshold as the promotion price for the available promotion;
re-ranking the set of available promotions based on the respective adjusted relevance scores to generate a re-ranked set of available promotions; and
presenting a subset comprising a top-N ranked promotions of the re-ranked set of available promotions to the computing device associated with the particular consumer.
18. The system of claim 17 , wherein the relevance score associated with the available promotion is generated by multiplying the promotion price by a conversion rate per impression for the available promotion.
19. The system of claim 17 , wherein
the respective adjusted relevance score represents an estimated minimum gross revenue per impression for the available promotion.
20. The system of claim 17 , wherein the operations further comprise:
in circumstances where the promotion price is below the price threshold associated with the promotion category of the available promotion,
generating an exponential conversion rate multiplier based on the promotion price; and adjusting the relevance score by multiplying the relevance score by the exponential conversion rate multiplier.
21. The system of claim 17 , wherein the price threshold associated with the promotion category of the available promotion is determined based on a combination of business strategies.
22. The system of claim 21 , wherein the combination of business strategies is associated with configurable parameters including at least one of a sweet spot price range for the promotion category, a lowest price point for the promotion category, a highest price point for the promotion category, a price exponent for the promotion category, or a threshold price point below which the exponential conversion rate multiplier is applied to a conversion rate per impression for the promotion category associated with the available promotion.
23. The system of claim 21 , wherein the combination of business strategies is associated with promotion data representing attributes of promotion performance, or user data representing attributes of consumer behavior.
24. The system of claim 17 , wherein dynamically adjusting each relevance score further comprises:
determining an active time session for the particular consumer; and
dynamically adjusting each relevance score in circumstances where the particular consumer is in a highly active time session.