RECOMMENDATION OF ELECTRONIC OFFERS BASED ON UNIVERSAL SCORING FUNCTIONS
According to an embodiment, a data processing system for managing electronic offers comprises: a first logic module adapted to receive a set of recommendation reference data comprising a plurality of items, the plurality of items including at least one product or service, and a plurality of offers; a second logic module adapted to determine a set of association scores for the plurality of items based on the recommendation reference data; a third logic module adapted to determine a universal score for each offer; and a fourth logic module adapted to rank the offers in a recommendation order based on the set of association scores and on the universal scores. In an embodiment, each universal score is based at least in part on a measure of revenue associated with one or more of: activation, recommendation, or redemption of the offer.
1 . A data processing system for managing electronic offers, the system comprising:
a first logic module adapted to receive a set of recommendation reference data comprising a plurality of items, the plurality of items including at least one product or service, and a plurality of offers;
a second logic module adapted to determine a set of association scores for the plurality of items based on the recommendation reference data;
a third logic module adapted to determine a universal score for each offer; and
a fourth logic module adapted to rank the offers in a recommendation order based on the set of association scores and on the universal scores.
2 . The data processing system of claim 1 , wherein each universal score is based at least in part on a measure of revenue associated with one or more of: activation, recommendation, or redemption of the offer.
3 . The data processing system of claim 2 , wherein the measure of revenue for at least one universal score is designed to be more favorable to one or more of: an offer distributor, an offer provider, or a consumer.
4 . The data processing system of claim 1 , wherein at least one universal score is based on a comparison of the historical performance of an offer to average offer performance across a plurality of offers.
5 . The data processing system of claim 1 , wherein at least one universal score is designed to maintain a predetermined distribution rate for an offer during a given period of time.
6 . The data processing system of claim 1 , wherein at least two universal scores are determined for different entities, and wherein the entities include one or more of: a set of consumers, a set of retailers, or a set of offer providers.
7 . The data processing system of claim 1 , wherein at least one universal score is further adjusted to artificially increase the recommendation ranking of a particular offer.
8 . The data processing system of claim 1 , wherein a recommended offer is distributed in response to a request for a receipt.
9 . The data processing system of claim 1 , wherein a recommended offer is included in a transaction receipt transmitted to a customer data processing system.
10 . The data processing system of claim 1 , wherein the recommendation reference data includes at least one of: a transaction log, a shopping list, a web clickstream for particular consumer entities, an offer clickstream for particular consumer entities, an offer impression log, an offer activation log, an offer redemption log, or a consumer behavioral log.
11 . The data processing system of claim 1 , wherein at least one association score is weighted.
12 . The data processing system of claim 10 , wherein at least one weight is adapted to increase offer-related revenue for an offer distributor, to increase an offer discount for a customer, or to increase a profit margin of a retailer or of an offer provider.
13 . The data processing system of claim 10 , wherein at least one weight is modified based on feedback from at least one of: an offer redemption, an offer activation, or a purchase.
14 . The data processing system of claim 10 , wherein at least one weight is based on at least one of: contextual transaction data or a historical transaction record.
15 . The data processing system of claim 1 , wherein at least one association score is: an item-to-item correlation score, offer-to-item correlation score, consumer-centric correlation score, item-to-event correlation score, event-to-event correlation score, consumer-to-event correlation score, or measure of the frequency of co-occurrence of a set of items.
16 . The data processing system of claim 1 , wherein at least one universal score is based on one or more of: a historical or projected impact of an offer on sales of an item, a historical or projected offer redemption rate, a historical or projected profit per offer impression, a historical or projected profit margin, a historical or projected offer volume, a historical or projected offer yield, a historical trend in offers made available by an offer provider, a historical or projected profit margin for the offer distributor, a historical or projected profit margin for an offer provider, a historical or projected impact of an offer on other offers, or a historical or projected impact of an offer on consumer behavior.
17 . A method comprising:
identifying a set of at least two offers responsive to a request;
calculating universal scores for each offer in the set of offers;
for each offer in the set of offers, calculating a function of: a first ratio of total activations for the offer to average total activations for all offers in a plurality of offers, a second ratio of total impressions for the offer to average total impressions for all offers in the plurality of offers, and a third ratio of total redemptions for the offer to average total redemptions for all offers in the plurality of offers;
ranking the set of offers based at least in part on the universal scores;
providing information about at least a subset of the ranked set of offers in response to the request;
wherein the method is performed by one or more computing devices.
18 . The method of claim 17 , wherein calculating a universal score comprises comparing historical performance of the offer to average offer performance for multiple offers in a group of offers
19 . The method of claim 17 , wherein calculating the universal score for comprises calculating a function of at least activations, redemptions, and impressions for the offer.
20 . The method of claim 17 , wherein calculating the universal score comprises calculating a function of at least offer inventory and offer end date
21 . The method of claim 17 , wherein calculating the universal scores comprises calculating predetermined universal scores based on a set of signals in advance of receiving the request.
22 . The method of claim 17 , wherein calculating the universal scores comprises calculating delta changes to the predetermined universal scores based on changes in the signals between the time that the predetermined universal scores were originally calculated and the time that the request was received.
23 . The method of claim 17 , wherein calculating the universal score for a particular item is based on a manually specified sponsorship score for the particular item.
24 . The method of claim 17 , wherein the first ratio, second ratio, and third ratio are weighted.
25 . The method of claim 17 , further comprising learning different weights for different contexts.
26 . The method of claim 17 , wherein the first ratio, second ratio, and third ratio are weighted based on data that is specific to certain groups of users
27 . The method of claim 17 , wherein the plurality of offers is the identified set responsive to the request.
28 . The method of claim 17 , wherein the plurality of offers is a set of all available offers, including offers not in the set responsive to the request.
29 . The method of claim 17 , wherein a universal score is determined for a new offer, or an offer for which there is not yet a threshold amount of data, as a ratio of average totals from a group of offers to average totals of the plurality of offers, wherein the group of offers is a subset of the plurality of offers, of which the new offer is part.
30 . A method for managing electronic offers, the method comprising:
receiving a set of recommendation reference data comprising a plurality of items, the plurality of items including at least one product or service, and a plurality of offers;
determining a set of association scores for the plurality of items based on the recommendation reference data;
determining a universal score for each offer; and
ranking the offers in a recommendation order based on the set of association scores and on the universal scores;
wherein the method is performed by one or more computing devices.