System and method for automatic generation and delivery of personalized content

An application programming interface (API) server for real-time delivery of personalized content, the API server comprising:

• one or more processors;
• a non-transitory computer-readable medium;
• and
• instructions stored on the computer-readable medium that implement an application programming interface (API), wherein the instructions, when implemented by the one or more processors, cause the API server to:
• detect that a delivery server invoked the API in response to a delivery event corresponding to the delivery server detecting that a client device is accessing an email provided to the client device by the delivery server, the email formatted in accordance with a template;
• receive, via the API, (i) a request for a set of offers including a user identifier associated with a user of the client device, (ii) offer query criteria based on the template, and (iii) an indication of a preferred model from which offers of the set of offers should be generated, wherein the offer query criteria includes (a) an indication of one or more zones of the email to be populated using the set of offers and (b) a quantity of offers to be populated in each zone of the one or more zones;
• build the set of offers by:
• retrieving, via the API from a database server using an API request, (i) candidate offers specific to the user that have been generated, prior to receiving the request for the set of offers, by a machine learning model trained on historical data indicative of a plurality of interactions of a plurality of users with an enterprise, the candidate offers including an incentive for the user to interact with the enterprise, and (ii) an identification of a model used to generate each of the candidate offers, wherein retrieving the candidate offers includes retrieving, for each candidate offer of the candidate offers, one or more candidate zones to which the candidate offer corresponds;
• retrieving, for each of the candidate offers, a rank based on calculating at least one affinity score indicating an estimated affinity of the user for the corresponding candidate offer;
• and
• based on the offer query criteria,
• identifying the set of offers, from the candidate offers, to deliver to the delivery server, wherein identifying the set of offers includes filtering the candidate offers based on (i) the quantity of offers and (ii) whether the one or more candidate zones of each of the candidate offers are included in the one or more zones, and
• filtering the candidate offers based on whether the model, for each of the candidate offers, corresponds to the preferred model;
• and
• transmit, via the API, the set of offers to the delivery server to cause the delivery server to populate the email with content corresponding to the set of offers, in accordance with the template, wherein the client device displays the email including the content when the client device accesses the email.

A non-transitory computer-readable medium storing thereon instructions that implement an application programming interface (API) for real-time delivery of personalized content, wherein the API is configured to at least:

• detect that a delivery server invoked the API in response to a delivery event corresponding to the delivery server detecting that a client device is accessing an email provided to the client device by the delivery server, the email formatted in accordance with a template;
• receive, from the delivery server, (i) a request for a set of offers including a user identifier associated with a user of the client device, (ii) offer query criteria based on the template, and (iii) an indication of a preferred model from which offers of the set of offers should be generated, wherein the offer query criteria includes (a) an indication of one or more zones of the email to be populated using the set of offers and (b) a quantity of offers to be populated in each zone of the one or more zones;
• build the set of offers by:
• retrieving, from a database server using an API request, (i) candidate offers specific to the user that have been generated, prior to receiving the request for the set of offers, by a machine learning model trained on historical data indicative of a plurality of interactions of a plurality of users with an enterprise, the candidate offers including an incentive for the user to interact with the enterprise, and (ii) an identification of a model used to generate each of the candidate offers, wherein retrieving the candidate offers includes retrieving, for each candidate offer of the candidate offers, one or more candidate zones to which the candidate offer corresponds;
• retrieving, for each of the candidate offers, a rank based on calculating at least one affinity score indicating an estimated affinity of the user for the corresponding candidate offer;
• and
• based on the offer query criteria,
• identifying the set of offers, from the candidate offers, to deliver to the delivery server, wherein identifying the set of offers includes filtering the candidate offers based on (i) the quantity of offers and (ii) whether the one or more candidate zones of each of the candidate offers are included in the one or more zones, and
• filtering the candidate offers based on whether the model, for each of the candidate offers, corresponds to the preferred model;
• and
• transmit the set of offers to the delivery server to cause the delivery server to populate the email with content corresponding to the set of offers, in accordance with the template, wherein the client device displays the email including the content when the client device accesses the email.