Customer-level cross-channel message planner

• US 9,959,551 B1
• Filed: 12/18/2014
• Issued: 05/01/2018
• Est. Priority Date: 12/18/2014
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Schermerhorn, Charles Porter, Edwards, Jr., David John, Kondal, Tegdeep, Parsons, Donald Arthur, Shah, Nisarg, Shanbhag, Vishal Mangesh
• Primary Examiner(s): Holmes, Michael B
• Application Number:
  US14/575,999
• Time in Patent Office: 1,230 Days
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 30/0244 Optimization
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: H04L 12/1859 adapted to provide push services, e.g. data channels
• CPC Class Codes: H04L 51/214 using selective forwarding
• CPC Class Codes: H04L 67/62 Establishing a time schedule for servicing the requests
• CPC Class Codes: H04L 67/63 Routing a service request depending on the request content or context

Abstract

Techniques for providing a messaging plan may be provided. A message that conforms to a channel agnostic format may be received. In an example, a messaging plan that includes a channel for delivering the message may be determined based at least in part on a score determined by a machine learning module utilizing metrics of a user. The message may be provided to the user according to the messaging plan. The messaging plan may be updated for subsequent messages based at least in part on updated metrics of the user received in response to the user interacting with the message.

Claims

1. A computer-implemented method comprising:

receiving, by a computer system, a message intended for a user associated with an electronic marketplace, the message conforming to a channel agnostic format;

analyzing the message for relevancy to the user based at least in part on information associated with the user via the electronic marketplace;

calculating a score based at least in part on machine learning utilizing input indicating an interaction by the user with previously provided messages, the score corresponding to a probability that the user will interact with the message when provided via a particular message channel of a plurality of message channels, a unique message channel of the plurality of message channels being unique with respect to other unique message channels;

determining, by the computer system, a cross-channel messaging plan for the message and the user based at least in part on the score and the relevancy of the message to the user, the cross-channel messaging plan identifying one or more message channels of the plurality of message channels and a delivery time period of a plurality of delivery time periods to associate with the unique message channel of the one or more message channels;

selecting the particular messaging channel from the plurality of messaging channels based at least in part on the score;

providing the message to the user via a delivery system associated with the particular messaging channel;

in response to providing the message to the user,

receiving a metric associated with the user; and

updating the cross-channel messaging plan for a subsequent message to the user based at least in part on the metric.

2. The computer-implemented method of claim 1, wherein the machine learning utilizes the information associated with the user via the electronic marketplace and implicit user preferences obtained from the user interacting with provided messages to generate the score.

3. The computer-implemented method of claim 1, wherein determining the cross-channel messaging plan for the message and the user is further based at least in part on implicit user preferences and explicit user preferences, the implicit user preferences obtained by analyzing the metric associated with the user.

4. The computer-implemented method of claim 3, wherein the particular messaging channel comprises at least one of an email channel, a short message service (SMS) channel, a telephone channel, a mobile telephone channel, or a stand-alone application channel.

5. The computer-implemented method of claim 1, wherein the channel agnostic format of the message includes information indicating at least one of a type of message, an identifier for the user, a start time period, an expire time period, metadata associated with a category of an item offered by the electronic marketplace, metadata associated with a value of the item offered by the electronic marketplace, or electronic marketplace attributes.

6. The computer-implemented method of claim 1, wherein the cross-channel message plan includes metadata for each messaging channel of the plurality of messaging channels, the metadata identifying at least one of the score, an optimal delivery time period, and an expiration time period.

7. A non-transitory computer-readable storage medium storing computer-executable instructions that, when executed by a computer system, configure the computer system to perform operations comprising:

receiving a notification that conforms to a channel agnostic format;

determining a messaging plan that identifies a channel of a plurality of channels and a delivery time period of a plurality of delivery time periods associated with the channel for delivering the notification to a user, said determining based at least in part on a score measuring engagement by the user with the notification provided via the channel, the channel being unique with respect to other channels of the plurality of channels;

providing the notification via the channel identified in the messaging plan; and

updating the messaging plan for at least one subsequent notification based at least in part on a metric representing the user that is determined in response to the user engaging with the notification.

8. The non-transitory computer-readable storage medium of claim 7, wherein providing the notification to the user via the channel identified in the messaging plan includes providing the notification to a channel notification queue.

9. The non-transitory computer-readable storage medium of claim 8, wherein a delivery system accesses the notification from the channel notification queue using a first application programming interface (API) that accesses all provided notifications in the channel notification queue during a specific time period or using a second API that accesses a notification immediately upon being provided to the channel notification queue.

10. The non-transitory computer-readable storage medium of claim 9, wherein the delivery system translates the notification from the channel agnostic format to a channel specific format.

11. The non-transitory computer-readable storage medium of claim 7, wherein determining the messaging plan is further based at least in part on the score satisfying a threshold.

12. The non-transitory computer-readable storage medium of claim 11, wherein the threshold is unique to the notification and the user.

13. The non-transitory computer-readable storage medium of claim 7, wherein determining the messaging plan is further based at least in part on an implicit delivery time preference of the user that is determined based at least in part on the metric representing the user.

14. The non-transitory computer-readable storage medium of claim 7, wherein the metric representing the user indicates at least one of a preferred notification channel, a preferred delivery time, or a preferred content of the notification.

15. A computer system comprising:

a memory configured to at least store computer-executable instructions, and

a processor in communication with the memory and configured to execute the computer-executable instructions to at least;

analyze a message for relevancy to a particular user based at least in part on metadata associated with the message and information associated with the particular user;

generate a score by applying machine learning to a metric associated with the particular user, the score corresponding to a probability that the particular user interacts with the message delivered via a channel of a plurality of channels during a time period of a plurality of time periods associated with the channel, the channel being unique with respect to other channels of the plurality of channels;

determine the channel and the time period for delivering the message to the particular user based at least in part on the score and the relevancy of the message to the particular user;

deliver the message to the particular user via the channel during the time period; and

in response to the user interacting with the message that is delivered,

receive an updated metric associated with the particular user, and

determine and updated channel and an updated time period for delivering a subsequent message to the user based at least in part on the updated metric.

16. The computer system of claim 15, wherein the processor is further configured to execute the specific computer-executable instructions to at least analyze the message to obtain topic information of the message and determine frequency of other messages with similar topic information.

17. The computer system of claim 16, wherein analyzing the message for relevancy to the particular user is further based at least in part on the determined frequency of other messages with similar topic information.

18. The computer system of claim 15, wherein the processor is further configured to execute the specific computer-executable instructions to at least maintain aggregated information for users regarding channels preferred by the users and delivery times preferred by the users.

19. The computer system of claim 18, wherein the processor is further configured to execute the specific computer-executable instructions to at least utilize the aggregated information to generate the score for the particular user when the metric associated with the particular user are insufficient.

20. The computer system of claim 15, wherein the processor is further configured to execute the specific computer-executable instructions to at least prioritize the message with reference to the particular user, the message being one of a plurality of messages intended for the particular user, the prioritization of the message based at least in part on the relevance of the message to the particular user.