Systems and methods for recurrent and adaptive electronic notifications

1. A system for generating adaptive electronic notifications comprising:

at least one processor; and

at least one memory device comprising instructions that when executed configure the at least one processor to perform operations comprising:

generating an item list for notifications to a client ID, the item list being associated with category IDs and product IDs, items in the item list being associated with priority scores and lifecycle timers;

selecting a first item from the item list for notification, the first item being selected having a highest priority score and a lowest lifecycle timer, the first item being associated with a first product code and a first category ID;

encoding a first landing page by encoding the first product code and a first tracking cookie in a first URL, the first URL being configured for redirection to a first webpage of the first item by a redirect server;

transmitting, to a client device, a first notification with a payload comprising the first landing page and a first message;

after transmitting the first notification, modifying at least one of the priority scores and updating at least one of the lifecycle timers based on determining, using the first tracking cookie, whether the client device visited the first landing page; and

after modifying the priority scores, selecting a second item from the item list for notification, the second item being associated with a second product code and a second category ID different from the first category ID.

2. The system of claim 1 , wherein the operations further comprise:

encoding a second landing page by encoding the second product code and a second tracking cookie in a second URL, the second URL being configured for redirection to a second webpage of the second item by the redirect server;

in response to determining a notification time lapsed after transmitting the first notification, transmitting, to the client device, a second notification with a second payload comprising the second landing page and a second message;

after transmitting the second notification, modifying at least one of the priority scores and updating at least one of the lifecycle timers based on determining, using the second tracking cookie, whether the client device visited the second landing page; and

selecting a third item from one of the item list for notification, the third item being associated with a third product code and a third category ID, the third category ID being different from the first category ID and the second category ID.

3. The system of claim 2 , wherein the operations further comprise:

encoding a third landing page by encoding the third product code and a third tracking cookie in a third URL, the third URL being configured for redirection to a third webpage of the third item by the redirect server.

4. The system of claim 1 , wherein:

determining whether the client device visited the first landing page comprises searching the first tracking cookie in redirection logs of the redirect server; and

modifying the priority scores and updating the lifecycle timers:

in response to determining the client device visited the first landing page, increasing a priority score of the first item and resetting a lifecycle timer of the first item; and

in response to determining the client device did not visit the first landing page, decreasing the priority score and delaying the lifecycle timer.

5. The system of claim 1 , wherein generating the item list comprises:

identifying, from a database, products that are ordered periodically and assigning the product IDs;

categorizing the products in product categories and assigning the category IDs;

calculating average times between repeat purchases of the products;

ranking each one of product categories by determining a category affinity and assigning higher rankings to the product categories with smallest average times; and

assigning the priority scores by:

assigning initial priority scores based on previous purchases associated with the client ID;

increasing the priority scores of items with higher ranked product categories;

increasing the priority scores based on an inclusion function that increases the priority scores of items recently included in the item list;

increasing the priority scores proportionally to a number of times the client ID has ordered the product; and

decreasing the priority scores proportionally with respective ones of the lifecycle timers.

6. The system of claim 5 , wherein generating the item list further comprises assigning the lifecycle timers equal to the average times.

7. The system of claim 1 , wherein:

the first tracking cookie encodes a source ID specifying a campaign source and the client ID; and

generating the notification comprises:

retrieving a plurality of images associated with the first item;

selecting a smallest image from the plurality of images as a first product image;

evaluating whether the payload is below a threshold weight; and

in response to determining the payload is above the threshold weight, generating a transformed image for notification.

8. The system of claim 1 , wherein generating the item list comprises:

displaying a graphical user interface to a manager, the graphical user interface comprising the category IDs and input icons for category rank, maximum number of notifications, and days between notifications; and

the operations further comprise:

collecting campaign goals and manager-defined parameters from the graphical user interface; and

adjusting the priority scores and lifecycle timers based on the campaign goals and the manager-defined parameters.

9. The system of claim 1 , wherein the operations further comprise:

generating a training dataset associating transmitted notifications and redirection logs from the redirect server;

training a machine learning model using the training dataset;

employing the machine learning model to identify the notifications with highest click through rates; and

adjusting the priority scores and rankings of the category IDs based on the click through rates.

10. The system of claim 1 , wherein:

the first notification is a first push notification;

the payload further comprises a first product image; and

the first tracking cookie encodes an index to the first message and the first product image;

transmitting the first push notification comprises:

selecting the first message from a legacy message or a testing message, the legacy message being different from the testing message;

selecting the first product image from a legacy image or a testing image, the legacy image being different from the testing image; and

the operations further comprise identifying a message and image combination with highest clickthrough rate based on a message-image combination selected from a group comprising: legacy message and legacy image, legacy message and testing image; testing message and legacy image; or testing message and testing image.

11. A computer-implemented method for generating adaptive electronic notifications

comprising:

generating an item list for notifications to a client ID, the item list being associated with category IDs and product IDs, items in the item list being associated with priority scores and lifecycle timers;

selecting a first item from the item list for notification, the first item being selected having a highest priority score and a lowest lifecycle timer, the first item being associated with a first product code and a first category ID;

encoding a first landing page by encoding the first product code and a first tracking cookie in a first URL, the first URL being configured for redirection to a first webpage of the first item by a redirect server;

transmitting, to a client device, a first notification with a payload comprising the first landing page and a first message;

after transmitting the first notification, modifying at least one of the priority scores and updating at least one of the lifecycle timers based on determining, using the first tracking cookie, whether the client device visited the first landing page; and

after modifying the priority scores, selecting a second item from the item list for notification, the second item being associated with a second product code and a second category ID different from the first category ID.

12. The computer-implemented method of claim 11 , further comprising:

encoding a second landing page by encoding the second product code and a second tracking cookie in a second URL, the second URL being configured for redirection to a second webpage of the second item by the redirect server;

in response to determining a notification time lapsed after transmitting the first notification, transmitting, to the client device, a second notification with a second payload comprising the second landing page and a second message;

after transmitting the second notification, modifying at least one of the priority scores and updating at least one of the lifecycle timers based on determining, using the second tracking cookie, whether the client device visited the second landing page; and

selecting a third item from one of the item list for notification, the third item being associated with a third product code and a third category ID, the third category ID being different from the first category ID and the second category ID.

13. The computer-implemented method of claim 12 , further comprising:

encoding a third landing page by encoding the third product code and a third tracking cookie in a third URL, the third URL being configured for redirection to a third webpage of the third item by the redirect server.

14. The computer-implemented method of claim 11 , wherein:

determining whether the client device visited the first landing page comprises searching the first tracking cookie in redirection logs of the redirect server; and

modifying the priority scores and updating the lifecycle timers:

in response to determining the client device visited the first landing page, increasing a priority score of the first item and resetting a lifecycle timer of the first item; and

in response to determining the client device did not visit the first landing page, decreasing the priority score and delaying the lifecycle timer.

15. The computer-implemented method of claim 11 , wherein generating the item list comprises:

identifying, from a database, products that are ordered periodically and assigning the product IDs;

categorizing the products in product categories and assigning the category IDs;

calculating average times between repeat purchases of the products;

ranking each one of product categories by determining a category affinity and assigning higher rankings to the product categories with smallest average times; and

assigning the priority scores by:

assigning initial priority scores based on previous purchases associated with the client ID;

increasing the priority scores of items with higher ranked product categories;

increasing the priority scores based on an inclusion function that increases the priority scores of items recently included in the item list;

increasing the priority scores proportionally to a number of times the client ID has ordered the product; and

decreasing the priority scores proportionally with respective ones of the lifecycle timers.

16. The computer-implemented method of claim 15 , wherein generating the item list further comprises assigning the lifecycle timers equal to the average times.

17. The computer-implemented method of claim 11 , wherein:

the first tracking cookie encodes a source ID specifying a campaign source and the client ID; and

generating the notification comprises:

retrieving a plurality of images associated with the first item;

selecting a smallest image from the plurality of images as a first product image;

evaluating whether the payload is below a threshold weight; and

in response to determining the payload is above the threshold weight, generating a transformed image for notification.

18. The computer implemented method of claim 11 , wherein generating the item list comprises:

displaying a graphical user interface to a manager, the graphical user interface comprising the categories IDs and input icons for category rank, maximum number of notifications, and days between notifications; and

the method further comprises:

collecting campaign goals and manager-defined parameters from the graphical user interface; and

adjusting the priority scores and lifecycle timers based on the campaign goals and the manager-defined parameters.

19. The computer-implemented method of claim 11 , further comprising:

generating a training dataset associating transmitted notifications and redirection logs from the redirect server;

training a machine learning model using the training dataset;

employing the machine learning model to identify the notifications with highest click through rates; and

adjusting the priority scores and rankings of the category IDs based on the click through rates.

20. An apparatus comprising:

one or more processors; and

one or more memory devices comprising instructions that configure the one or more processors to:

generate an item list for notifications to a client ID, the item list being associated with category IDs and product IDs, items in the item list being associated with priority scores and lifecycle timers;

select a first item from the item list for notification, the first item being selected having a highest priority score and a lowest lifecycle timer, the first item being associated with a first product code and a first category ID;

encode a first landing page by encoding the first product code and a first tracking cookie in a first URL, the first URL being configured for redirection to a first webpage of the first item by a redirect server;

transmit, to a client device, a first notification with a payload comprising the first landing page and a first message;

after transmitting the first notification, modify at least one of the priority scores and updating at least one of the lifecycle timers based on determining, using the tracking cookie, whether the client device visited the first landing page;

after modifying the priority scores, select a second item from the item list for notification, the second item being associated with a second product code and a second category ID, different from the first category ID;

encode a second landing page by encoding the second product code and a second tracking cookie in a second URL, the second URL being configured for redirection to a second webpage of the second item by the redirect server;

in response to determining a notification time lapsed after transmitting the first notification, transmit, to the client device, a second notification with a second payload comprising the second landing page and a second message;

after transmitting the second notification, modify the priority scores and updating the lifecycle timers based on determining, using the second tracking cookie, whether the client device visited the second landing page; and

select a third item from one of the item list for notification, the third item being associated with a third product code and a third category ID, the third category ID being different from the first category ID and the second category ID.