Granted Patent B2

US 12,596,720 · App. 18/815,993 · Granted Apr 7, 2026

Inventors: Manuel Ventero Pena (Cambridge, MA); Carola Leiva (Brookline, MA); Alex Riina (Cambridge, MA); Stanley Wu (Chicago, IL); Robert Huselid (Stamford, CT); Annika Sougstad (Boston, MA); Alexandra Holness (Framingham, MA); Jordan Cora Chase (Salem, MA); Christopher Fox (Westport, CT); Jingwen Wu (Somerville, MA); Xiao Liu (Orinda, CA); Timothy Tadros (Boston, MA); Nicolas Asnes (Cohasset, MA); Elliot Penson (Braintree, MA); Dillon Jones (Medford, MA)

Assignee: Klaviyo, Inc.

G06F16/2457G06F16/2452

Abstract

Claims (68)

1 . An apparatus, comprising:

a segmentation server;

a user server connected through a network to the segmentation server;

a plurality of sub-user computing devices connected through the network to the user server and the segmentation server;

wherein the segmentation server is configured to:

a. receive a segmentation description from a user of the user server;

b. retrieve user data from a database;

c. translate the segmentation description to a standardized segmentation system vernacular based on the segmentation description and the retrieved user data;

d. identify a set of individual criteria contained within the standardized segmentation description vernacular mapped to variable names;

e. express the standardized segmentation system vernacular as a logical statement in terms of the variable names;

f. parse each individual criteria of the set of individual criteria against sub-user data of the user data of the database in parallel yielding a representation of each criteria of the set of individual criteria;

g. identify sub-users of a segment by logically combining the representation of each criteria according to the logical statement; and

h. perform an electronic action directed to the sub-users identified by the segment, comprising:

loading user-tracking code to a browser of each of the sub-users when each of the sub-users loads a webpage of the user;

sending electronic messages to each of the sub-users identified by the segment, wherein different electronic messages include a display having a different behavior;

sensing, by the user-tracking code, actions of each of the sub-users in response to receiving the electronic messages; and

adjusting the different behavior of the display of the different electronic messages based on the sensed actions to improve a user interface of the display of the sub-users.

2 . The apparatus of claim 1 , wherein the segmentation server is further configured to:

create suggested segment descriptions;

display the suggested segment descriptions;

wherein the segment description received from the user is based on the suggested segment descriptions.

3 . The apparatus of claim 1 , wherein the segmentation server is further configured to:

compare list names and metrics called out in the segmentation description against the metrics and information retrieved from the user data to find a best match for each;

wherein the best matches for each of the list name and metrics are translated to the standardized segmentation system vernacular based on the segmentation description and the retrieved user data.

4 . The apparatus of claim 1 , wherein the segmentation server is further configured to:

display the standardized segmentation system vernacular to the user;

sense user actions by the user;

feed the sensed user actions back to a translator that performs the translation.

5 . The apparatus of claim 1 , wherein the yielded representation of each criteria of the set of individual criteria is a JSON (JavaScript object syntax) object.

6 . The apparatus of 1 , further comprising adjusting future translations based on the sensed actions.

7 . The apparatus of 1 , further comprising adjusting parsing of future individual criteria based on the sensed actions.

8 . The apparatus of claim 1 , wherein the segmentation server is further configured to:

generate two or more versions of identified sub-users of two of more segmentation versions based on two or more versions of the standardized segmentation system vernacular;

electronically send electronic messages to the two or more versions of identified sub-users;

monitor actions of the two or more versions of identified sub-users based on responses to receiving the electronic messages; and

rank the two or more versions of the standardized segmentation system vernacular based on the monitored actions.

9 . The apparatus of claim 8 , wherein the segmentation server is further configured to train future translations of segmentation descriptions to the standardized segmentation system vernacular based on the rankings of the two or more versions of the standardized segmentation system vernacular.

10 . The apparatus of claim 8 , wherein the segmentation server is further configured to train future parsing based on the rankings of the two or more versions of the standardized segmentation system vernacular.

11 . The apparatus of claim 1 , wherein the electronic messages are electronically generated based on message inputs of the user, and further comprising tuning the generation of the electronic messages based on sensing actions of the user and based on sensing actions of the sub-user in response to electronically receiving the electronic message.

12 . The apparatus of claim 1 , further comprising:

selecting one or more send times for the electronic messages; and

adjusting send times for future generated electronic messages based on sensed actions of the sub-users in response to receiving the electronic messages.

13 . The apparatus of claim 1 , wherein a selectable button is located on the display of the electronic messages based on the sensed actions.

14 . A method, comprising:

receiving a segmentation description from a user of the user server;

retrieving user data from a database;

translating the segmentation description to a standardized segmentation system vernacular based on the segmentation description and the retrieved user data;

identifying a set of individual criteria contained within the standardized segmentation description vernacular mapped to variable names;

expressing the standardized segmentation system vernacular as a logical statement in terms of the variable names;

parsing each individual criteria of the set of individual criteria against sub-user data of the user data of the database in parallel yielding a representation of each criteria of the set of individual criteria;

identifying sub-users of a segment by logically combining the representation of each criteria according to the logical statement; and

performing an electronic action directed to the sub-users identified by the segment, comprising:

loading user-tracking code to a browser of each of the sub-users when each of the sub-users loads a webpage of the user;

sending electronic messages to each of the sub-users identified by the segment, wherein different electronic messages include a display having a different behavior;

sensing, by the user-tracking code, actions of each of the sub-users in response to receiving the electronic messages; and

adjusting the different behavior of the display of the different electronic messages based on the sensed actions to improve a user interface of the display of the sub-users.

15 . The method of claim 14 , further comprising:

displaying the standardized segmentation system vernacular to the user;

sensing user actions by the user;

feeding the sensed user actions back to a translator that performs the translation;

training the translator with the sensed user action for future translations.

16 . The method of claim 14 , further comprising:

adjusting future translations based on the sensed actions.

17 . The method of claim 14 , further comprising:

adjusting parsing of future individual criteria based on the sensed actions.

18 . The method of claim 14 , further comprising adjusting future translations based on the sensed actions.

19 . The method of claim 14 , further comprising adjusting parsing of future individual criteria based on the sensed actions.

20 . The method of claim 14 , wherein the electronic messages are electronically generated based on message inputs of the user, and further comprising tuning the generation of the electronic messages based on sensing actions of the user and based on sensing actions of the sub-user in response to electronically receiving the electronic message.

Assignments (1)

Continuity (1)

References Cited (43)