Automatically managing email communications using indirect reply identity resolution
Methods and systems are enclosed herein for automatically managing email communication between a group of users and a group of target prospects. A sequence of outbound emails is automatically sent on behalf of a user to a prospect. Based upon the prospect's inbound replies (or lack thereof) the system will perform preconfigured actions, such as stopping automated communications and deferring to the user for manual action.
1. A non-transitory computer-readable medium comprising memory with instructions encoded thereon for automatically managing email communication between a user and a known prospect, the instructions, when executed by one or more processors, causing the one or more processors to perform operations, the instructions comprising instructions to:
detect an email received at an email inbox of the user, the email received responsive to a given message of a sequence;
extract metadata from a header of the email;
determine that the metadata matches information in an entry of a database, the entry corresponding to the known prospect;
determine a classification for the email by applying content of the email to a machine learning model and receiving as output from the machine learning model the classification; and
transmit an instruction to generate a next message of the sequence in accordance with an updated activity, the updated activity selected from a plurality of candidate activities based on the classification.
2. The non-transitory computer-readable medium of claim 1 , wherein the instructions to determine the classification comprise instructions to:
identify at least a portion of the content of the email; and
determine the classification based on the portion of content.
3. The non-transitory computer-readable medium of claim 2 , wherein the instructions to identify at least the portion of content of the email comprise instructions to extract a prefix from a subject line of the email, and wherein the instructions to determine the classification based on the portion of content comprise instructions to determine the classification based on the prefix.
4. The non-transitory computer-readable medium of claim 3 , wherein the classification is determined to be at least one of an out-of-office classification and a bounce classification.
5. The non-transitory computer-readable medium of claim 2 , wherein the content comprises one or more keywords, and wherein the instructions to determine the classification based on the portion of content is based on at least one of the one or more keywords corresponding to the classification.
6. The non-transitory computer-readable medium of claim 1 , the instructions further comprising instructions to:
determine a sender email address of the email;
determine whether the sender email address corresponds to a given classification; and
responsive to determining that the sender email address corresponds to the given classification, determine the classification to be the given classification.
7. The non-transitory computer-readable medium of claim 6 , wherein the classification is determined to be a bounce classification.
8. The non-transitory computer-readable medium of claim 1 , wherein the classification is determined to be a reply classification.
9. The non-transitory computer-readable medium of claim 1 , wherein the next message of the sequence is a message scheduled to be transmitted on a basis of a particular time.
10. The non-transitory computer-readable medium of claim 9 , wherein the updated activity is selected based on a pre-programmed workflow indicating an activity that is to occur responsive to receiving a given email from the known prospect having the classification.
11. A method for automatically managing email communication between a user and a known prospect, the method comprising:
detecting an email received at an email inbox of the user, the email received responsive to a given message of a sequence;
extracting metadata from a header of the email;
determining that the metadata matches information in an entry of a database, the entry corresponding to the known prospect;
determining a classification for the email by applying content of the email to a machine learning model and receiving as output from the machine learning model the classification; and
transmitting an instruction to generate a next message of the sequence in accordance with an updated activity, the updated activity selected from a plurality of candidate activities based on the classification.
12. The method of claim 11 , wherein determining the classification comprises:
identify at least a portion of the content of the email; and
determine the classification based on the portion of content.
13. The method of claim 12 , wherein identifying at least the portion of content of the email comprises extracting a prefix from a subject line of the email, and wherein determining the classification based on the portion of content comprises determining the classification based on the prefix.
14. The method of claim 13 , wherein the classification is determined to be at least one of an out-of-office classification and a bounce classification.
15. The method of claim 12 , wherein the content comprises one or more keywords, and wherein determining the classification based on the portion of content is based on at least one of the one or more keywords corresponding to the classification.
16. The method of claim 11 , the method further comprising:
determining a sender email address of the email;
determining whether the sender email address corresponds to a given classification; and
responsive to determining that the sender email address corresponds to the given classification, determining the classification to be the given classification.
17. The method of claim 16 , wherein the classification is determined to be a bounce classification.
18. The method of claim 11 , wherein the classification is determined to be a reply classification.
19. The method of claim 11 , wherein the next message of the sequence is a message scheduled to be transmitted on a basis of a particular time.
20. A system comprising:
memory with instructions encoded thereon for automatically managing email communication between a user and a known prospect; and
one or more processors that, when executing the instructions, are caused to perform operations comprising:
detecting an email received at an email inbox of the user, the email received responsive to a given message of a sequence;
extracting metadata from a header of the email;
determining that the metadata matches information in an entry of a database, the entry corresponding to the known prospect;
determining a classification for the email by applying content of the email to a machine learning model and receiving as output from the machine learning model the classification; and
transmitting an instruction to generate a next message of the sequence in accordance with an updated activity, the updated activity selected from a plurality of candidate activities based on the classification.