IP Library Granted Patent US 10,931,623
Granted Patent B2
US 10,931,623 · App. 15/884,273 · Granted Feb 23, 2021

Introducing a new message source into an electronic message delivery environment

Inventors: Michael O'Brien (Skerries, IE); Michael Axiak (Somerville, MA)
Assignee: HUBSPOT, INC.
H04L51/34G06N3/08G06N7/005G06N20/00G06Q10/107G06Q30/0248G06Q30/0277H04L51/08H04L51/12H04L51/18H04L51/22H04L51/28H04L61/2007
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,931,623
App. No.
15/884,273
Filed
Jan 30, 2018
Granted
Feb 23, 2021
Kind
B2
Examiner
YE, ZI
Art Unit
2455
USPC
709/206
Abstract

Provided herein is an electronic message management platform that enables management and execution of electronic message campaigns while appropriately managing challenges presented by spam filters, black lists, and domain blocking technologies, and that includes elements for managing an electronic message campaign based on dynamic conditions, quality measures, engagement factors, and other measures, factors and conditions, including a rate of negative delivery events and a rate of negative engagement events.

Claims (34)

1. A method of introducing a resource for a new sender for sending email messages that mitigates a risk that the resource is marked for rejection of the email messages, comprising:

processing a list of recipients provided by the new sender with a quality scoring algorithm that detects recipients that score below a baseline new sender introductory score threshold;

identifying a first subset of recipients from the list of recipients as excluded from receiving the email messages from the new sender during a new sender introduction phase based at least in part on the quality scoring algorithm;

calculating a probability of engagement for each email message, the probability of engagement including a likelihood that a recipient will engage with each email message;

sending the email messages to a subset of the list of recipients that are not in the first subset of recipients during the new sender introduction phase, wherein the email messages that are more likely to engage are sent before the email messages that are less likely to engage according to the probability of engagement for each email message; and

adjusting the new sender introduction phase based on a rate of negative engagement events for the email messages sent during the new sender introduction phase.

2. The method of claim 1 , wherein the quality scoring algorithm indicates to exclude recipients that have resulted in at least one hard bounce of a message previously sent to a recipient.

3. The method of claim 1 , wherein the quality scoring algorithm comprises calculating a recipient quality score based on at least one message engagement event for a recipient with respect to at least one past message.

4. The method of claim 1 , wherein the quality scoring algorithm comprises applying at least one model for calculating the probability of engagement to identify recipients for excluding from message delivery during the new sender introduction phase.

5. The method of claim 4 , wherein the at least one model is at least one of a logistic regression model or a beta distribution model.

6. The method of claim 1 , further comprising not sending the email messages to the first subset of recipients that are identified during the new sender introduction phase.

7. The method of claim 1 , further comprising adjusting the baseline new sender introductory score threshold based on a measure of recipient engagement with the email messages sent to the subset of recipients.

8. The method of claim 1 , wherein the quality scoring algorithm determines a quality score based on a number of recipient open events and recipient message link engagement events recorded for a recipient.

9. The method of claim 1 , wherein the quality scoring algorithm determines a quality score based on a reputation for a source of recipient contact information.

10. The method of claim 1 , wherein identifying the first subset of recipients is based on a recipient status indicative of a message address of a recipient being invalid.

11. The method of claim 1 , wherein identifying the first subset of recipients is based on a recipient status indicative of a recipient opting out from receiving message from senders that are similar to the new sender.

12. The method of claim 1 , wherein identifying the first subset of recipients is based on a recipient status indicative of a recipient designating a new sender domain as a source of spam.

13. The method of claim 1 , wherein adjusting the new sender introduction phase comprises stopping the new sender introduction phase.

14. The method of claim 1 , further comprising sending the email messages to a sample number of recipients of the subset of the list of recipients and wherein the adjusting the new sender introduction phase is based on at least one of the rate of negative delivery events and the rate of negative engagement events for the email messages sent to the sample number of recipients during the new sender introduction phase.

15. The method of claim 1 , further comprising sending the email messages to a sample number of recipients of the subset of the list of recipients and wherein adjusting the new sender introduction phase comprises skipping sending the email messages to a remainder of the subset of the list of recipients when at least one of the rate of negative delivery events and the rate of negative engagement events for the email messages exceed a threshold.

16. A non-transitory computer readable medium having stored thereon software instructions that, when executed by a processor, cause the processor to perform a method of introducing a resource for a new sender for sending email messages that mitigates a risk that the resource is marked for rejection of the email messages, the method comprising:

processing a list of recipients provided by the new sender with a quality scoring algorithm that detects recipients that score below a baseline new sender introductory score threshold;

identifying a first subset of recipients from the list of recipients as excluded from receiving the email messages from the new sender during a new sender introduction phase based at least in part on the quality scoring algorithm;

calculating a probability of engagement for each email message, the probability of engagement including a likelihood that a recipient will engage with each email message;

sending the email messages to a subset of the list of recipients that are not in the first subset of recipients during the new sender introduction phase, wherein the email messages that are more likely to engage are sent before the email messages that are less likely to engage according to the probability of engagement for each email message; and

adjusting the new sender introduction phase based on a rate of negative engagement events for the email messages sent during the new sender introduction phase.

17. The non-transitory computer readable medium of claim 16 , wherein the quality scoring algorithm indicates to exclude recipients that have resulted in at least one hard bounce of a message previously sent to a recipient.

18. The non-transitory computer readable medium of claim 16 , wherein the quality scoring algorithm comprises calculating a recipient quality score based on at least one message engagement event for a recipient with respect to at least one past message.

19. The non-transitory computer readable medium of claim 16 , wherein the quality scoring algorithm comprises applying at least one model for calculating the probability of engagement to identify recipients for excluding from message delivery during the new sender introduction phase.

20. The non-transitory computer readable medium of claim 19 , wherein the at least one model is at least one of a logistic regression model and a beta distribution model.

21. The non-transitory computer readable medium of claim 16 , wherein identifying the first subset of recipients is based on a recipient status indicative of a recipient opting out from receiving message from senders that are similar to the new sender.

22. The non-transitory computer readable medium of claim 16 , wherein adjusting the new sender introduction phase comprises stopping the new sender introduction phase.

23. The non-transitory computer readable medium of claim 16 , further comprising sending the email messages to a sample number of recipients of the subset of the list of recipients and wherein the adjusting the new sender introduction phase is based on at least one of a rate of negative delivery events and the rate of negative engagement events for the email messages sent to the sample number of recipients during the new sender introduction phase.

24. The non-transitory computer readable medium of claim 16 , further comprising sending the email messages to a sample number of recipients of the subset of the list of recipients and wherein adjusting the new sender introduction phase comprises skipping sending the email messages to a remainder of the subset of the list of recipients when at least one of the rate of negative delivery events and the rate of negative engagement events for the email messages exceed a threshold.

Assignments (2)
SECURITY INTEREST Recorded Feb 12, 2026
From: HUBSPOT, INC.
To: BANK OF AMERICA, N.A.
Reel/Frame 074818/0036 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 14, 2018
From: O'BRIEN, MICHAEL; AXIAK, MICHAEL
To: HUBSPOT INC.
Reel/Frame 044928/0561 →
Continuity (2)
Provisional Application 62451974 · Jan 30, 2017
Related Publication 20180219830A1 · Aug 2, 2018
Cited By (5)
US 12,271,847 US 12,271,926 US 12,493,582 US 12,511,256 US 12,688,474