US2023410020A1PendingUtilityA1
Systems and methods for real-time lead grading
Est. expiryDec 31, 2040(~14.4 yrs left)· nominal 20-yr term from priority
G06Q 10/06393
48
PatentIndex Score
0
Cited by
0
References
0
Claims
Abstract
A method for a lead grading platform can include: receiving a lead from a lead vendor, wherein the lead comprises lead attribute values; converting the lead attribute values to benchmark values; assigning the lead to a pre-determined cluster and a pre-determined sub-cluster based on the benchmark values; determining a grade for the lead based on the pre-determined cluster and the pre-determined sub-cluster; and responding to the grade being greater than a threshold grade by delivering the lead to a lead consumer.
Claims
exact text as granted — not AI-modified1 . A method for a lead grading platform, the method comprising:
receiving a lead from a lead vendor, wherein the lead comprises a plurality of lead attribute values; converting the plurality of lead attribute values to a plurality of benchmark values; assigning the lead to a pre-determined cluster and a pre-determined sub-cluster based on the plurality of benchmark values; determining a grade for the lead based on the pre-determined cluster and the pre-determined sub-cluster; and responding to the grade being greater than a threshold grade by:
delivering the lead to a lead consumer.
2 . The method of claim 1 , wherein converting the plurality of lead attribute values to the plurality of benchmark values comprises:
determining for a lead attribute value, V, a corresponding benchmark value B A=V , according to:
B
A
=
V
=
CRate
A
=
V
CRate
wherein CRate A=V is given by:
CRate
A
=
V
=
Conv
e
r
s
ions
A
=
V
L
e
a
d
s
A
=
V
wherein Conversions A=V is a number of conversions of leads with the lead attribute value, V, previously delivered to the lead consumer, and wherein Leads A=V is a number of leads with the lead attribute value, V, previously delivered to the lead consumer, and wherein CRate is given by:
CRate
=
C
o
n
v
e
r
s
i
o
n
s
L
e
a
d
s
wherein Leads is a total number of leads previously delivered to the lead consumer, and wherein Conversions is a total number of conversions resulting from the total number of leads previously delivered to the lead consumer.
3 . The method of claim 1 , wherein assigning the lead to the pre-determined cluster and the pre-determined sub-cluster based on the plurality of benchmark values comprises:
mapping the plurality of benchmark values to the pre-determined cluster and the pre-determined sub-cluster using a trained neural network.
4 . The method of claim 1 , the method further comprising:
responding to the grade being less than the threshold grade by:
not delivering the lead to the lead consumer.
5 . The method of claim 1 , wherein the plurality of lead attribute values comprise one or more of:
personal contact information; lead source information; and product information.
6 . The method of claim 5 , wherein the personal contact information includes one or more of:
first name; last name; email address; phone number; mailing address; and residence address.
7 . The method of claim 5 , wherein the lead source information includes one or more of:
campaign; campaign tactic; and lead provider.
8 . The method of claim 5 , wherein the product information includes one or more of:
type of product; price of product; product name; and product characteristics.
9 . A lead grading system, comprising:
a processor; and non-transitory memory storing instructions, that when executed cause the processor to:
receive a plurality of leads and a plurality of conversion event records for the plurality of leads, wherein each of the plurality of leads comprises a plurality of lead attribute values;
convert the plurality of lead attribute values into a plurality of benchmark values based on the plurality of conversion event records;
filter the plurality of benchmark values to produce a plurality of filtered benchmark values;
encode the plurality of leads as a plurality of lead vectors comprising the plurality of filtered benchmark values;
cluster the plurality of lead vectors to form a plurality of clusters;
determine average conversion probabilities for each of the plurality of clusters; and
grade the plurality of clusters based on the average conversion probabilities of each of the plurality of clusters.
10 . The lead grading system of claim 9 , wherein the processor is configured to filter the plurality of benchmark values by:
determining an average benchmark value for a subset of the plurality of benchmark values corresponding to a first attribute; determining if the average benchmark value is within a pre-determined range; and responding to the average benchmark value being within the pre-determined range by:
filtering out the subset of the plurality of benchmark values from the plurality of benchmark values.
11 . The lead grading system of claim 9 , wherein the processor is configured to filter the plurality of benchmark values by:
determining a degree of correlation between a first subset of the plurality of benchmark values and a second subset of the plurality of benchmark values, wherein the first subset of the plurality of benchmark values corresponds to a first attribute, and wherein the second subset of the plurality of benchmark values corresponds to a second attribute; and responding to the degree of correlation being greater than a threshold by removing one of the first subset of the plurality of benchmark values or the second subset of the plurality of benchmark values.
12 . The lead grading system of claim 9 , wherein the processor, when executing the instructions, is further configured to:
map the plurality of lead vectors from a first lead vector space to a second lead vector space, wherein the second lead vector space is of lower dimension than the first lead vector space.Join the waitlist — get patent alerts
Track US2023410020A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.