IP Library Granted Patent US 11,368,752
Granted Patent B2
US 11,368,752 · App. 17/186,688 · Granted Jun 21, 2022

Optimization of broadcast event effectiveness

Inventors: Justin Manus (Oakland, CA); Sean Odlum (Santa Monica, CA); Anand V. Bodapati (Los Angeles, CA)
Assignee: Bliss Point Media, Inc.
H04N21/44222G06N7/005G06N20/00G06Q10/06393H04N21/2407H04N21/25H04N21/812
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 11,368,752
App. No.
17/186,688
Granted
Jun 21, 2022
Kind
B2
Abstract

Introduced herein are methods and systems for determining machine learning marketing strategy. For example, a computer-implemented method according to the disclosed technology includes steps of identifying one or more business metrics to be driven by a marketing plan; generating one or more response functions of the business metrics by performing a machine learning process on a marketing dataset; optimizing a spending subject of the marking plan subject to constraints to generate a marketing strategy based on multiple decision variables; and presenting the marketing strategy to an advertiser.

Claims (85)

1. A computer-implemented method for determining measurable responses to broadcast events, comprising:

storing, in memory, a broadcast event history, the broadcast event history including metadata of a plurality of broadcast events;

identifying, by a processor from the metadata, a first time stamp for a first broadcast event of the plurality of broadcast events and a second time stamp for a second broadcast event of the plurality of broadcast events, each broadcast event respectively including a medium and a channel;

determining a quantifiable metric to tie to the first and second broadcast events, the quantifiable metric including any of: views, downloads, phone calls, clicks, purchases, sign-ups, signatures, logins, or application interactions;

receiving, by the memory, a plurality of quantifiable metric performance history data including the quantifiable metric measured as a function of time, wherein the plurality of quantifiable metric performance history data does not identify attribution to specific broadcast events;

assigning, by the processor, a first time period to the first broadcast event and a second period of time to the second broadcast event based upon a recorded time of a change in the quantifiable metric in the quantifiable metric performance history data subsequent to a conclusion of each respective broadcast event, wherein the first and second time periods are temporal periods after the first and second time stamps respectively;

assigning, by the processor, a first portion of the quantifiable metric performance history data to the first broadcast event, the first portion associated with the first time period;

determining, by the processor, an overlapping portion of the quantifiable metric performance history data, the overlapping portion including quantifiable metric performance history data from the first portion and measured after the second time stamp;

determining, by the processor, that the overlapping portion is not attributable to the second broadcast event;

assigning, by the processor, a second portion of the quantifiable metric performance history data to the second broadcast event, the second portion associated with the second time period and not including the overlapping portion;

comparing, by the processor, the first portion to the second portion; and

determining, by the processor, an effectiveness rating of the first broadcast event as relative to the second broadcast event based upon said comparing of the first portion and the second portion.

2. A computer-implemented method for determining measurable responses to broadcast events, comprising:

receiving, by a memory, a plurality of quantifiable metric performance history data including a quantifiable metric measured as a function of time, wherein the plurality of quantifiable metric performance history data does not identify attribution to specific broadcast events;

assigning, by a processor, a first time period to a first broadcast event and a second period of time to a second broadcast event, each broadcast event respectively including a time stamp, a medium, and a channel, wherein a length for each of the first and second time periods is assigned based upon a recorded time of a change in the quantifiable metric in the quantifiable metric performance history data subsequent to the time stamp of each respective broadcast event, wherein the first and second time periods are temporal periods after each of the first and second broadcast events conclude respectively;

assigning, by the processor, a first portion of the quantifiable metric performance history data to the first broadcast event, the first portion associated with the first time period;

determining, by the processor, an overlapping portion of the quantifiable metric performance history data, the overlapping portion including quantifiable metric performance history data from the first portion and measured after the second time stamp;

determining, by the processor, that the overlapping portion is not attributable to the second broadcast event;

assigning, by the processor, a second portion of the quantifiable metric performance history data to the second broadcast event, the second portion associated with the second time period and not including the overlapping portion;

comparing, by the processor, the first portion to the second portion; and

determining, by the processor, an effectiveness rating of the first broadcast event as relative to the second broadcast event based upon said comparing of the first portion and the second portion.

3. The method of claim 2 , wherein the quantifiable metric are any of key performance indicators (KPIs) including views to a website or an application, sign-ups of the website or an application, downloads of an application, clicks on the website or application, verified purchases, signatures, logins to the website or application, phone calls, and detected user activity on the website or application.

4. The method of claim 2 , wherein the medium is any of:

print;

picture;

video;

audio; or

multimedia;

and the channel is any of:

a website;

a television channel;

a streaming video or video on demand service;

a podcast personality;

a radio station; or

signage.

5. The method of claim 2 , further comprising:

updating, in the memory, the quantifiable metric performance history data;

updating, by the processor, the first portion and the second portion based on said updating of the quantifiable metric performance history data; and

determining, by the processor, an updated effectiveness rating of the first broadcast event as relative to the second broadcast event based upon said updating of the quantifiable metric performance history data and the first and second portions.

6. The method of claim 2 , wherein each broadcast event further includes an emotional tone and the emotional tone is used as a further basis for said determining an effectiveness rating.

7. The method of claim 2 , wherein each broadcast event further includes an event length and the event length is used as a further basis for said determining an effectiveness rating.

8. The method of claim 2 , wherein budget, day of week, national or local availabilities, inventory status, age suitability, or spending timeframe is used as a further basis for said determining an effectiveness rating.

9. The method of claim 2 , further comprising:

submitting one or more orders to media platforms based on the effectiveness rating.

10. A computer-implemented method for determining measurable responses to broadcast events, comprising:

receiving, by a memory, a plurality of quantifiable metric performance history data including a quantifiable metric measured as a function of time, wherein the plurality of quantifiable metric performance history data does not identify attribution to specific broadcast events;

receiving, by the memory, a broadcast event history including metadata of a plurality of broadcast events, the broadcast events each having a time stamp, a channel, and a medium, wherein each of the time stamps, channels, and mediums are varied across the plurality of broadcast events;

assigning, by a processor, a first portion of the quantifiable metric performance history data to a first broadcast event, the first portion associated with a first time period after a first time stamp for the first broadcast event;

determining, by the processor, an overlapping portion of the quantifiable metric performance history data, the overlapping portion including quantifiable metric performance history data from the first portion and measured after a second time stamp for a second broadcast event;

determining, by the processor, that the overlapping portion is not attributable to the second broadcast event;

assigning, by the processor, a second portion of the quantifiable metric performance history data to the second broadcast event, the second portion associated with a second time period after the second time stamp and not including the overlapping portion;

assigning, by the processor, an effectiveness period to each of the broadcast events, wherein the effectiveness period of each broadcast event is a period of time after a conclusion of each respective broadcast event; and

determining an effectiveness rating of each of the time stamps, channels, and mediums based on a status of the quantifiable metric during the effectiveness period of each of the broadcast events.

11. The method of claim 10 , wherein the quantifiable metric are any of key performance indicators (KPIs) including views to a website or an application, sign-ups of the website or an application, downloads of an application, clicks on the website or application, verified purchases, signatures, logins to the website or application, phone calls, and detected user activity on the website or application.

12. The method of claim 10 , wherein the medium is any of:

print;

picture;

video;

audio; or

multimedia;

and the channel is any of:

a website;

a streaming video or video on demand service;

a television channel;

a podcast personality;

a radio station; or

signage.

13. The method of claim 10 , wherein each broadcast event further includes an emotional tone and an event length, and wherein said determining includes determining the effectiveness rating for the emotional tone and the event length.

14. The method of claim 10 , further comprising:

determining, for each combination of the time stamp, the channel and the medium, an associated cost; and

assigning a weight to each effectiveness rating based on an inverse relationship to the associated cost.

15. The method of claim 10 , further comprising:

determining a combination effectiveness rating for each combination of the time stamp, the medium, and the channel.

16. The method of claim 15 , further comprising:

generating a hierarchical Bayes model to identify the combination effectiveness ratings of unrepresented combinations of the time stamp, the channel, and the medium are not represented by the plurality of broadcast events.

17. The method of claim 15 , further comprising:

determining a positive or negative combinatorial effect of two or more broadcast events with substantially similar time stamps based on the combination effectiveness ratings.

18. The method of claim 10 , further comprising:

supplementing the quantifiable metric performance history data with household level set-top box data from a data aggregator.

19. The method of claim 10 , further comprising:

identifying, within the quantifiable metric performance history data, a regime change; and

dynamically determining a cutoff of the quantifiable metric performance history data based on the regime change.

20. The method of claim 19 , further comprising:

determining that a portion of the quantifiable metric performance history data before the cutoff is a baseline portion and the quantifiable metric performance history data after the cutoff is an active portion; and

generating a campaign rating based on a comparison between the active portion to the baseline portion.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 4, 2021
From: MANUS, JUSTIN; ODLUM, SEAN; BODAPATI, ANAND V.
To: BLISS POINT MEDIA, INC.
Reel/Frame 055497/0684 →
Continuity (5)
Continuation In Part 16595648 · Oct 8, 2019
Continuation 16280765 · Feb 20, 2019
Continuation 15495561 · Apr 24, 2017
Provisional Application 62441883 · Jan 3, 2017
Related Publication 20210185395A1 · Jun 17, 2021