	US 12,437,316 B1
	Multi-channel attribution based on timing and number of exposures relative to conversion events
Susmita Ghose, Mountain View, CA (US); and Ashish Ramgopal Baldua, San Jose, CA (US)
Assigned to ALPHONSO INC., Mountain View, CA (US)
Filed by Alphonso Inc., Mountain View, CA (US)
Filed on Jan. 29, 2020, as Appl. No. 16/775,688.
Claims priority of provisional application 62/908,944, filed on Oct. 1, 2019.
Int. Cl. G06Q 30/00 (2023.01); G06Q 30/0242 (2023.01); G06Q 30/0272 (2023.01)

CPC G06Q 30/0243 (2013.01) [G06Q 30/0246 (2013.01); G06Q 30/0272 (2013.01)]

6 Claims

1. An automated method for performing multi-channel attribution for conversion events associated with a plurality of consumers, each consumer having an associated consumer identifier, wherein each conversion event is associated with a brand or product that has corresponding media advertising electronically delivered on a plurality of different media-based delivery channels to the plurality of consumers via a plurality of media devices associated with respective consumers, the method comprising:

(a) detecting a plurality of conversion events for the brand or product, and identifying a consumer identifier and conversion time associated with each of the conversion events, wherein the consumer identifier associated with each of the conversion events is identified using one or more device graphs;

(b) electronically detecting media advertising exposure by the plurality of media devices for each of the conversion events by:

(i) monitoring the consumers' media devices using electronic monitoring devices, and

(ii) performing automatic content recognition (ACR) on media content played by the media devices using an ACR server, wherein the ACR server uses one or more of digital signature and digital fingerprint comparisons to identify the media content played by the media devices,

and for each media advertising exposure, identifying:

(i) the respective consumer identifier associated with the media advertising exposure,

(ii) an exposure time relative to the time of the conversion event, and

(iii) the delivery channel of the media advertising exposure; and

(i) creating a recency histogram of the exposure times without regard to which delivery channel the media advertising exposure occurred, wherein the recency histogram includes bins and a height for each bin, the bins representing different exposure times relative to the time of the conversion event, and the height representing a count of the number of media advertising exposures that occurred at each of the different exposure times relative to the time of the conversion events, and wherein the exposure times are for the plurality of consumers, and wherein all media advertising exposures are counted when creating the recency histogram, regardless of whether any specific ad exposure leads to a conversion,

(ii) normalizing the recency histogram to a standard probability distribution, and thereby deriving weights for each of the exposure times,

(iii) assigning the weights to each media advertising exposure,

(iv) calculating for each consumer identifier an attribution per delivery channel by normalizing the sum of the weights to equal one, and

(v) calculating an overall attribution for the conversion events associated with the plurality of consumers for each of the delivery channels by averaging the attributions calculated for each consumer identifier.