System, method, and computer program for providing similar product recommendations for non-merchant publishers based on publisher preferences

Abstract

To provide a system, method, and computer program for automatically providing similar product recommendations for sold-out products promoted on non-merchant publisher sites.SOLUTION: A computer system with a product database provides a platform that enables non-merchant publishers to search products in the database, create call-to-actions on publisher sites for products in the database, and automatically receive similar product recommendations when end users of the publisher sites attempt to purchase promoted products that are now sold out. Once a publisher creates a call-to-action for a product that subsequently becomes sold out, similar product recommendations are made taking the publisher's preferences into account without requiring any action on the publisher's part to change the publisher's site. Publisher preferences may be learned by the system based on publisher behavior or inputted by each publisher.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates generally to electronic commerce systems, and more particularly to a system that makes similar product recommendations on behalf of non-merchant publishers based on the publishers' preferences.

[0002] [CROSS REFERENCE TO RELATED APPLICATIONS] This application claims the benefit of U.S. Provisional Patent Application No. 62 / 884,627, filed August 8, 2019, entitled "System, Method, and Computer Program for Providing Similar Product Recommendations for Non-Merchant Publishers based on Publisher Preferences," the contents of which are incorporated by reference as if set forth in their entirety herein. [Background technology]

[0003] The role of influencers and bloggers in commerce is rapidly growing. Many consumers, especially the younger generation, prefer to purchase products that have been vetted and recommended by trusted influencers. Providers of digital shopping platforms, such as "SHOPSTYLE", manage a product database and search engine that are accessible to third-party, non-merchant publishers (e.g., bloggers, influencers, etc.). Non-merchant publishers can search the product database and provide links to products in the database on their sites. When a user of the publisher's site clicks on a linked product, the user is redirected to the web page of the merchant that corresponds to that product. The user's shopping session on the merchant's site is associated with the digital shopping platform. The digital shopping platform receives a commission for sales made during that session and provides a portion of the commission to the publisher. Summary of the Invention [Problem to be solved by the invention]

[0004] The Internet "shelf life" of blog posts (or other posts by non-merchant publishers) tends to be much longer than the availability period of the products recommended in the post, so if a product mentioned in a post is sold out, it may be useful to recommend a similar product to the user. [Means for solving the problem]

[0005] The present disclosure relates to a system, method, and computer program for automatically recommending similar products when a recommended product is sold out on a non-merchant publisher's site. The recommendations of similar products are filtered with respect to the publisher's preferences. A computer system having a product database provides a platform that allows a non-merchant publisher to search for products in the database, create calls-to-actions (e.g., "buy" buttons) on the publisher's site for products in the database, and automatically receive recommendations of similar products when end users of the publisher's site attempt to purchase a recommended product that is now sold out. When a publisher creates a call-to-action for a product that will soon be sold out, similar product recommendations are made without requiring any action on the part of the publisher to change the publisher's site. The recommendations of similar products are tailored to each publisher and take into account the publisher's preferences. In some embodiments, the publisher's preferences are learned by the system based on the publisher's behavior, and in other embodiments, the publisher's preferences are input by each publisher.

[0006] In one embodiment, providing the platform includes: managing a product database having products from a plurality of vendors, each product entry in the database being associated with a product ID, an associated vendor, a link to a web page on the associated vendor's site, and an indicator of whether the associated vendor has the product in stock; providing a user interface to the product database that enables a plurality of non-merchant publishers, with each of the plurality of non-merchant publisher sites associated therewith, to search the product database and create a plurality of calls to action at each of the plurality of non-merchant publisher sites related to products in the product database; providing an application programming interface to the system that enables the system to accept call-to-action selections from end-users of a plurality of non-merchant publisher sites regarding products in a product database; the application programming interface accepting a call-to-action selection from an end user at a non-merchant publisher site for an item in the product database that is out of stock, the call-to-action selection being entered through the non-merchant publisher site and sent from a client application executing at the non-merchant publisher site; For each call-to-action selection accepted for an out-of-stock item, identifying a plurality of products in the product database that are similar to the out-of-stock product by comparing the products to at least one of product images and product metadata in the product database; filtering the identified similar products according to one or more publisher preferences of the non-merchant publisher associated with the call-to-action selection; allowing the end user who made the call-to-action selection to view information regarding one or more identified and filtered similar products; and Includes. [Brief description of the drawings]

[0007] [Figure 1A] 1 is a flow chart illustrating a method for providing similar product recommendations on a digital shopping platform accessible to multiple non-merchant publishers, according to one embodiment. [Figure 1B] 1 is a flow chart illustrating a method for providing similar product recommendations on a digital shopping platform accessible to multiple non-merchant publishers, according to one embodiment. [Diagram 2] 2 is a flow chart illustrating an embodiment of the method of FIG. 1 in which product information is displayed and a call-to-action request is initiated using a product widget. [Diagram 3] 2 is a flow chart illustrating another embodiment of the method of FIG. 1. [Figure 4] FIG. 1 is a block diagram illustrating an example of a system architecture. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] The present disclosure relates to a system, method, and computer program for automating similar product recommendations when a recommended product on a non-merchant publisher's site is sold out. The similar product recommendations are filtered with respect to the publisher's preferences. In particular, a digital shopping platform is provided that allows a third-party non-merchant publisher to search a product database having products from multiple merchants, create calls to action on the non-merchant publisher's site for products in the database, and automatically receive similar product recommendations when end users of the publisher's site attempt to purchase a recommended product in the database that is now sold out. When the publisher creates a call to action for a product that will soon be sold out, similar product recommendations are made without requiring any action on the publisher's part to change the publisher's site. The platform is accessible to multiple non-merchant publishers who recommend products from a merchant. The method is performed by a computer system (the "system"). As used herein, the term "publisher" refers to a non-merchant publisher who recommends products from a third-party merchant. "Publisher Site" means the content of a Publisher's website or mobile application.

[0009] 1A and 1B illustrate a method for providing similar product recommendations on a digital shopping platform accessible to multiple non-merchant publishers. The system manages 110 a product database having products from multiple merchants. In one embodiment, each product entry is associated with a product ID, a product title and / or description, a product image, an associated merchant, a link to a web page at the associated merchant's site, and an indicator of whether the associated merchant has the product in stock.

[0010] The system provides a publisher user interface to the product database (step 120). This user interface allows multiple non-merchant publishers, each with multiple publisher sites associated with them, to register with the system, access the product database, and create calls to action on the multiple publisher sites for products in the product database. One example of a call to action is a "Buy" button or link that redirects an end user of the non-merchant publisher's site to a merchant's product web page where the end user can purchase the product featured on the publisher's site.

[0011] The system provides an application programming interface (API) (step 130). The application programming interface enables the system to accept call-to-action selections from end users on non-merchant publisher sites for products in the product database through client applications running on the non-merchant publisher sites. End users initiate the call-to-action from the non-merchant publisher sites. The system can accept call-to-action requests from many non-merchant publisher sites. Examples of client applications that can run on the non-merchant publisher sites include web browsers and mobile applications.

[0012] When the system receives a call-to-action request to purchase a product, it determines whether the product is in stock with the associated merchant (steps 135, 140). If so, the system redirects the end user's client application (e.g., a web browser) to a page on the merchant's site for the product (step 145). The end user's shopping session at the merchant's site is associated with the system. The system typically provides the publisher with a portion of the commission received from the merchant for the product(s) purchased during that session.

[0013] When the system receives a call to action to purchase an out-of-stock item, the system identifies 150 multiple items in the product database that are similar to the out-of-stock item by comparing the product images and / or product metadata in the product database. For example, the system may search the product database for items with similar metadata or the same GTIN number (a 14-digit number used to identify items, goods, or services for sale). Visual image analysis of the product images may be used to identify similar items. The visual image analysis may be performed by the system or a third party source. In one embodiment, the system uses a combination of metadata search and visual image analysis to identify similar items.

[0014] Identifying similar products in step 150 may mean identifying similar products in real time or retrieving information about similar products that have been previously identified. In certain embodiments, the system may identify similar products prior to receiving a call-to-action request and then retrieve information about the identified similar products when the call-to-action request is received. For example, the system may track products in a database for which non-merchant publishers have created call-to-actions, identify similar products as those products become out of stock, and cache this information for immediate retrieval when a call-to-action request is received.

[0015] The system filters the identified similar products by one or more publisher preferences of the non-merchant publisher associated with the call-to-action request (step 160). In one embodiment, the publisher preferences are explicit preferences specified by the publisher when the publisher registered or updated their profile on the system. In another embodiment, the system looks up the publisher preferences based on the publisher's past behavior. For example, the system can track the merchants associated with the products in the product database linked to the publisher, the usual price range of the products linked to the publisher, the type of product, etc. The system can filter out the identified similar products that do not match the publisher's past behavior. For example, the system can filter out the identified similar products from merchants that have never been linked to the publisher or filter out products that are not in the price range of the products that the publisher normally recommends. After filtering the search results for similar products by one or more publisher preferences, the system allows the end user who made the call-to-action selection to view information about the identified and filtered one or more similar products (step 170).

[0016] In one embodiment, the system allows publishers to create calls to action for products in the product database using "product widgets." A product widget is software code that provides specific functionality in a web browser or mobile application for a product associated with the widget. For example, a product widget may display a product image, a name or short description, and a "Buy" button to purchase the displayed product, and an end user may select the "Buy" button to initiate a call to action request to the system.

[0017] In one embodiment, the system provides a user interface for a publisher. The user interface allows a publisher to create an instance of a widget and associate the instance with a product in a product database. The user interface displays one or more widgets. Each widget type can display product information separately. The user interface allows a publisher to select a widget and associate one or more products in the product database with the selected widget instance. Each instance of a widget has a unique widget ID. Each widget ID is associated with a product identifier for the product associated with the widget instance. The widget ID is also associated with the publisher ID of the publisher who created the widget instance. The user interface allows a publisher to copy software code for the widget instance. The publisher can then incorporate the code into the publisher's website or mobile application.

[0018] Figure 2 illustrates an embodiment of the method of Figure 1 in which product information is displayed and a call-to-action request is initiated using a product widget. When an end user loads a publisher's site in a browser or opens a publisher's mobile application, a product widget in the publisher's site is initialized and provides its widget ID to the system (steps 210, 220). As previously mentioned, each widget ID is associated with one or more product IDs, and the system identifies the product IDs associated with the widget ID (step 230).

[0019] The system retrieves product metadata associated with each identified product ID (step 240). The metadata may include an image of the product, the product brand, the product name or short description, the merchant selling the product, the price, and current availability (i.e., in stock). The metadata may be obtained from a cached product index or a product database. The system provides some or all of the product metadata to the requesting widget (step 250).

[0020] The widget uses the product metadata to display product information on the publisher's site (step 260). For example, the widget may display a product image, name, and price for each product associated with the widget. The widget may also display a call-to-action, such as a "Buy" button, for each product associated with the widget. The widget may display the product information and call-to-action directly on the publisher's site or in a frame or window on the site.

[0021] When the user selects a call-to-action to purchase the product, the widget notifies the system of the call-to-action selection along with the associated product ID (step 270). The system uses the product ID to look up availability information in a product database or cached product index. If the product is in stock, the server redirects the user's browser to the retailer web page associated with the product ID (step 280). If the product is out of stock, the system notifies the widget and provides it with metadata about similar products in the product database filtered for the publisher's preferences (step 280). As previously discussed, similar products can be identified by searching product metadata in the system and / or by performing visual analysis to identify products that are visually similar.

[0022] When the widget is notified that the product is out of stock, it indicates that the product is out of stock on the publisher's site (step 290). For example, the widget may display a notification that the product is out of stock and a call-to-action that allows the user to view similar products. If the user clicks on the call-to-action to view similar products, the widget displays a window, frame, or modal window (e.g., an iframe) with the similar products. Each similar product is displayed with a call-to-action to purchase the product. If the user clicks on the call-to-action for a similar product, the system repeats step 280 and, if applicable, step 290 for the similar products.

[0023] In certain embodiments, the system notifies the widget of any out-of-stock items associated with the widget when the widget is initialized, in which case the first call-to-action displayed by the widget on the publisher's site may be a call-to-action to view similar items.

[0024] FIG. 3 illustrates another embodiment of the method illustrated in FIG. 1. In this embodiment, the system allows a user to create a link to a product in the product database and embed the link in the form of a call-to-action (e.g., a "buy" button) directly on the publisher's site. When the user clicks on the call-to-action to purchase a product on the publisher's site, the user's client application (e.g., a web browser or a mobile app) notifies the system of the call-to-action and provides the relevant product ID (steps 310, 320, and 330). The system then uses the product ID to determine whether the product is in stock by looking up product information in a cached product index or product database (step 340). If the product is in stock with the merchant associated with the product ID, the server retrieves the appropriate merchant's web page address for the product (step 350). If the product is out of stock, the system creates a web page with similar products in the product database filtered for the publisher's preferences (step 350). The system then redirects the client application to the merchant's web page or a page of similar products depending on whether the product is in stock (step 360). The client application displays the appropriate merchant web page or similar product web page (step 370). If the user selects a product on the similar product web page that is also out of stock, steps 350-370 are repeated for the selected similar product. The similar products can also be filtered for purchase likelihood in addition to the publisher's preferences.

[0025] An example of the system configuration is shown in Figure 4. The system includes a product database 470 and a server 410 that executes a number of software modules. The software modules include the following: · A Publisher UI module 460 that creates a user interface that allows publishers to register with the system and search the product database. · A widget UI module 435 that creates a user interface that allows publishers to create widgets or calls to action for products in the product database 470. · A database interface module 440 that searches, retrieves, and stores information in a product database 470. · An API 455 that receives and processes call-to-action requests from client applications that are viewing the publisher's site. · Similar product identification module 445 for identifying products similar to the out-of-stock product. Module 445 can interface with third party systems, such as systems that identify similar products based on image analysis. A publisher preference filter module 450 that filters similar products based on learned or explicit publisher preferences.

[0026] As described above, server 410 interfaces with publisher sites 420 and merchant websites 430 to provide product information to end users, process call-to-action requests, and redirect customers to merchant sites.

[0027] The methods described herein may be implemented in systems configured in other manners and are not limited to the system configuration shown in FIG.

[0028] The methods described herein are embodied in software and performed by a computer system (comprising one or more computing devices) executing the software. Those skilled in the art will appreciate that a computer system has one or more memory units, disks, or other physical computer-readable storage media for storing software instructions, and one or more processors for executing the software instructions.

[0029] Those skilled in the art will recognize that the present invention can be embodied in other specific forms without departing from the spirit or essential characteristics thereof. Accordingly, the above disclosure is intended to be illustrative, but not limiting, of the scope of the invention, which is set forth in the following claims.

Claims

1. 1. A method for enabling similar product recommendations at a plurality of non-merchant publisher sites that recommend products of a merchant, the method being executed by a computer system, the method comprising: The non-merchant publisher is not affiliated with the merchant, and the recommendation of a similar product is made in relation to a sold-out product; managing a product database having products from a plurality of vendors, each product entry in the database being associated with a product ID, an associated vendor, a link to a web page at the associated vendor's site, and an indicator of whether the product is in stock at the associated vendor; providing a user interface to the product database that enables a plurality of non-merchant publishers, with each of the plurality of non-merchant publisher sites associated therewith, to search the product database and to create calls to action at the plurality of non-merchant publisher sites related to products in the product database; providing an application programming interface to said system that enables said system to accept call-to-action selections from end-users of said plurality of non-merchant publisher sites regarding products in said product database; the application programming interface accepting call-to-action selections from end users of the plurality of non-merchant publisher sites for out-of-stock products in the product database, the call-to-action selections being entered through the non-merchant publisher sites and sent from client applications executing on the non-merchant publisher sites; For each of said call-to-action selections received for an out-of-stock item, identifying a plurality of products in the product database that are similar to the out-of-stock product by comparing product images and / or product metadata in the product database; filtering the identified similar products according to one or more publisher preferences of the non-merchant publisher associated with the selection of the call to action; enabling the end user who selected the call to action to view information regarding one or more identified and filtered similar products; and The method includes:

2. The method of claim 1 , wherein the publisher preferences regarding non-seller publishers are determined by the system based on past behavior of the non-seller publishers.

3. The method of claim 1 , wherein the publisher preferences with respect to non-seller publishers are based on explicit preferences provided to the system by the non-seller publishers.

4. 2. The method of claim 1, wherein the one or more publisher preferences include a category or group of merchants associated with the non-merchant publisher site from which the call-to-action request was made, and the system filters out similar products from merchants not included in the category or group of merchants.

5. 2. The method of claim 1, wherein providing the user interface that enables the non-merchant publisher to create a call-to-action includes providing a user interface that enables the non-merchant publisher to create a product widget, associate the product widget with one or more products in the product database, and embed the product widget on the non-merchant publisher's site.

6. A non-transitory computer readable medium having a computer program that, when executed by a computer system, causes the computer system to perform a method for enabling similar product recommendations at a plurality of non-merchant publisher sites that promote products of a merchant, the method comprising: The non-merchant publisher is not affiliated with the merchant, and the recommendation of a similar product is made in relation to a sold-out product; The method comprises: managing a product database having products from a plurality of vendors, each product entry in the database being associated with a product ID, an associated vendor, a link to a web page at the associated vendor's site, and an indicator of whether the product is in stock at the associated vendor; providing a user interface to the product database that enables a plurality of non-merchant publishers, with each of the plurality of non-merchant publisher sites associated therewith, to search the product database and to create calls to action at the plurality of non-merchant publisher sites related to products in the product database; providing an application programming interface to said system that enables said system to accept call-to-action selections from end-users of said plurality of non-merchant publisher sites regarding products in said product database; the application programming interface accepting call-to-action selections from end users of the plurality of non-merchant publisher sites for out-of-stock products in the product database, the call-to-action selections being entered through the non-merchant publisher sites and sent from client applications executing on the non-merchant publisher sites; For each of said call-to-action selections received for an out-of-stock item, identifying a plurality of products in the product database that are similar to the out-of-stock product by comparing product images and / or product metadata in the product database; filtering the identified similar products according to one or more publisher preferences of the non-merchant publisher associated with the selection of the call to action; enabling the end user who selected the call to action to view information regarding one or more identified and filtered similar products; and A non-transitory computer readable medium comprising:

7. The non-transitory computer-readable medium of claim 6 , wherein the publisher preferences regarding non-seller publishers are determined by the system based on past behavior of the non-seller publishers.

8. 7. The non-transitory computer-readable medium of claim 6, wherein the publisher preferences for non-seller publishers are based on explicit preferences provided to the system by the non-seller publishers.

9. 7. The non-transitory computer-readable medium of claim 6, wherein the one or more publisher preferences include a category or group of merchants associated with the non-merchant publisher site from which the call-to-action request was made, and the system filters out similar products from merchants not included in the category or group of merchants.

10. 7. The non-transitory computer-readable medium of claim 6, wherein providing the user interface that enables the non-merchant publisher to create a call-to-action includes providing a user interface that enables the non-merchant publisher to create a product widget, associate the product widget with one or more products in the product database, and embed the product widget on the non-merchant publisher's site.

11. 1. A computer system for enabling similar product recommendations at a plurality of non-merchant publisher sites that recommend products of a merchant, comprising: The non-merchant publisher is not affiliated with the merchant, and the recommendation of a similar product is made in relation to a sold-out product; one or more processors; one or more memory units coupled to the one or more processors and configured to store instructions that, when executed by the one or more processors, cause the system to perform operations; Equipped with The process comprises: managing a product database having products from a plurality of vendors, each product entry in the database being associated with a product ID, an associated vendor, a link to a web page at the associated vendor's site, and an indicator of whether the product is in stock at the associated vendor; providing a user interface to the product database that enables a plurality of non-merchant publishers, with each of the plurality of non-merchant publisher sites associated therewith, to search the product database and to create calls to action at the plurality of non-merchant publisher sites related to products in the product database; providing an application programming interface to said system that enables said system to accept call-to-action selections from end-users of said plurality of non-merchant publisher sites regarding products in said product database; the application programming interface accepting call-to-action selections from end users of the plurality of non-merchant publisher sites for out-of-stock products in the product database, the call-to-action selections being entered through the non-merchant publisher sites and sent from client applications executing on the non-merchant publisher sites; For each of said call-to-action selections received for an out-of-stock item, identifying a plurality of products in the product database that are similar to the out-of-stock product by comparing product images and / or product metadata in the product database; filtering the identified similar products according to one or more publisher preferences of the non-merchant publisher associated with the selection of the call to action; enabling the end user who selected the call to action to view information regarding one or more identified and filtered similar products; and 2. A computer system comprising:

12. 12. The system of claim 11, wherein the publisher preferences with respect to non-seller publishers are determined by the system based on past behavior of the non-seller publishers.

13. 12. The system of claim 11, wherein the publisher preferences with respect to non-seller publishers are based on explicit preferences provided to the system by the non-seller publishers.

14. 12. The system of claim 11, wherein the one or more publisher preferences include a category or group of merchants associated with the non-merchant publisher site from which the call-to-action request was made, and the system filters out similar products from merchants not included in the category or group of merchants.

15. 12. The system of claim 11, wherein providing the user interface that enables the non-merchant publisher to create a call to action includes providing a user interface that enables the non-merchant publisher to create a product widget, associate the product widget with one or more products in the product database, and embed the product widget on the non-merchant publisher's site.

Images