Predictive Shipping Method for E-Commerce Site
A delivery mechanism uses a recommender and prediction system to determine if a customer may want an item before he/she specifically requests such. Physical inventory ordered from an e-commerce site is thus preemptively distributed and maintained in transit.
Latest Media Queue, LLC Patents:
The present application claims priority to and is a continuation of both Ser. No. 10/874,412 (now U.S. Pat. No. 8,738,541) and Ser. No. 11/369,660, both of which in turn claim the benefit under 35 U.S.C. 119(e) of the priority date of Provisional Application Ser. No. 60/482,371 filed Jun. 25, 2003, and all of which are hereby incorporated by reference.
FIELD OF THE INVENTIONThe present invention relates to electronic commerce systems and methods of providing selections, shipments and exchanges of rental items. The invention has particular applicability to applications in which entertainment media are rented on a temporary or short term basis by subscribers over the Internet.
BACKGROUNDThe Internet is now being used for a number of commercial purposes, including purchase and rental of movie films in different media formats. One such popular website is maintained by Netflix®, where subscribers can search, review and select movie titles (in DVD media format). If a particular title is available, the subscriber's choice is then placed into a rental selection “queue.” During an interactive online session, a subscriber can select a number of titles, and then prioritize them in a desired order for shipment within the selection queue.
After the movie title selection session is over, the system proceeds to ship the desired titles in the order requested by the subscriber. The selections are mailed to the user in special packaging, which include return mailers pre-printed and adapted for such movie media. This minimizes the inconvenience to the subscriber, but there is an indeterminate processing and mailing delay therefore associated with each selection. In most instances, a selection is mailed the same day, but even in such cases, there is typically a three day transaction period associated between the selection, processing, shipment, mail transit and receipt of a particular title.
After shipment, these titles then appear in a list identified essentially as items that are outstanding (i.e., movies that have not yet been returned by the user) within a “titles out” queue. Under one option of the Netflix terms of service, the system places a limit on the number of outstanding titles that a subscriber may have at any one time (typically, a function of the level of service agreement, with more $$/month resulting in more titles). Thus, if the user selects more movies than they can acquire at one time, the remaining movie titles remain as un-shipped items in the rental selection queue. Additional titles in the rental selection queue are only shipped to the user after the system logs a returned item from that same user. This happens automatically, so the user does not need to return to the Netflix website to request the shipment. At the same time, however, the user is not consulted in the decision to ship the next item; it happens without his/her approval or advance notice.
The aforementioned system is further described in U.S. Pat. No. 6,584,450, which is hereby incorporated by reference. The '450 patent explains the above policy in terms of a “MAX” out approach for servicing rental customers. Notably, the aforementioned disclosure also mentions that the MAX out approach can only be altered under certain exceptions, such as where the customer agrees to pay a surcharge.
A limitation of the aforementioned system is that while the system automatically ships the next items in the rental queue it fails to consider whether the user may in fact want to “veto” the next item in the queue, either because their tastes have changed, or if they might have already seen the title. As users sometimes fill up their request queues months in advance with dozens of titles, and they typically only see 6-8 movies per month, there is a strong possibility that one more of such titles will be stale after several months.
A further limitation, of course, is that the user must always return another title before the next title in their queue is shipped to them. While this is somewhat useful for inventory tracking purposes, the user is handicapped by the fact that after they posit the returned title in the mail, they have little or no control over how quickly or accurately such return will be processed by the mail, or the rental service.
Finally, it is conceivable that on many occasions, from the perspective of the service provider, it will be desirable to ship another title to a customer, regardless of a status of their delivery queue, and without applying a surcharge. In other words, in order to promote certain titles, retain certain customers, or balance supply/demand, a service provider may want to automatically ship more than the “Max Out” figure to a particular customer, based on some triggering event.
SUMMARY OF THE INVENTIONAn object of the present invention, therefore, is to overcome the aforementioned limitations of the prior art.
Another object is to implement a method of delivering rentals which allows users to be part of the process of determining which titles they should receive;
A further object is to implement a method of delivering rentals which is based on an automatic, time based delivery for rental titles, and not on a returned/received status of other rental titles in the customer's possession;
Yet another object is to implement a method of delivering rentals which is based on automatically selecting titles of interest for the customer using a recommendation system, and using such criteria, rather than user specified criteria, for automatically shipping items to a user.
A first aspect of the invention, therefore, concerns a method of delivering rented media to a subscriber comprising the steps of: providing a maximum number of media items (Nmax) which the subscriber can have out at any moment in time; delivering a media item to the subscriber in accordance with a priority specified by the subscriber; incrementing a count of media items out (Nout); determining if Nout>=Nmax; when Nout>=Nmax, setting a timer to a predetermined delay; delivering another media item to the subscriber after expiration of said predetermined delay, even if Nout>=Nmax. In this fashion, a number of media items in excess of Nmax can be in the subscriber's possession and/or in transit to/from the subscriber.
Thus, in preferred embodiments, distributions of media items to subscribers are provided at least partially on a time basis, and not on a subscriber capacity basis. This allows for more flexibility in reducing returns monitoring. When Nout=Nmax+1, returns from the subscriber are then monitored to identify if Nout should be decremented. Typically the media items are DVDs or other entertainment content identified in a delivery queue by the subscriber, ordered in a sequence of desired receipt by the subscriber.
In some embodiments a delinquency period for the subscriber is monitored, which delinquency period is a period of time in which Nout>=Nmax for the subscriber. Depending on how tardy a particular subscriber is, automatic timed deliveries can be suspended. Further still, in some instances, the predetermined delay can be function of how many titles are out, when a prior title was delivered to the subscriber, or demand for the media item.
Another aspect of the invention covers a method of delivering rented media to a subscriber comprising the steps of: providing a maximum number of media items (Nmax) which the subscriber can have out at any moment in time; delivering a media item to the subscriber in accordance with a priority specified by the subscriber; incrementing a count of media items out (Nout); determining if Nout>=Nmax; determining if an overcapacity triggering event has occurred;
delivering another media item to the subscriber after expiration of said predetermined delay, even if Nout>=Nmax. Again, a number of media items in excess of Nmax can be in the subscriber's possession and/or in transit to/from the subscriber.
In typical environments, the overcapacity triggering event can be one or more of the following: a promotion for the media item, a demand associated for the media item, a timer associated with the subscriber, a recommender system prediction given for the subscriber, or a status of a delivery queue maintained for the subscriber.
Still another aspect of the invention concerns a method of delivering rented media to a subscriber comprising the steps of: providing a maximum number of media items (Nmax) which the subscriber can have out at any moment in time; delivering a first media item to the subscriber; monitoring returns from the subscriber and a delivery capacity for the subscriber to identify if a second media item should be sent to the subscriber; selecting at least two candidate media items by choosing a first title identified by the subscriber and a second title automatically selected for the subscriber based on preferences of the subscriber; presenting said at least two candidate media items to the subscriber to permit such subscriber to select which one should be delivered next in sequence. Thus, the subscriber is consulted before said second media item is sent.
In preferred embodiments, the media are rented over the Internet, and the subscriber is consulted by email. To assist the subscriber in identifying suitable candidate items, the subscriber is directed to a website to select from additional candidate media items. The second titles are preferably selected by a recommender system operating at an Internet site.
It will be understood from the Detailed Description that the inventions can be implemented in a multitude of different embodiments. Furthermore, it will be readily appreciated by skilled artisans that such different embodiments will likely include only one or more of the aforementioned objects of the present inventions. Thus, the absence of one or more of such characteristics in any particular embodiment should not be construed as limiting the scope of the present inventions. While described in the context of a rental system, it will be apparent to those skilled in the art that the present teachings could be used in any Internet based rental or purchase system that employs a queue of some form.
While the present preferred embodiment is directed to a process to be used by a rental service for distributing DVD movie titles, it will be understood by those skilled in the art that, as used herein, the term “media item” or “playable media item” is intended to have its broadest meaning to include any machine readable media readable (including software programs) by a subscriber machine player (such as a DVD player, a computer, a video recorder/receiver, an audio CD ROM Player, and the like). In addition, electronic files which are downloaded and/or streamed from the Internet or some other distribution point (Satellite, Cable) are also encompassed in this definition. Furthermore, other queable content/items that are not machine readable could be distributed in this fashion as well, including human readable materials such as printed matter in the form of books, magazines, and other general purpose items. Other useful articles can be rented and distributed in accordance with the present teachings as well.
As seen in
At step 120, the system begins shipping titles to the customer, if a slot is available in his/her shipping queue. Assume that K represents the net sum of titles that are out to the customer, accounting for returns, that the user has a free slot in their delivery queue, and that they have selected a title that is otherwise available for shipping. If K<Nmax, the title is shipped, and afterwards, the system increments the counter of titles out to the customer, to reflect that K=K+1.
From there, at step 130, a determination is made if K>=Nmax. According to the prior art above, no further titles would be sent to the customer, unless the customer agrees to a surcharge, or until he/she returns a title. In the present invention, however, an additional title is automatically sent out after a fixed time delay set at step 135, or in accordance with other triggering events identified below. It will be noted, of course, that this time delay can be designed in accordance with the needs of an inventory management system, and in a manner that does not result in an excessive number of titles being shipped unnecessarily. For example, the delay could be set to 1 day, 2 days, or even hours.
The important distinction is that no monitoring need be done for returns by the customer between steps 130 and 135. In other words, the present invention automatically ships another title, even past the “max out” limit, based on a time event, and not a customer originated event (i.e., such as a return of a title, or a receipt of a title).
After step 135, a check can be made, if desired for returns made by the customer. If so, a slot is opened in the subscriber's shipping queue, K is decremented accordingly, and the process continues again at step 120.
At step 140, a check is made to see if in fact the number of titles out the customer K=Nmax+1. If not, a second delay is again introduced at step 145; it will be appreciated that this second delay could be the same or different as the first delay above. The process then proceeds to step 120 as noted earlier.
If at step 140 K is indeed=Nmax+1, the system then moves into a monitoring mode at step 150, wherein the shipping queue is examined for the customer's account to see if there is an available slot. If a return is made, a slot is opened, so that once K<Nmax, another title is shipped in response to an expiration of a third timer delay at step 165.
It will be noted that the above approach, of course, varies from the prior art, in that it is possible, of course, to ship an additional title beyond that contractually requested to the subscriber. It is possible, as well, that the subscriber may “sit” on the titles. Nonetheless, given that the aforementioned rental systems operate by incentivizing customers to exchange inventory, it is unlikely that a subscriber will be motivated to simply sit on an extra title that they did not bargain for. In other words, customers are aware that they cannot get new titles without exchanging older titles, so they are not going to handicap their own potential monthly “turn” of titles by sitting on some set number of titles that is in excess of their limit.
It is also perhaps more efficient to separate the returns and shipping processes, so that the programming complexity and monitoring is reduced.
Furthermore, in many cases, of course, it is possible that a title will be shipped out in the morning, while a return is processed from the user later that day. In this respect, the increase in “turns” is rather minimal, since the exchange would occur on that day anyway. Even in cases where the expected “turn” is accelerated by a day or two, there may be other commercial advantages in preemptively shipping a title in an automated fashion according to a fixed schedule.
The primary point, as noted above, is to de-couple the events of a rental return and a rental shipment, and to make the deliveries/shipments of titles caused by a time based event, rather than on monitoring and logging an additional event. The time based event, as well as the amount of the delay can be tailored on an application by application basis, depending on the characteristics of the subscribers, the type of inventory, etc. If a subscriber returns a title before the delay has expired (i.e., the slot opens up, it is not necessary to account for the return anyway, and the present invention presents no worse penalty than the prior art.
In another variation, an additional check can be made at step 140, to see if the subscriber is someone who makes a habit of retaining more than their allotted number of titles. If such person has had more than their allotment during the past N days (where N can be a billing cycle) the system may “veto” the extra shipment, and simply fall through to step 150. Alternatively, the delay for such person could be set longer.
In yet another variant, the time delay could be set to be a function of the number of the title, so that there is only a small delay in shipping the second title, and a larger delay in shipping the last or extra title. Another approach, of course, is to tie the delay for the Nmax+1 title to some fixed time after the shipment of the earliest one shipped to the customer, again, on the expectation that the user is very likely to return the earliest title within such time frame.
In some cases, of course, it may be desirable to only send low-demand titles, or promotional titles as the Nmax+1st title. These may not have been specifically requested by the customer. The advantage of such titles is that they can supplant a subscriber's request for a potentially more valuable or popular title.
As noted herein, such return/shipping process is expected to be embodied in a variety of software programs, routines, etc., that run on one more servers coupled to the Internet. In other embodiments these programs and routines will run directly on a computing machine at the subscriber's premises, such as a digital video recorder, receiver, etc.
As with step 110 above, step 210 collects customer item selection criteria.
As with step 120 above, step 220 ships a new title if a slot is available.
At step 130, an additional title (beyond the customer's maximum allotment Nmax) is shipped, based on a “triggering event.” The triggering event may be an expiration for a timer, a promotion for a particular title, a prediction that the customer will desire the title, and/or a determination made based on observing a status of a customer's queue.
In the promotional case, the title may be a low demand title, a promotional title, etc., which “costs” little to the service provider, and helps to increase the subscriber satisfaction.
In the last case, for example, the system may note that the customer has selected a number of titles which have long waits. Rather than the user seeing nothing for long periods of time (because no titles will be delivered if they are unavailable), the system can, using its own criteria (separate from the user collected criteria) select a title and automatically ship it to the customer.
Again, this is preferably done without a surcharge. The user's shipping queue is thus expanded (albeit temporarily) at step 240.
While the Netflix system is known to “randomly” send additional titles to customers, in excess of their maximum allotment, the applicant does not believe that such system operates based on a promotional event, or on monitoring the customer's queue. The advantage of this approach, as well, is that the system can “throttle” and control demand for new releases. That is, the system can preemptively send titles to a subscriber that are not as high in demand as the titles already in a subscriber's queue. Again, the cost in sending another title may be less than the cost of securing a very popular title, or not satisfying another customer with such title.
Furthermore, an ancillary benefit to the service provider is that they can determine, from available inventory, particular titles which may be of interest to subscribers, and ship them as a way of “filling” subscriber needs in advance. This can be done, again, by auto-recommender systems, such as by the Cinematch system employed by Netflix, or other comparable system. In other words, every title preemptively shipped by the service provider acts to supplant a subscriber choice that may be harder to satisfy (i.e., because the desired title is scarce). For example, if the service determines that a customer is highly likely to enjoy title A, and if the title is shipped to the subscriber, this fills an available spot in the shipping queue and reduces the expectation/opportunity by the subscriber of receiving another (potentially more popular) title. This further frees up an opportunity for the service provider to ship the more popular title to another subscriber, thus giving more flexibility in responding to demand. Furthermore, a service provider can determine which titles have a higher profit margin (i.e., lower sharing expenses with distributors, studios, etc.) and control the distribution of titles more closely using cost/profit as an additional factor as well.
As with step 110 above, step 310 collects customer item selection criteria.
At step 320 a determination is made to see if there is an available shipping slot in the subscriber's queue. Assuming there is, at step 330, the system then determines an appropriate title that should be moved into the customer's shipping queue and send to the subscriber.
Unlike the prior art, however, the title is not shipped automatically, simply because the customer has returned another title. Instead, at step 340, a notification is made to the subscriber, by email or otherwise, alerting them to the selected choice that will be sent to them next. In some cases, of course, the selected choice may simply be the next item in the customer's queue, in the order they presented it. In other instances, however, the system can automatically select a new title, and present it to the subscriber for acceptance. Thus, in the latter case, the “selection criteria” are provided by the rental provider system, and not by the subscriber. In some instances, both a user-specified title and a rental provider specified title can be identified, so that the user can make a choice between the two. The user can elect, through the selection criteria entered at step 310, to what extent they wish to have the system make such recommendations.
The notification/confirmation request provided at step 340 is then ratified at step 350. The ratification by the subscriber can take a variety of forms, some of them explicit (i.e., responding “yes” to an email) or implicit (i.e., not providing an objection to the choice within a designated time period). In some cases, a subscriber may set up such notification controls in advance, so that the system always makes a selection on their behalf without further prompting. Other forms will be apparent to those skilled in the art. The distinction, again, is that the system incorporates the subscriber's desires in the shipping process, and does not mechanically decide to ship the next title without first consulting with the customer.
Thus, an acceptance at step 350 may be an actual or “constructive” confirmation. An actual confirmation would require, for example, that the subscriber either visit the site to remove a tentative designation, or confirm the shipment in a later e-mail, etc. A constructive confirmation can be based on time, for example, so that the system will automatically ship the title unless the subscriber objects to such selection within a predetermined (controlled by the user) number of days/hours.
In another instance, the subscriber can be presented with a set of alternatives, or be directed to the website to select another option different from the default choice that would otherwise be sent. This, again, can be used to increase traffic to a provider site, by exposing the subscribers to new materials that they may be unaware of since their last visit.
This approach also has an advantage in that the service provider can also intervene, and automatically recommend a title to be shipped to the subscriber, in lieu of the choice (i.e., the next in the queue) made by the subscriber. A variety of artificial intelligence, or preference determination systems are known in the art for predicting preferences of individuals. These systems work based on such simple factors as demographics, prior selections, expressed preferences, questionnaires, etc.
In the present system, an auto-recommend feature can be based, for example, on prior ratings given by a subscriber to certain movies which he/she has seen, and/or just based on the identity if all prior rentals. In this instance, the system uses prior ratings by default, as these tend to more accurately reflect a particular individual's tastes. However, the invention is by no means limited to any particular mechanism, and it will be understood that any suitable system can be employed for this purpose. The only criterion, of course, is that recommender should work in the background, seamlessly and without extensive burdensome participation by the subscriber so that the queue is automatically replenished.
The above process, notification and acceptance procedures, etc., can all be easily implemented by skilled artisans in a hardware/software form suitable for execution on an online service provider system.
Again, using the present invention, the system is allowed to use a customer acceptance, and/or its own selection criteria, rather than a receiving event, to control shipping of titles. This provides additional advantages, including the fact that titles selected and shipped by the service provider in this fashion act to supplant a subscriber choice that may be harder to satisfy (i.e., because the desired title is scarce). For example, if the service determines that a customer is highly likely to enjoy title A, if the subscriber agrees to select title A, and if the title is shipped to the subscriber, this fills an available spot in the shipping queue and reduces the expectation/opportunity by the subscriber of receiving another (potentially more popular) title. This further frees up an opportunity for the service provider to ship the more popular title to another subscriber, thus giving more flexibility in responding to demand. Furthermore, a service provider can determine which titles have a higher profit margin (i.e., lower sharing expenses with distributors, studios, etc.) and control the distribution of titles more closely using cost/profit as an additional factor as well.
Finally, it should be noted that this type of proactive inventory management helps to expand the reach and desirability of the system to other types of would-be renters, including persons who are too busy, lazy, or unable to determine appropriate choices on their own. Moreover, since some renters are impulsive, they can be presented automatically with title choices to increase their participation rates as well.
It will be understood by those skilled in the art that the above are merely examples of processes for a media distribution system, and that countless variations on the above can be implemented in accordance with the present teachings. A number of other conventional steps that would be included in a commercial application have been omitted, as well, to better emphasize the present teachings.
Furthermore, it will be apparent to those skilled in the art that the processes of the present invention, including those illustrated in
Claims
1. An automated method of delivering items to a customer of an e-commerce website comprising the steps of:
- delivering a first item to the customer in response to a specific request by the customer for such first item to a website computing system;
- wherein said first item is an article shipped by the e-commerce website to the first customer;
- generating a prediction with a recommender system for a second item which is likely to be of interest to said customer, said prediction being based at least in part on said customer's prior item selections; and
- distributing said second item to the first customer based on said prediction and prior to an explicit request by the first customer for said second item;
- wherein delivery of said second item from inventory of the e-commerce site is initiated in advance of said first customer requesting such item from the e-commerce website.
2. The method of claim 1, wherein said items are identified in a particular order in a delivery queue by the subscriber.
3. The method of claim 1, wherein said items are machine playable media.
4. The method of claim 1 wherein said second item is correlated by the recommender system to said first item.
5. The method of claim 1 wherein said items are mailed to the first customers.
6. The method of claim 1 wherein said second item is selected based on a profit margin associate with said item, such that high profit margin items are more likely to be shipped in advance of an express request.
7. The method of claim 1 wherein said second item is a promotional item selected by the e-commerce site computing system based on a promotional event and which is shipped to a plurality of customers in advance of such customers requesting such item.
8. The method of claim 1 wherein said second item is selected based on a demand for such item by a plurality of customers of the e-commerce site.
9. The method of claim 1 wherein said second item is selected based on monitoring an availability of status of items previously selected by the first customer and maintained in a customer selection list of items by the e-commerce website.
10. The method of claim 1 wherein inventory of the e-commerce site is managed by controlling shipments of said second item to a plurality of customers based on an automated timing schedule and not in response to requests of said second item from such customers.
11. The method of claim 1 wherein said second item is provided at no charge to said first customer.
12. The method of claim 1 wherein said second item is selected based on a determination by said website computing system that it will increase satisfaction of said first customer with said e-commerce site.
13. The method of claim 1 wherein said second item is selected based on a determination by said website computing system of a cost of shipping such item and whether said second customer is likely to enjoy such second item.
14. The method of claim 1 wherein processing and mailing delays associated with customer selections at the e-commerce site are reduced by customer prediction based shipping.
15. The method of claim 1 further including steps: receiving a request for a third item from said first customer which is more popular than said second item; shipping said second item to said first customer and said third item to a second customer instead of said first customer based on both customer satisfaction and cost factors.
16. The method of claim 1 wherein said distributing of said second item to the first customer is performed at least in part using physical mail delivery.
17. The method of claim 16 wherein said second item is packaged with a mailer that includes a pre-printed physical return mailer.
18. The method of claim 1 wherein the e-commerce website distributes a plurality of items to customers based on a plurality of predictions prior to explicit requests, such that a portion of inventory of such e-commerce is maintained in transit.
19. The method of claim 1 wherein said second item is selected and distributed to said first customer based on a time based event determined for such customer and not an explicit request by such customer.
20. An automated method of delivering items to a customer of an e-commerce website comprising the steps of:
- maintaining a current request list of a set of items selected by a first customer at the e-commerce website;
- delivering a first item to the customer from said set of items in response to a specific request by the customer for such first item to a website computing system;
- wherein said first item is an article shipped by the e-commerce website to the first customer and is associated with a charge paid by the first customer;
- generating a prediction with a recommender system for a second item which is likely to be of interest to said first customer, said prediction being based at least in part on said first customer's prior item selections and said current request list maintained for said first customer; and
- shipping said second item to the first customer based on said prediction and prior to an explicit request by the first customer for said second item to be shipped;
- wherein delivery of said second item from inventory of the e-commerce site is initiated in advance of said first customer requesting such item to be shipped from the e-commerce website; and
- further wherein said second item is shipped and delivered without any charge to said first customer as a promotional item.
Type: Application
Filed: May 20, 2014
Publication Date: Nov 20, 2014
Applicant: Media Queue, LLC (Tulsa, OK)
Inventor: John Nicholas Gross (Berkeley, CA)
Application Number: 14/282,477
International Classification: G06Q 10/08 (20060101); G06Q 30/06 (20060101);
