Optimizing digital video distribution
Systems, methods, and apparatus include computer programs encoded on a computer-readable storage medium, including a system for ranking videos. Videos are identified that have been presented at client devices. For each video, session start data is identified that specifies a lead video that initiated presentation to a user during a presentation session. For each lead video, presentation times over multiple user sessions are determined, a scaled presentation time is obtained, user sessions for which the lead video initiated presentation of videos are identified, and an aggregate video presentation time attributable to the lead video is determined. For each given video, a presentation score is determined based on a scaled presentation time of the lead video relative to a sum of the aggregate video presentation times for the lead videos. The videos are ranked based on the presentation scores. A user interface is updated to present the ranked videos.
1. A system comprising:
one or more processors; and
one or more memory devices including instructions that, when executed, cause the one or more processors to perform operations comprising:
identifying, for a given video, session start data specifying that the given video was a lead video that initiated video presentation to one or more users during multiple user sessions;
for each of at least some user sessions among the multiple user sessions:
classifying the lead video as one of an in-service initiated video presentation or a remotely initiated video presentation; and
applying a scaling factor to one or more presentation times for remotely initiated video presentations of the lead video, wherein the scaling factor increases a value corresponding to remotely initiated video presentations relative to in-service video presentations;
determining a video presentation time attributable to the given video as a lead video based on a total presentation time of other videos during the multiple user sessions that were initiated with the given video as the lead video; and
generating a creator score for a creator that supplies the given video based on the presentation time attributable to the given video as the lead video.
2. The system of claim 1 , the operations further comprising:
generating a presentation score for the given video based on the video presentation time attributable to the given video as the lead video;
ranking the given video among other videos based on the presentation score; and
updating a user interface to present a highest ranked portion of the ranked videos at a client device according to the ranking.
3. The system of claim 2 , wherein ranking the given video based on the presentation score comprises ranking the given video to reduce a number of videos previewed by the user before the given video is presented to the user based on the presentation score and a friend of the user identifying the given video for the user.
4. The system of claim 3 , wherein applying a scaling factor to one or more presentation times for remotely initiated video presentations of the lead video comprises:
applying a first scaling factor to one or more presentation times for in-service initiated video presentations of the lead video; and
applying a second scaling factor to one or more presentation times for remotely initiated video presentations of the lead video, wherein the first scaling factor is lower than the second scaling factor.
5. The system of claim 4 , the operations further comprising:
identifying the remotely initiated video presentations of the lead video based on referrer information included in a request to present the lead video, wherein the referrer information specifies one of a third-party website that directed a user to an online video distribution service, a third-party native application that directed a user to the online video distribution service, or shared link that directed a user to the online video distribution service;
collecting timestamps indicating playback start times of the lead video;
obtaining pings generated during playbacks of the lead video; and
determining various presentation times for the playbacks of the lead video based on the timestamps and the pings.
6. The system of claim 5 , the operations further comprising:
identifying search queries that resulted in the lead video being identified to various users in search results;
for each of the search queries, determining a portion of the various users that initiated presentation of the given video through interaction with the search results; and
determining, based on the determined portions, search scaling factors to apply to the determined presentation time of the lead video for the presentations of the given video that were initiated through user interaction with the search results that identified the lead video.
7. The system of claim 1 , the operations further comprising:
ranking the creator among other creators based, at least in part, on the creator score.
8. A method comprising:
identifying, by one or more processors and for a given video, session start data specifying that the given video was a lead video that initiated video presentation to one or more users during multiple user sessions;
for each of at least some user sessions among the multiple user sessions:
classifying the lead video as one of an in-service initiated video presentation or a remotely initiated video presentation; and
applying a scaling factor to one or more presentation times for remotely initiated video presentations of the lead video, wherein the scaling factor increases a value corresponding to remotely initiated video presentations relative to in-service video presentations;
determining, by the one or more processors, a video presentation time attributable to the given video as a lead video based on a total presentation time of other videos during the multiple user sessions that were initiated with the given video as the lead video; and
generating, by the one or more processors, a creator score for a creator that supplies the given video based on the presentation time attributable to the given video as the lead video.
9. The method of claim 8 , further comprising:
generating a presentation score for the given video based on the video presentation time attributable to the given video as the lead video;
ranking the given video among other videos based on the presentation score; and
updating a user interface to present a highest ranked portion of the ranked videos at a client device according to the ranking.
10. The method of claim 9 , wherein ranking the given video based on the presentation score comprises ranking the given video to reduce a number of videos previewed by the user before the given video is presented to the user based on the presentation score and a friend of the user identifying the given video for the user.
11. The method of claim 10 , wherein applying a scaling factor to one or more presentation times for remotely initiated video presentations of the lead video comprises:
applying a first scaling factor to one or more presentation times for in-service initiated video presentations of the lead video; and
applying a second scaling factor to one or more presentation times for remotely initiated video presentations of the lead video, wherein the first scaling factor is lower than the second scaling factor.
12. The method of claim 11 , further comprising:
Identifying the remotely initiated video presentations of the lead video based on referrer information included in a request to present the lead video, wherein the referrer information specifies one of a third-party website that directed a user to an online video distribution service, a third-party native application that directed a user to the online video distribution service, or shared link that directed a user to the online video distribution service;
collecting timestamps indicating playback start times of the lead video;
obtaining pings generated during playbacks of the lead video; and
determining various presentation times for the playbacks of the lead video based on the timestamps and the pings.
13. The method of claim 12 , further comprising:
identifying search queries that resulted in the lead video being identified to various users in search results;
for each of the search queries, determining a portion of the various users that initiated presentation of the given video through interaction with the search results; and
determining, based on the determined portions, search scaling factors to apply to the determined presentation time of the lead video for the presentations of the given video that were initiated through user interaction with the search results that identified the lead video.
14. The method of claim 13 , further comprising:
ranking the creator among other creators based, at least in part, on the creator score.
15. One or more non-transitory computer-readable media having instructions stored thereon that, when executed by one or more processors, cause performance of operations comprising:
identifying, for a given video, session start data specifying that the given video was a lead video that initiated video presentation to one or more users during multiple user sessions;
for each of at least some user sessions among the multiple user sessions:
classifying the lead video as one of an in-service initiated video presentation or a remotely initiated video presentation; and
applying a scaling factor to one or more presentation times for remotely initiated video presentations of the lead video, wherein the scaling factor increases a value corresponding to remotely initiated video presentations relative to in-service video presentations;
determining a video presentation time attributable to the given video as a lead video based on a total presentation time of other videos during the multiple user sessions that were initiated with the given video as the lead video; and
generating a creator score for a creator that supplies the given video based on the presentation time attributable to the given video as the lead video.
16. The one or more non-transitory computer-readable media of claim 15 , the operations further comprising:
generating a presentation score for the given video based on the video presentation time attributable to the given video as the lead video;
ranking the given video among other videos based on the presentation score; and
updating a user interface to present a highest ranked portion of the ranked videos at a client device according to the ranking.
17. The one or more non-transitory computer-readable media of claim 16 , wherein ranking the given video based on the presentation score comprises ranking the given video to reduce a number of videos previewed by the user before the given video is presented to the user based on the presentation score and a friend of the user identifying the given video for the user.
18. The one or more non-transitory computer-readable media of claim 17 , wherein applying a scaling factor to one or more presentation times for remotely initiated video presentations of the lead video comprises:
applying a first scaling factor to one or more presentation times for in-service initiated video presentations of the lead video; and
applying a second scaling factor to one or more presentation times for remotely initiated video presentations of the lead video, wherein the first scaling factor is lower than the second scaling factor.
19. The one or more non-transitory computer-readable media of claim 18 , the operations further comprising:
identifying the remotely initiated video presentations of the lead video based on referrer information included in a request to present the lead video, wherein the referrer information specifies one of a third-party website that directed a user to an online video distribution service, a third-party native application that directed a user to the online video distribution service, or shared link that directed a user to the online video distribution service;
collecting timestamps indicating playback start times of the lead video;
obtaining pings generated during playbacks of the lead video; and
determining various presentation times for the playbacks of the lead video based on the timestamps and the pings.
20. The one or more non-transitory computer-readable media of claim 19 , the operations further comprising:
identifying search queries that resulted in the lead video being identified to various users in search results;
for each of the search queries, determining a portion of the various users that initiated presentation of the given video through interaction with the search results; and
determining, based on the determined portions, search scaling factors to apply to the determined presentation time of the lead video for the presentations of the given video that were initiated through user interaction with the search results that identified the lead video.