Question | Answer | Product | Catergory | Sub | Date | Edited By | Used |
What happens in VA if a user close down the window during the video – will we collect the data at all or report the watched minutes? | If the user returns to the website then we will match the cookie up and close out the video play. If this revisit was greater than 30 minutes after the window was closed then in the metrics we will truncate the watched minutes to 30 minutes as per the IAB definition of a video session. | VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
What happens if a user during a video keep the video running and writing in a new destination url and leaves the video for another site – will we collect and report the watched minutes? | Yes
we will but it is dependent on a number of implementation requirements. These are: 1: Event 49 (periodic play) is coded in the video player. If this is not provided then we assume 30 minutes duration as per the IAB definition of a video session. 2: allowScriptAccess is set to “always” if you are implementing the flash based variant of the combined beacon. This is to allow the flash version of the combined beacon to execute some page-level JavaScript. This is not required if you are implementing the fully native JavaScript version of the combined beacon as the website would already permission JavaScript to execute. |
VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
Are VA measuring all type of Browsers or are there any one(s) that we can’t measure? It seems that Omniture and Nedstad has problems with Opera | The
combined beacon does not measure browser activity, only video player
activity. As such, there should not be
any browser limitation as long as the browser officially supports the video
player technology the publisher is using. For a definitive answer on Opera I would need to know what issues Omniture has. I suspect it is with the page-level tag. Opera is actually quite good at supporting most video player types. |
VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
Will all browser generate time spent (and the full viewing time) | It is not dependant on browser, but on the video player types we support. Currently this is Flash AS2 and AS3 variants, Silverlight 1,2 & 3, Windows Media Player 9+ | VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
Are we measuring all types of Operating systems? | It is not dependant on operating system, but on the video player types we support. If the operating system can support our officially supported players then there should not be an issue. | VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
TV4 has partner sites, like expressen.se – and expressen.se imbedding the TV4 player in their site – can TV4 track in VA the number of users that has watched their video per partner site and understand the total number of plays and users? | Yes, we can track the referring domain. However, for Flash versions of the combined beacon we do require allowScriptAccess to be set to “always” so that page-level JavaScript can be executed. | VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
Is collected data surface in the reporting UI in real-time? | Data
is not currently processed and surfaced in real time. Data received before 12 midnight in the time zone declared for a specific publisher will be available in the reporting UI by 9am next day, in the time zone declared for the specific publisher. This is the worst case scenario where a video stream start has been registered but an explicit video stop has not been sent. i.e. up to 8 hours after the video start was logged. |
VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
Will the beacon generate a lot of network traffic to the Nielsen data collection server? | No.
The optimal (default) configuration is to cache all events other than the
initial video play start and video play end. The first time a visitor plays a video on any tagged publisher site, the API will make a call to fetch the beacon code from the Nielsen data collection server. The beacon will then be cached on the users machine. For each video play, a web request to the Nielsen data collection server is made upon receiving the first play event (event 5) after seeing a load video event (event 3). At the end of each video play (upon receiving a video stop event 7), or the detection of a video termination through browser close or URL change, another web request is made to the Nielsen data collection server is made to flush all the detected events during the users video viewing session; scrubbing back and forth, pause, resume, volume up / down. |
VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
How is the Referring URL for a video derived? | The
Nielsen beacon uses standard JavaScript script (“document.referrer”) to
extract the referring URL from the header. It should be understood that this property can be blocked or manipulated which can lead to misleading information being returned to the beacon when it makes this inquiry. Common reasons (but not limited to) for this blocking are:- - The referring URL in the web page header has been removed or NULL out. - Anti-virus software on the user PC has masked or removed the referring URL in the web page header. - The referring URL in the request header can be manipulated / changed by the web server. You would need to ask the webmaster of the client website. - The web server can block the JavaScript that is used by the beacon to extract the referring URL from the document.referrer property in the HTML DOM. You would need to ask the webmaster of the client website. |
VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
The Video Stream will not playback in the Reporting UI SynchPlayer. Why? | For
each video stream start recorded by the beacon, the second parameter in the
event 3 or event 15 must supply the full file path URL for play back
purposes. Assuming that the full path
is supplied for each video file, then the following conditions also have to
be met:- a) The video file is not encrypted. i.e. RTMPe are usually not playable outside of the publisher’s player. b) The video file is accessible to the public internet. If the publisher is hosting the video file behind a secure mechanism that requires the video player to pass a hideen sign-on or other mechanism to get access then the Nielsen Video Analytics UI may not have permission to access the file URL. c) The file path URL is still valid. The URL may have been valid at the time the stream was recorded but subsequently the publisher may have deleted it, or moved it to another location. d) The file is an unsupported video type. At this time we support playback of FLV video. Note: Nielsen Video Analytics does not store the video file in it’s database. It stores the URL reference to the video file. |
VA | Auditing | 11/1/2010 | Kenny Lau | PUBLIC | |
Are Combined Beacon messages compatible with existing tags? | Combined Beacon messages are a strict superset of existing messages for VideoCensus and SiteCensus. Current messages will continue to work. | VA | Daily DAV | 11/1/2010 | Kenny Lau | TAM | |
Why Can’t I use VideoCensus to pull the same information? | VideoCensus
is a syndicated tool that will show only a number of defined reports such a
total Streams Count, Title of stream, Uniqe Browsers, etc. However, for troubleshooting purposes the DAV files will provide you with a number of cross-field information such as VCID (CI & C6), MarketView Brand, Parent, and Channel. |
VA | Daily DAV | 11/1/2010 | Kenny Lau | TAM | |
Where & how do I access the daily DAV files? | The
daily DAV files are dumped daily by our Au engineers on the global FTP server
at export02-au.imrworldwide.com Inquire your local TAM team/manager for login&password. Alternatively, email rt.censusproducts@nielsen.com (Level2 in melbourne) for help. The folder you should look at is probably the /dav_stats/vc-daily-us |
VA | Daily DAV | 11/1/2010 | Kenny Lau | TAM | |
What Qualifies as an Ad? | •
The video is streamed from a 3rd party server (eg: Atlas or DoubleClick) or
from a location on a publisher’s CDN that is used for serving ads. • The video is shown in rotation where other advertising appears (ex: within a commercial “pod” or “chapter break” between segments of an episode). • The video is clearly not the primary viewing experience and intended to be shown as advertising. |
VA | General | Ad | 11/1/2010 | Kenny Lau | PUBLIC |
Are Promotional Videos counted as Ads? | •
Promos are only counted when not shown in an ad rotation spot. • For example- on a page for a specific show- the user is seeking out the promotional content as their primary viewing experience. • If the promo is shown in an ad rotation position (ex: within a commercial “pod” or “chapter break” between segments of an episode) it will be counted as an ad. |
VA | General | Ad | 11/1/2010 | Kenny Lau | PUBLIC |
My organization has mandated that we should not force cookie tracking technology on our users. Is there anything I need to do in the video player / API? | There is no beacon specific change required. Individual users may visit a privacy page on the Nielsen website and “opt-out” of tracking. Any subsequent video views / plays that trigger the beacon will not be tracked for that user | VA | General | Cookie | 11/1/2010 | Kenny Lau | PUBLIC |
We do not, or cannot use Flash cookies. Do you have a video beacon technology that uses regular browser cookies? | Yes. You should use the Javascript API in native mode which will force the use of the Javascript beacon for collecting metrics and will use a browser cookie. | VA | General | Cookie | 11/1/2010 | Kenny Lau | PUBLIC |
Do we need to capture all the events? | No. Load Video and Play are the required events. All other events are optional or are automatically generated. But, more data capture leads to better measurement of a viewer’s experience in a single video play. When aggregated across a large number of views, this results in accurate profiles of user experience with videos. | VA | General | Events | 11/1/2010 | Kenny Lau | PUBLIC |
What’s an Event 49 (aka Periodic Play)? | In
order to provide accurate Video Analytic metrics, it is strongly advised to
generate a “periodic play” event, or Event 49. This does not result in any
additional network traffic as the periodic plays are cached and subsequently
flushed to the server upon receiving the video stop/end event. In some cases the event is not |
VA | General | Events | 11/1/2010 | Kenny Lau | PUBLIC |
What is the suggested interval for sending Event 49 (aka Periodic Play) ? | Nielsen suggests that an Event 49 be sent every 5 seconds for Ad’s (“preroll”, “midroll”, and “postroll”) and every 30 seconds for “content”. | VA | General | Events | 11/1/2010 | Kenny Lau | PUBLIC |
Do we need to capture all the API events? | No. Load Video and Play are the required events. All other events are optional or are automatically generated. But, more data capture leads to better measurement of a viewer’s experience in a single video play. When aggregated across a large number of views, this results in accurate profiles of user experience with videos. | VA | General | Events | 11/1/2010 | Kenny Lau | PUBLIC |
Why do we need a beacon anyway? | As the number and types of events that need capture goes up, efficient messaging back to the server becomes paramount. View counts can be achieved without a beacon. Duration metrics are problematic – more accurate numbers need more messages and affect the cost of service. % viewed and other metrics are impractical without a beacon (i.e. code of some sort attached to the player). Emerging metrics for strict auditing, content-ad interaction and interactive videos are impossible without a beacon. | VA | General | Events | 11/1/2010 | Kenny Lau | PUBLIC |
For the purposes of metrics we record the video stop before the actual end of the video content (Also known as the “black screen”). Can we send the stop event before the actual end of the video? | You can send the stop video (event 7) any time you wish as long as the analytics team understand that you are doing so. If possible you should keep track of the playhead position and throw the stop video (event 7) if the user scrubs forward through the time point in question. | VA | General | Events | 11/1/2010 | Kenny Lau | PUBLIC |
How long does it take to tag videos with the combined beacon? | This depends upon a variety of factors, including the type of player, the number of events required to support the service(s) the client is tagging for, the clients development bandwidth etc…. Some implementations are very simple and take a matter of hours, some implementations take up to a month of iterative collaboration between the client and a Nielsen TAM. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Does the combined beacon use a cookie? | Yes. A flash cookie. If flash cookies have been disabled then it defaults to a page level cookie, but only in this instance. We use the flash cookie because it is more accurate and persistent. If you have questions surrounding the slight discrepancies which may be caused by this, please contact your Nielsen TAM representative. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Which video players are supported? | Flash Players (AS2 & AS3), Brightcove (v2 & v3), ThePlatform (PDK4), Javascript, Silverlight via Javascript, Windows Media Player via Javascript, JW Player. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Which Nielsen services does the combined beacon support? | Video Analytics, Video Census, IAG, MI Streaming, SC Streaming | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Does the combined beacon only support certain types of browser? | The combined beacon does not measure browser activity, only video player activity. As such, there should not be any browser limitation as long as the browser officially supports the video player technology the publisher is using. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Is there any particular advantage to using the flash implementation as apposed to the javascript implementation? | Apart from any technology allegiances a particular customer may have, there are no specific performance reasons why you would use one over the other. It could be argued that the flash implementation is preferable from a portability/embed point of view because once compiled into the player the beacon is guaranteed to follow any player placement. The javascript implementation is ultimately a page level <script> implementation and therefore susceptible to omission from an embed player when placed on a website outside of the publisher’s immediate control. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Does the combined beacon affect the viewer experience or cause any latency on the player | No, there is no visible affect on the users viewing experience. The beacon is designed to cache all events captured between the video start and video end. Because there is no network traffic generated during user viewing time then the user experience is preserved. Buffering also minimizes the number of round trips to the Nielsen server. Upon video end detected (or browser URL change) the beacon transmits all events since the video start up to the Nielsen server as one POST request. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Where do I find the client documentation regarding implementation? | You may go to the following link: http://www.nielsenonlinesupport.com/clientsupport/index.html | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
What is the formula for Attentiveness? | 1)
Baseline Attentiveness Score: Time
spent in seconds * 100 / Length of the clip in seconds. Time spent is the
number of seconds the video is play.
It takes into account Play, Pause, Skip, Rewind and other events. If this value is more than 100, it is set
to 100. It is the basic score and goes
through the following adjustments. 2) Focus Factor adjustment: Focus Factor is the % of time focus was on the video window rather than something else on the viewer’s computer. Focus Factor has a weight of 20%, so score is modified like this Attentiveness = baseline score * (0.8 + (Focus Factor/100) * 0.2) 3) Positive event adjustment: Any increase in volume adds 5 to score. Going Full screen adds 10 score. 4) Repeat view adjustment (*for content only): If the same user is seeing the same clip for the second time, the score for the second time will be at least as much as the score for his first play. Repeat Viewing adjustment is not done for Ads. |
VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
If I see a bug/want to suggest a feature who do I contact? | Email to Sophie Coleman: sophie.coleman@nielsen.com | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
What timezone does VA report in? | The default timezone for VA is PST. All clients should be asked for a timezone preference upon starting implementation so this can be configured correctly on the glanceguide back end. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Why do the stream count numbers vary between Video Analytics and SC Streaming/MI Streaming/VideoCensus? | There are a number of reasons that you should expect to see a slight (less than 5% over time) discrepancy between census counts and VA counts, for both uniques and streams. This is due to a variety of factors and is acceptable. If the count varies by more than 5% over an extended period this is too high a discrepancy, please contact your Nielsen TAM Representative. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
How much past data is retained in a Video Analytics account? | VA retains a rolling 12 months of back data. | VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
List of reports available? | Steam
views/Unique browser Stream views/Session Cumulative stream views Beginning to end views Bounce rate Average % complete Total viewing time Average viewing time Average session duration Audience exposure Average attentiveness Category performance Sub-category performance Top referring domains Top countries Top regions Period analysis (by hour) Audience loss (by second) |
VA | General | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Does the combined beacon affect end-user video viewing? | No.
The beacon is a small, lightweight code that has no visible components, is
entirely in the background and is designed not to interfere with video
experience in any way. 5. The code implemented on the page looks correct. However, the beacon is not firing at all times. Sometimes it would work in IE, but never in FireFox. The problem seems to be specifically affecting MA operating systems running IE and Firefox browsers. occurring when the page where the player is loading on has more than 10 SWF objectings concurrently. to be limited to the MAC operating system. In fact, there seems to be a limit as to the number of concurrently running swf files that a single page can run in the MAC OS or in FireFox on a Mac |
VA | General | Performance | 11/1/2010 | Kenny Lau | PUBLIC |
When the user browses away from the video page during video playback, a GET record is sent out with all the previously buffered events. How VA underdstands that the user browsed away rather then simply stopped the video? | In any instance when the user stops or interrupt the video playback , and Event 7 is recorded, to indicate that a controlled stop was requested. | VA | General | Performance | 11/1/2010 | Kenny Lau | PUBLIC |
Can a click event triggered outside of the flash player area be captured by the Nielsen video combined beacon ? | No. The video beacon is only for detecting video player events, not web page level events. | VA | General | Performance | 11/1/2010 | Kenny Lau | PUBLIC |
How do you detect that a user has abruptly closed the browser / video player or abruptly navigated away from the URL hosting this viewing of the video? | You
must enable Javascript access (see section in this document entitled “Website
Configuration Required for Flash Based Players”) to Flash in order for us to
detect these user behaviours. Flash
detects the state change and triggers the JavaScript that generates the
appropriate event for our processing.
This gets stored in the beacon buffer and gets sent to the Nielsen
server next time a user watches another video, or in fact any video on any
publishers website that is tagged with our beacon technology. The VA backend processing will process messages for a given video stream that are time stamped up to eight hours after the initial video start message (event 15 or event 3 + 5). If an explicit video stop event (event 7) or an unload video event (event 2) is not seen within that time frame for a video stream then the video will be deemed stopped and the stream count credited. The duration of that video will be approximated based on the last event received for that video stream. This could be any API event; pause, stop, forward, back, periodic play update (event 49) |
VA | General | Performance | 11/1/2010 | Kenny Lau | PUBLIC |
It is already the 7th of the month and the Demographics data is not showing for the previous month. Why? | Demographic data for the previous month is typically available on the 11th of the month. i.e April data will be available in the UI on the 11th of May. | VA | General | Performance | 11/1/2010 | Kenny Lau | PUBLIC |
What is the attentiveness score? | The
attentiveness score is a computed metric that indicates how engaging the
video content is. The greater the score, the more engaged the user base
is. Maximum score is 100. It is computed from the following user
behavior when a video is in play: - Viewing: Time measurement - Focus: Is the video player in a foreground window (in focus) or background window - Interaction: Positive event adjustment such as volume change. - Affinity: Repeat viewing |
VA | General | Performance | 11/1/2010 | Kenny Lau | PUBLIC |
Why do we need a plugin anyway? | As the number and types of events that need capture goes up, efficient messaging back to the server becomes paramount. View counts can be achieved without a plugin. Duration metrics are problematic – more accurate numbers need more messages and affect the cost of service. % viewed and other metrics are impractical without a plugin (i.e. code of some sort attached to the player). Emerging metrics for strict auditing, content-ad interaction and interactive videos are impossible without a plugin. | VA | General | Plugin | 11/1/2010 | Kenny Lau | PUBLIC |
Are Combined Beacon messages compatible with existing tags? | Combined Beacon messages are a strict superset of existing messages for VideoCensus and SiteCensus. Current messages will continue to work. | VA | General | 11/1/2010 | Kenny Lau | PUBLIC | |
When auditing a Flash implementation (both FLVPlayback component and NetStream object) I notice that that the resume play (event 5) sent after the forward seek (event 8) actually has a time point fractionally less than that of the time point indicated in parameter 2 of the seek event (event 8). In other words, when the user skips video content, the flash engine event reports the resulting time point for the play event to be a fraction of a second ahead of the resume play time point. | Apparently, this has to do with the Flash event engine being not 100% precise in reporting time points in the instance of users skipping video content. This has been known and aknowledged by our developers team. Their feedback is that this kind of behavior would not upset the analytics computation adversely. Therefore, it can be safely ignored. | VA | Implementation | Flash | 11/1/2010 | Kenny Lau | PUBLIC |
What is AllowScriptAccess and why it is important that it is implemented correctly? | Once the video starts playing and the load/start record is sent to the Nielsen servers, all the subsequent events resulting from the user interacting with the player buttons (stop,pause,forward,play,etc) are buffered in a local cache and then sent periodically as a POST record to the Nielsen servers. One event we need to make sure is captured correctly is when the user browses away from the video page while the video is still playing. In this instance, a GET record is triggered so that the buffered information is sent to the Nielsen servers. If this doesn’t happen, the video stats that will result will be missing useful information. For example: all of the user interaction during playing time (forward, stop, pause), will be lost when the user browses away. For this to happen, AllowScriptAccess must be set to “always” within the http page where the video content is called. | VA | Implementation | Flash | 11/1/2010 | Kenny Lau | PUBLIC |
How can I make sure that AllowScriptAccess is set to “always” in the http code calling for the video content? | In
order to successfully have the GET record sent out after a user browses away
from a video page where a video is playing,
the following Flash parameter must be set in the HTML code where the
SWF file (video content) is
loaded. If embedded in a <OBJECT> tag like I did in this blog, the parameter will look something like this: <object classid=”clsid:d27cdb6e-ae6d-11cf-96b8-444553540000″ width=”400″ height=”300″ codebase=”http://download.macromedia.com/pub/shockwave/cabs/flash/swflash.cab#version=6,0,40,0″><embed type=”application/x-shockwave-flash” width=”400″ height=”300″ src=”http://www.nielsenonlinesupport.com/mauro/video/flash/Africa_NOL_FLVPlayback_music.swf” quality=”high” allowfullscreen=”false” allowscriptaccess=”always” align=”middle” bgcolor=”#e5e5e5″ name=”Africa FLVPlayback Method”></embed></object> When the user browses away from the video page, our script will send out a similar record to indicate that the video stream has been interrupted (and also will send any record describing the interaction with the player until that moment): http://secure-us.imrworldwide.com/cgi-bin/m?ci=us-502202&c6=vc,c03&cc=1&du=0&tl=dav1%2Dafrica%2Dflvplayback&ou=http%3A%2F%2Fwww%2Enielsenonlinesupport%2Ecom%2Fncb%2Fvideos%2F&tp=ggHEX40%3D%3Cm%20v%3D1%20c%3DmKOu4X8f3TA59Pt5lwf0L799Kmrq10rK%3E%23%144%28%0D%0B%1BC%1DF%7FV%3A%22a%3C1f%40G%1E%02%7FSE%13%22%1Bh%28r%3F%2A%7F%06%18%1E%02cQY%17%21%0D%3C%2Fp%23x%60QD%0D%0B%1FC%1DF%7FV%3A%22a^|^^^|^^1259714443934,49,0,100|||1259714443934,2^|^^c%2B%09%2A%7BsuJQ%19%3Ab%18%5E%7B%3C%2Fm%3 |
VA | Implementation | Flash | 11/1/2010 | Kenny Lau | PUBLIC |
Our Flash player executes the content video start a split second before it executes the preroll. The user does not visually see the content video before the preroll video but it does result in the video play event for the content executing before the video play event start executing for the preroll Ad. How would the beacon handle this scenario? | It is OK to send the start beacon for the content before the start beacon of the preroll Ad. Typically we like to see video stream start and end beacon calls supplied as discrete pairs of calls, however we can accept a Ad video start without seeing the stop event for the already sent content video start. What we CANNOT accept is another additional Ad video start or content video start without first receiving a stop event for the already “open” Ad video and / or content video stream. i.e. we only allow one level of call nesting. | VA | Implementation | Flash | 11/1/2010 | Kenny Lau | PUBLIC |
We do use a Flash player but do we have to use the Flash variant of your API and beacon? | We do not mandate that you use the Flash API but we do strongly encourage you to do so. The Flash API gets bound / compiled into your player and therefore is guaranteed to “travel” with your player if it gets embedded in other publisher websites. This ensures that all end-user plays get captured and registered. If you used the Javascript API to instrument your Flash player then you would have to make sure other publishers that embedded your video player also inserted the Javascript API and instrumentation which would be an additional management overhead. | VA | Implementation | Flash | 11/1/2010 | Kenny Lau | PUBLIC |
Why has my Nielsen TAM asked me to set cisuffix to “gg”? | Many
of our existing customers already have the Video Census beacon in place. This new super beacon is designed to also
send the same information to Video Census.
Many of our customers have communicated a resistance to the extra work
in taking the old Video Census beacon down.
As such, we have to allow for the situation where the old Video Census beacon and new super beacon execute on the same property. If left unaddressed, this would result in double-counts in Video Census. This flag, when set to “gg”, indicates to our servers that we should ignore the super beacon for the purposes of counts in Video Census. At some future point, once all the properties in a given client account are tagged with the new super beacon then the old VC beacon can be removed and the cissuffix set to “”. At that point, Video Census will then be fed counts from the new super beacon. |
VA | Implementation | Get Started | 11/1/2010 | Kenny Lau | PUBLIC |
Is it advisable to implement the javascript native mode when other methods are implemented in the same page AND with the same VCID? | No.
Among all methods described, the java script Native method will use solely a
standard page-level cookie while the java script Combo method will try and
use standard page-level cookie before
attenpting using the page-level cookie. Considering the mix of cookies used,
and considering that the page-level are less
persistent then flash cookies, it is not advised to have a mix type of
cookies processed. In fact, the end
result could easily be a UB that doesn’t reflect correctly and uniformly the
Nielsen unified video beacon published in VideoCensus and VideoAnalytics.
Considering that the other player is carrying along with the fact that the
other player is using a Flash-based
cookie, will impact the calculation of the UBs in our back-end servers. Among possible work-arounds: a) switch the native method implementation to a wrapper method b) implement a java script wrapper within the Flash player to support page-level cookies |
VA | Implementation | JavaScript/Flash | 11/1/2010 | Kenny Lau | PUBLIC |
If I am using a player based on Action Script 3. which implementation method should I use? | It
really depends on your personal preferences . Flash Players ActionScript 3 Advantage is that the beacon will always go where ever the player goes. If the player gets embedded on another website then it will automatically go with it. Disadvantage is that the video player has to be opened up to make any upgrades to the beacon in the future. Javascript API Advantage is that the video player source code does not have to be changed. Disadvantage is that the customer has to make sure the Javascript code is placed around the video player in every instance. Also, it can be difficult to mandate that the Javascript API is in place with every embedded instance. |
VA | Implementation | JavaScript/Flash | 11/1/2010 | Kenny Lau | PUBLIC |
I have a number of different player technologies on the website that use different cookie technologies (i.e. Flash cookie with Flash player, browser cookie with Silverlight) but I wish to report the data into the same Nielsen Analytics Account, which Nielsen beacon should I use? | You
should choose your cookie type and stick with it for all players. This is important from a methodology
perspective as the persistence of a particular cookie can vary. i.e. Flash cookies tend to be more
difficult to remove from a users environment and therefore that affects the
Unique Viewer metric. The Javascript API in native mode will use a browser cookie. The Flash AS2 API and AS3 API will use a Flash cookie. The Javascript API in Combo/SWF mode will use a flash cookie if it can, otherwise it will use a browser level cookie. |
VA | Implementation | JavaScript/Flash | 11/1/2010 | Kenny Lau | PUBLIC |
I have added the <ggxp_gg> parameter in the action script code so to allow compatibility with Site census streaming. However, I don’t see the PG variable being fired. Why? | In
order to work properly, you need to make sure that in ggCom.as, the right
version of the GG SWF file is called: ggce353.swf in the following line of
code: GGSWFADDRESS = “http://secure-” + _nolggGlobalParams.sfcode + “.imrworldwide.com/novms/gn/3/ggce353.swf”; |
VA | Implementation | 11/1/2010 | Kenny Lau | PUBLIC | |