Global Metrics for Measuring Electronic Media

AD CLICK
"An AD IMPRESSION clicked on by a valid USER."

AD IMPRESSION
"A file or a combination of files sent to a valid USER as an individual advertisement as a result of that USER's request being received by the server."

In effect, one request by a valid USER should result in one AD IMPRESSION being claimed. Using this standard measurement on the certificate is optional. It is usually only relevant when auditing ad servers.

The method of counting AD IMPRESSIONs can be either:

AD INSERTION
"The insertion of an advertisement image source tag into the content file prior to the delivery of the file to the browser as measured by the advertisement serving software (exclusive of non-qualifying activity and internal users)."


AD REQUEST
"The initial request of an advertisement from the browser as measured by the server that "redirects" a browser to the specific location of the advertisement (exclusive of non-qualifying activity and internal users)."

AUDIO STREAM
"A STREAM for an audio file."

AUTOMATED PAGE IMPRESSION
"A PAGE IMPRESSION sent to a valid USER as a result of an automatic process."

All Audit Certificates must carry a breakdown of the totals of AUTOMATED PAGE IMPRESSIONS from the overall PAGE IMPRESSION total where such traffic forms 5% or more of the overall PAGE IMPRESSION count.

CHAT DURATION (HTML ONLY)
"The total elapsed time in seconds between the first and last time stamp recorded for each valid USER's CHAT IMPRESSIONs."

In order to measure CHAT DURATION, a first and last CHAT IMPRESSION record must exist for each UNIQUE USER. Therefore, USERs making single-record CHAT IMPRESSIONS, and any such CHAT IMPRESSIONS, are excluded.

This metric should be used alongside the CHAT IMPRESSION and UNIQUE CHAT USER metrics.

CHAT IMPRESSION (HTML)
"A file, or combination of files, sent to a valid USER while the USER is involved in an interactive Chat session (i.e. a PAGE IMPRESSION generated by a Chat URL.). Chat URLs show the input of one or more concurrent USERS, visible to each other, updated frequently, so as to enable a text based conversation."


All Audit Certificates must carry a breakdown of the totals of CHAT IMPRESSIONS from the overall PAGE IMPRESSION total where such traffic forms 5% or more of the overall PAGE IMPRESSION count.

DISTRIBUTION ENDED:
"The date and time expressed in hours and minutes (GMT) when the last email in a campaign/distribution was sent to the last address on the distribution list."

This would be the timestamp of the last message in the mailing.

DISTRIBUTION (GROSS):
"The total number of emails sent as part of a single campaign/distribution to all (SMTP) addresses on the distribution list."

This number includes EMAILS BOUNCED.

DISTRIBUTION STARTED:
"The date and time expressed in hours and minutes (GMT) when the first email in a campaign/distribution was sent to the first address on the distribution list."



This would be the timestamp of the first message in the mailing.

DOUBLE OPT-IN:
"A unique SMTP address that has been added to the distribution list as a result of a confirmation email sent as a reply to the list owner by the address user after the user has previously opted in."

This differs from Opt-In in that the address user specifically has to confirm their interest in subscribing by sending a confirmation email as a reply to the list owner. User Duplication "The total number of identical unique device IDs in the distribution list expressed as a percentage of the gross distribution, rounded down to the nearest whole number."

This must be declared whenever UNIQUE SMS USERs are certified.

DISTRIBUTION (NET):
"The total number of emails successfully sent as part of a single campaign/distribution to all (SMTP) addresses on the distribution list."

This number excludes EMAILS BOUNCED.

DUPLICATION:
"The total number of identical SMTP addresses in the distribution list expressed as a percentage of the gross distribution, rounded down to the nearest whole number."

Note that this means exact, case-insensitive duplication of addresses.

EMAILS BOUNCED:
"The total number of emails in the gross distribution that generated an NDN (Non-Delivery-Notice) measured at least 24 hours after the time of the last email sent in that single mailing."

Any SMTP hard delivery error (e.g. "Address owner not recognised", "Delivery Failure") counts as a bounce. Re-queuing or delayed delivery do not count.

FLASH IMPRESSION
"An interactive media file, or combination of files, sent to a valid USER as a result of that USER's request being received by the server."

FLASH IMPRESSIONS can be included within the headline PAGE IMPRESSION total on a certificate, but must be broken out where such traffic forms 5% or more of the overall PAGE IMPRESSION count.

FLASH DURATION
"The total elapsed time in seconds between the serving of an interactive file and any subsequent file to a valid USER."

Often, there is no way of knowing from the server log(s) whether the USER is still interacting with the executable. If no record of a subsequent file can be determined, then the duration is deemed to be zero. A method of creating the time stamps required for this metric is to create two frames, the first containing the executable and the second containing a blank HTML page which has a META Refresh tag set to 30 seconds and an easily identifiable URI query string. The resulting log file entries can be used to calculate FLASH DURATION.

This metric should be used alongside the FLASH IMPRESSION metric.

INTERACTIVE TV AD CLICK
"An INTERACTIVE TV AD IMPRESSION clicked on by a valid USER."

INTERACTIVE TV AD IMPRESSION
"A file or a combination of files sent to a valid USER as an individual advertisement as a result of that USER's request being received by the server."

In effect, one request by a valid USER should result in one INTERACTIVE TV AD IMPRESSION being claimed. Using this measurement on the certificate is optional.

INTERACTIVE TV IMPRESSION
"A file or combination of files intended for an Interactive TV device, sent to a valid USER as a result of that USER's request being received by the server."

INTERACTIVE TV IMPRESSIONS can be included within the headline PAGE IMPRESSION total on a certificate, but must be broken out where such traffic forms 5% or more of the overall PAGE IMPRESSION count.

MESSAGE DUPLICATION
"The total number of identical combinations of the same unique device ID and message found in the activity logs, expressed as a percentage of the gross distribution, rounded down to the nearest whole number."

This must be declared whenever SMS MESSAGEs are certified.

OPT-IN:
"A unique SMTP address that has been added to the distribution list as a result of a positive action by the address user."

This action includes, but is not limited to, the address user clicking on a subscription link or sending a subscription email.

PAGE IMPRESSION
"A file, or combination of files, sent to a valid USER as a result of that USER's request being received by the server"

In effect, one request by a valid USER should result in one PAGE IMPRESSION being claimed. In most cases, a single request from a USER causes the server to send several files to satisfy the request. For example, the server may send an html file followed by several associated graphics images and audio files. A single request from a user may also cause the server to send several additional HTML files to build a frameset. The site must ensure that all additional, non-requested files are filtered out and excluded when counting the claimed number of PAGE IMPRESSIONs. Please note that files that contain specific types of advertising creative, such as banners or skyscrapers, and files that represent STREAMs are not valid for the counting of PAGE IMPRESSIONs but should be used separately to identify AD IMPRESSIONs or STREAMs

PDA SYNCH
"The total number of times that PDA devices synched with the site (or channel)."

This is measured by requiring one URL per site (or channel) to be non-cacheable. This then acts as the identifier for counting PDA SYNCHs. Measuring how many times this URL was requested in the Audit Period will therefore allow a count of total PDA SYNCHs. Break-outs by channel can also be supported by this approach.

This creates a comparable and indicative measure of a site's PDA audience. The agreed metric is not "PDA Users", as we are not identifying the receiving device. This may become possible in future.

PERCENTAGE BOUNCED:
"The total number of emails in the gross distribution that generated an NDN measured at least 24 hours after the time of the last email sent, expressed as a percentage of the gross distribution."

REFERRAL IN
"A PAGE IMPRESSION representing an arrival at the site by a valid USER from another identifiable site". Effectively, the event that needs to be detected is the arrival at the site by a USER who has come from another identifiable site. To do this requires testing whether this Page Impression has a Referrer field that is native to the site or is Null; any non-null, non-native values may be counted as REFERRALs IN.

REGISTERED ADDRESS:
"A contactable SMTP address in the distribution list that has requested the email service."

This is not a measure of individual people or addressees, because an individual address user may have more than one registered email address.

REPEAT UNIQUE USERS
"The total number of UNIQUE USERs who make more than one VISIT. "

This metric should be expressed as a percentage of the total UNIQUE USER figure when shown on certificates. SINGLE UNIQUE USERS+REPEAT UNIQUE USERS = TOTAL UNIQUE USERS.

SEARCH
"The first PAGE IMPRESSION sent to a valid USER as a result of that USER's search request being received by the server."

In effect, one search request by a valid User should result in one Search being claimed. This requires that the site, and hence the audit, can identify the first Page Impression served in response to a search request from a valid User and differentiate this first results page from any others.

The Search total for a site is distinct from its Page Impression total. Searches do NOT therefore contribute to the headline Page Impressions certified.

SINGLE UNIQUE USERS
"The total number of UNIQUE USERs who make only a single VISIT."

This metric should be expressed as a percentage of the total UNIQUE USER figure when shown on certificates. SINGLE UNIQUE USERS+REPEAT UNIQUE USERS = TOTAL UNIQUE USERS.

SMS MESSAGE
"A file or combination of files sent to a valid USER's SMS-enabled device by the server."

Typically, SMS devices are mobile phones or PDAs. One file sent by the server should result in one SMS MESSAGE being claimed. This metric only measures the opportunity for the SMS MESSAGE to be sent to the USER, since the serving of an SMS MESSAGE does not guarantee that a USER actually received it.

Generally, the SMS server sends a single file to the device, comprising a single SMS MESSAGE. The server log file can be interrogated to establish a count of SMS MESSAGEs. Only one file per message is deemed valid. Therefore, the client must filter out and exclude any additional files when claiming a number of SMS MESSAGEs.

STREAM
"An indicator of a media file stream sent to a valid USER as a result of that USER's request being received by the server."

In effect, one request by a valid USER should result in one STREAM being claimed. The indicator of the stream can be either the request for the streamed file itself, or else the opening of the media player which is intended to play the stream. This metric allows the measurement of the total number of non-interactive streamed content downloads sent to a USER.

This metric is designed to measure only the number of times that a USER has requested the streaming process. Therefore, it requires USER interaction. The methodology used to count STREAMS can involve data derived from (a) the web server; or (b) the streaming server; and/or (c) the media player application.

The speed of connection can be an important factor in measuring streamed content. However, this variable is not part of the scope of the standard ABCE audit.

STREAMS may optionally be broken out into Live and On-Demand, and also into AUDIO and VIDEO STREAMS.

The STREAM total for a site is distinct from its Page Impression total. STREAMS do NOT therefore contribute to the headline Page Impressions certified.

STREAM DURATION
"The total elapsed time in seconds between the first and last recorded indicator of each stream sent to a valid USER."

This metric should be used alongside the STREAM metric. Please note that sites may additionally state:

The average elapsed time in seconds (calculated by dividing the total elapsed time in seconds for all streams in the audit period divided by the total number of STREAMs in the audit period) for all valid streams in the audit period.
and/or
A range of elapsed times in seconds, breaking down the total number of streams of a given length for all valid streams in the audit period.

UNIQUE CHAT USER (HTML CHAT ONLY)
"The total number of UNIQUE USERs who make one or more CHAT IMPRESSIONs."

This is typically a count of all the UNIQUE USERs who have contributed to the total CHAT IMPRESSIONS.

UNIQUE FLASH USER
"The total number of UNIQUE USERs who make one or more FLASH IMPRESSIONs."

This is typically a count of all the UNIQUE USERs who have contributed to the total FLASH IMPRESSIONS.

UNIQUE HOST
"The total number of unique IP addresses."

UNIQUE INTERACTIVE TV USER
"The total number of UNIQUE USERs who make one or more INTERACTIVE TV IMPRESSIONs."

This is typically a count of all the UNIQUE USERs who have contributed to the total INTERACTIVE TV IMPRESSIONS.

UNIQUE SEARCH USER
"The total number of UNIQUE USERs who make one or more SEARCHes."

UNIQUE SMS USER
"The total number of unique device IDs recorded in the log files for the SMS MESSAGEs sent."

This will typically be a count of all the unique telephone numbers.

UNIQUE STREAM USER
"The total number of UNIQUE USERs who make one or more STREAMs."

This is typically a count of all the UNIQUE USERs who have contributed to the total STREAMS.

UNIQUE USER (USER)
"A unique and valid identifier. Sites may use (i) IP+User-Agent, (ii) Cookie and/or (iii) Registration ID."

Where a UNIQUE USER is calculated by IP+User-Agent, this definition may overstate or understate the real number of individual USERs concerned due to dynamic IP address allocation (for example by a dial-up Internet Service Provider) or to significant levels of uniformity in IP and browser configurations operating through a proxy. ABC ELECTRONIC will discuss with the site where they wish to use an alternative identifier (e.g. registration ID's for Registered UNIQUE USERs, cookies, etc.) which offer additional methods of defining different UNIQUE USERs visiting the site. Certain classes of traffic are not valid USERs and must be excluded. The VISIT metric will be calculated on the basis of the USER metric used.

UNIQUE USER DURATION
"The total time in seconds for all VISITs of two or more PAGE IMPRESSIONs, divided by the number of UNIQUE USERs making such VISITs."

In order to measure USER DURATION, a first and last PAGE IMPRESSION record must exist for each VISIT. Therefore, USERs making VISITs of only one page are excluded, since no interval can be established.

VIDEO STREAM
"A STREAM for a video file."

VISIT
"A series of one or more PAGE IMPRESSIONS, served to one USER, which ends when there is a gap of 30 minutes or more between successive PAGE IMPRESSIONS for that USER."

Using this standard measurement on the certificate is optional. A VISIT is effectively a burst of activity by a valid USER.

VISIT DURATION
"The total time in seconds for all VISITs of two or more PAGE IMPRESSIONs, divided by the total number of VISITs of two or more PAGE IMPRESSIONs"

In order to measure VISIT DURATION, a first and last PAGE IMPRESSION record must exist for each VISIT. Therefore, VISITs of only one page are excluded, since no interval can be established.

WAP AD CLICK
"A WAP AD IMPRESSION clicked on by a valid USER."

This metric is optional. The WAP AD CLICK total is usually the total number of WML requests for an ad file recorded in the relevant server log(s).

WAP AD IMPRESSION
"A file or a combination of files sent to a valid USER as an individual advertisement as a result of that valid USER's request being received by the server."

In effect, one request by a valid USER should result in one WAP AD IMPRESSION being claimed. The method of counting WAP AD IMPRESSIONs can be either:

Server Side Insertion: The WAP server requests a file from the Ad server after receiving a request for a WML file from a USER. A WAP AD IMPRESSION occurs when the Ad server returns the relevant file.

Direct Ad Serving: A WAP AD IMPRESSION occurs if the Ad server successfully serves or redirects to an Ad file when the USER's gateway requests an WML file.

WAP IMPRESSION
"A file or combination of files intended for a WAP device, sent to a valid USER as a result of that USER's request being received by the server."

WAP IMPRESSIONs can be included within the headline PAGE IMPRESSION total, but must be broken out where such traffic forms 5% or more of the overall PAGE IMPRESSION count.