Data Request – Viewability Vendor Reconciliation Project (Phase 3)
Media Rating Council
January 16, 2015
The Media Rating Council (MRC), in its ongoing effort to identify and address remaining systematic issues that cause discrepancies in the viewable impression measurements between MRC-accredited viewability measurement providers, is seeking specially structured data on viewability measurement results for production advertising campaigns.
While MRC has previously issued guidance to the industry on reasons why such discrepancies existed, and has prescribed certain procedures that accredited vendors must follow that address these, we are undertaking this additional step to ensure that any remaining material causes for systematic viewable impression counting differences are identified. The data supplied for this project will help to ensure that significant issues of this type will be addressed and resolved.
Our hope is that we will naturally get a varied set of data, ad types, etc., across market verticals, to ensure a complete view of the causes of differences. If in MRC’s judgement the coverage derived from the data submitted is insufficient to derive a complete view of vendor differences, additional specific data may be requested by MRC.
Confidentiality of Submitted Data: Data supplied to MRC for this project will be maintained under strict confidentiality. MRC will not share the identity of participants (marketers, agencies, media organizations and measurement vendors) and will not share results of campaign on an attributed basis with anyone – any results that are reported from this phase of testing will be generalized and anonymously reported by MRC.
The details of the data request are described below.
Each participating organization contributing data should identify a single point of contact who will be the lead in interacting with MRC and following up on questions. Please include the contact information for this contact, including e-mail and telephone contact with all communication.
The MRC personnel involved in this project are:
MRC telephone number: 212-972-0300
We are asking that those interested in supplying MRC with data for this purpose identify campaigns for which multiple (two or more) MRC-accredited viewability providers’ measurements can be tested and compared. This can be done in one of two ways:
- The preferred approach is for multiple vendors’ measurement tags to be applied to individual ads. If this approach is used, the order of the vendors’ tags as applied to the ad should be randomly assigned, so each vendor has an equal chance of being the first vendor tag, etc.
- Another acceptable approach is to tag each ad with only one measurement vendor tag, but to randomly assign which vendor tag is applied to the ad, so the end result is that there is an approximately equal distribution among the impressions measured by each vendor for each ad placement.
For each submitted campaign, participants should be prepared to supply MRC with the following information at the placement level:
- Specifics concerning the ad type, ad format, ad size, and any special features about the ad that should be considered (for example, whether it is an expandable ad, or a multi unit ad, a non-IAB standard ad unit, or any other attribute that distinguishes it from a typical display or video ad).
- The precise dates on which the ad measurements reported began and ended.
- The following specific measurements (by each viewability vendor included); specified in total and by publisher where possible:
- IAB valid served impressions
- Viewable impressions
- Non-viewable impressions (broken out by reasons for non-viewability, to extent these are known)
- Unmeasured impressions
- Measured rate
- Viewable rate
- Median time in view, if available
- Presence of cross domain iFrame
- Breakouts of the above measures by:
- Device type (desktop or mobile)
- Operating system
- Breakouts of the above measures by:
In addition, all participants will be asked to comply with the following test parameters:
- A short pre-test of the campaign should be conducted prior to the launch of data collection, and results shared with MRC, to ensure campaign set up and ad tagging are working properly.
- Viewability vendor results that are being compared in test campaigns should begin their measurements at the same time and end them at the same time (for reporting purposes). In other words, if two weeks of data are reported for Vendor A, the same two weeks should be reported for Vendor B, if Vendor A and Vendor B are to be compared against each other.
- While the length of the campaigns as reported for this purpose may vary, all reported campaigns should be a minimum of one week in duration.
- To the extent feasible, all participants will be asked to maintain any supporting data that exists for the submitted campaign, to allow for further in-depth investigation of certain cases as deemed necessary by MRC. This may included maintaining samples of the ad creative, records of the impression-level pixel and time viewability measurements, pertinent details concerning the ad’s placement (the site on which it appeared, etc.), and any other variables that would enable a fuller understanding of unexpected or unusual measurement results.
- Potential participants should indicate directly to the MRC in writing their plans to participate in this test program no later than February 20, 2015.
- Volunteering organizations should provide in writing to MRC information on basic details of the campaign(s) to be included in its testing (such as product category, advertiser, type of campaign, expected number of impressions and placements, expected length of campaign, expected length of test period, ad server to be used, metrics to be reported to MRC, etc.) at the earliest possible time, and no later than March 20th.
- All test data collection should begin no later than April 1st (all pre-testing must be completed prior to this date).
- All test data should be delivered to MRC no later than April 30th.
- Earlier submissions are encouraged.
Other Planned Viewablity Processes
MRC is planning additional efforts with IAB, advertising agencies and others to monitor the progress of viewable impression adoption, performance metrics and issues encountered on a longitudinal basis, using a combination of laboratory and campaign monitoring processes, but the specifications for that work are separate and currently under discussion. This is not a dependency for viewable impression adoption and can be considered a normal part of conversion to a new measurement metric.
Appendix: Viewability Vendors Accredited by MRC as of 1/16/15:
For Display Ads: Adloox, AdYapper, Alenty, Chartbeat, comScore vCE, DoubleVerify, GLAM Media, Google ActiveView, Integral Ad Science, Meetrics, Moat, Pixalate, RealVu, Sizmek, Yahoo! APT, WebSpectator.
For Video Ads: DoubleVerify, Meetrics, Moat, Telemetry, Videology.