State of Viewability Transaction 2015 – Frequently Asked Questions

General | Viewability Threshold | Mobile & Video | Programmatic | Make-Good | Guidelines | Impact



Q: What is IAB’s point of view on Viewability transactions?

A: On December 16, 2014, IAB released, State of Viewability Transaction 2015, a position paper, which includes seven principles we recommend that marketers, agencies, and publishers recognize and adhere to. The document can be found at IAB.net/viewability

Q: Where can we find more information on Viewability?

A: The shift from a served impression to a viewable impression is yet another step to greater accountability in digital media, but it is just that—a single step. It is also the foundational step in the journey to greater comparability with other media. This comparability paves the way toward cross-media platform measurement using GRPs that are based on apples-to-apples counts and toward better measurement of ad effectiveness. Such measurements were determined by the industry in the Making Measurement Make Sense (3MS) joint ANA-4As-IAB initiative to be of paramount importance to brand marketers’ abilities to achieve the best media allocations for their goals. See measurementnow.net for more information on 3MS.

Q: What is the relationship between the IAB’s December 2014 position paper and the MRC’s “The Viewable Impression Measurement Guidelines” document?

A: The IAB position paper fully supports the MRC standard. Moreover, the IAB position paper relies on work done by MRC after the standard was published.

back to top

Viewability Threshold

Q: What does the “70% threshold” refer to?

A: 70% refers to the proportion of measurable impressions in a campaign that meet the MRC guideline for viewable impressions, specifically for desktop display ads—50% of pixels must be in view for a minimum of one second, and for desktop video 50% for two seconds. In addition, the standard stipulates that for larger desktop ad units, 30% of pixels in view for one second constitutes a viewable ad. We support, we advocate, and we reiterate our support for the standards that the industry developed under the guidance of the MRC. The 70% threshold, put forth in the position paper, advocates for a 70% threshold across standard viewable impressions in a flight.

Q: How did you arrive at a 70% threshold? When do you expect to raise that percent?

A: The 70% number ties back to the wide variances IAB publisher members see among the viewability measurement vendors. Publishers report variances of 30%-40% which simply means that if the measured media viewability rate falls between 70%-100%, we can bill at the served impression count. We expect the measurement technology to improve over the next 12 months and also expect the variance to be reduced to an industry norm of less than 10% over time.

Q: How would you explain to a client that you are endorsing payment for 30% of activity that doesn’t result in an exposure?

A: IAB is not endorsing payment for activity that does not result in an exposure. We do not want advertisers to pay for non-viewable ads. The bounce rate in measurement is so high that we can’t accept that all of the ads that are designated as not viewable are, in fact, not viewable. This is NOT about asking advertisers to pay for ads that were not viewable; rather, it’s recognizing the short-comings of viewability reporting today. The discrepancies are too large to ignore and publishers will be forced to not charge for ads that are in fact viewable, but not measured as such.

Q: Is the 70% threshold a pit stop on the way to 100%?

A: That is the aspiration and intent. The 70% year-one recommendation is based on data seen at companies that show variability of 30-40%, both across vendors and within a specific vendor. As this variance narrows, so too will the recommended threshold.

Q: Does the 70% apply to all impressions in a campaign?

A: No, only measured impressions. The MRC notes that impressions in a campaign fall into three mutually exclusive buckets: 1) Viewable Impressions, 2) Non-Viewable Served Impressions, and 3) Impressions with Viewable Status Undetermined. The latter bucket is commonly referred to as non-measured impressions.

Q: Why are there non-measured impressions?

A: There are many technical challenges that make it impossible to measure all impressions for viewability. Non-measured units that have been shown to have poor measurability include custom skins, native ads, ads integrated into content, sponsorships, etc., all of which can be highly valued by advertisers. Mobile viewability measurement guidelines have yet to be developed by the industry as well.

Q: Is there a recommendation for what percentage of served impressions should be measured?

A: At this juncture, it is fair of any measurement vendor client to expect measurability rates upwards of 80%. That said, there is no official recommendation. The entire industry aims to get to a point where no impressions remain unmeasurable. We encourage the measurement companies to expand the types of ads that can be measured so that the measurable media bucket grows with the popularity of new ad executions. Until then, it is critical to maintain the distinction between measured and non-measured.

Q: How do we treat non-measured impressions in transactions?

A: Non-measureable impressions should be billed based on the served impression number. Non-measurable are assumed to be viewable since we cannot determine otherwise. We are explicitly not comfortable with applying viewability levels that are extrapolated from the measured impressions. With so many different line items on a plan, and multiple screens across desktop, mobile, and tablet, it’s hard to take the measurable media viewability rate and assign it to other line items that are not measurable.

back to top

Mobile & Video

Q: What is the stance on transacting off of mobile Viewablity?

A: MRC guidance to date and the IAB State of Viewability Transaction 2015 position paper apply to browser based desktop. IAB’s Mobile Marketing Center of Excellence have already spoken with the MRC about their plan to accelerate the development of standards for mobile measurement, and MRC will be overseeing that. Currently the MRC guidance does state in regard to mobile that two points should be noted: 1) “measurers of viewability of mobile browser-based web ads are encouraged to consider these guidelines in measurements until such time as guidance specifically designed for the measurement of viewability in mobile web based ads is created”, and 2) “… ad impressions served in an in-application environment are currently generally assumed to be viewable.” To note, MRC has already stated publicly that this second assumption is likely inaccurate and they will be issuing an amendment. Given the measurement challenges that remain untackled in mobile, it is recommended that mobile remain in the non-measureable bucket until MRC provides specific guidance and vendors become accredited against it.

Q: We are seeing large differences in Viewability vendor’s ability to track video vs. display. Video is often much lower. Does this 70% Viewability threshold apply to both display and video given this?

A: Yes, the 70% threshold applies to both. We have deliberately not differentiated between the two in the IAB State of Viewability Transaction 2015 position paper because we have a MRC standard in place for each. We do believe the 70% threshold works across both.  The difference is likely in a vendors’ ability to measure the impressions—not the ability to determine viewability of impressions in the measured pool. Non-measured impressions should be considered viewable.

Q: If VPAID may be necessary to assess Viewability, how can we expand adoption of the VPAID standard?

A: Adoption of VPAID is an important goal of the IAB. It is felt that the desire to improve video viewable impression measurement, the underpinning of the cross-platform GRP, will provide an important impetus to Video Suite adoption.

back to top


Q: Are these principles for all channels of desktop, e.g., Direct vs RTB?

A: Yes, the principles remain exactly the same, no matter how the transaction occurs, directly or in an automated fashion. Marketers should receive ads that are viewable based on the MRC definitions and, at present, given variances in the data, any campaign that delivers a 70% or higher viewable impression should be billed on served impressions. In practice, there are greater technical challenges in measuring programmatically traded media that need to be worked through.

back to top


Q: If Viewability is measured at the campaign level, what line item should the Viewable make good impressions be served from?

A: This decision should be made on a case by case basis. The IAB State of Viewability Transaction 2015 position paper simply states that “make-good impressions should be both viewable and generally consistent with inventory that was purchased in the original campaign.” Ultimately, this is about achieving the goals that you have promised to your client and this should be a key driver of make-good inventory considerations.

back to top


Q: Is there a recommendation to publishers for the number of ad units that should be available on a given page in order to achieve the highest Viewability measurement?

A: Not at this time, however, we plan to include site design principles in the 2015 3MS Educational Forum series. Please contact [email protected] or [email protected] if you would like to volunteer to showcase learnings and best practices during this session.

Q: What about impressions that are deemed “Viewable” but are actually fraudulent in nature?

A: Viewablity and fraud are completely separate issues, both of which are being aggressively addressed by cross ecosystem consortiums. While 3MS has been in place for several years, the ANA, 4As, and IAB partnered in Q4 2014 to create the Trustworthy Accountability Group (TAG) designed to fight ad fraud, malware, and the piracy of intellectual property head-on, while also increasing marketplace transparency. This jointly controlled body will operationalize principles around the critical issues that face the industry in this arena and will monitor the ecosystem for compliance and develop incentives for broad industry participation and consequences for untrustworthy actors. This has been announced in a series of public announcements over the past several months, and there will be more information about the TAG compliance program coming into the marketplace over the next several months.

Q: Will the IAB also help the industry better understand the impact of other factors, many of which are technical in nature, such as ad server latency, page build priorities, creative file sizes, daisy chain decisioning, etc.?

A: Yes. While it is convenient for buyers to say that below-the-fold ads are the source of non-viewable ads, it is often a result of the ads taking too long to load. While publishers have sped up the content portion of their pages, advertisers have slowed the ad portion with big files and a dizzying number of redirects. As a result the ad slots are often viewable but the user has moved down the page before the ad gets there.

Q: We have already heard that different MRC accredited vendors do not always produce the same results. Will IAB publish best practices for potential vendor discrepancies?

A: We are collaborating with the MRC who are working diligently with accredited vendors to improve measurement and narrow variances. IAB has no plans to separately address this issue.

back to top


Q: Has IAB discussed and made recommendations on who should pay for the measurements against a measured campaign?

A: We have not dealt with the cost. We are aware that in some cases the costs are quite high. Ultimately the market will determine which vendors to use and which ones not to use. We have been very clear in recommending that a buyer and a seller should agree on a single measurement vendor ahead of time—as this will help avoid costly, labor-intensive, error-prone manual processes of reconciling different sets of viewability numbers.

Q: Has any research been done to validate the increased ROI that the Viewability standard has achieved? Are advertisers seeing a higher return from campaigns that are priced out using Viewability currency versus standard impressions?

A: In 2014, early adopters published marketer campaign specific cases showing higher ROI, but there has not been an independent, third party study. The IAB Viewability Data Leadership Advisory Board welcomes any shared studies (send to [email protected])

Q: Will there be an assessment on how this impacts publishers from both revenue impact and technical capabilities?

A: Yes, this will occur continuously throughout 2015. In addition to the MRC and 3MS group, IAB has specific task forces assessing issues around data, sales, and technology. These groups, together with the IAB Tech Lab will continue to monitor the situation within our membership and we are prepared to course-correct as needed throughout this year of transition.

Q: We have heard that some agencies and marketers are asking for and promoting different standards?

A: There is only one “standard,” created by the MRC and supported by the ANA, 4A’s, and IAB. Naturally, in a free market, anyone can negotiate anything. IAB will provide guidance and we urge you to use this as well as the MRC documents to negotiate what you believe is appropriate, reflecting the agreed. Importantly, IAB continues to work hand in hand with marketers via the ANA and agencies via the 4As to find consensus and move the ecosystem forward.

Q: There seem to be more claims than facts in the market today, how can we move forward?

A: We are absolutely committed to using data to answer the outstanding issues, including what the most appropriate variance is. Our members tell us that it is 30%- 40% based on their data (leading to the 70% threshold guidance). We know that the variance is not “0%” which is what a demand of “100% viewability” implies. MRC is the third-party best equipped to gather the data and help the industry coalesce. We urge you to provide data and other evidence to them. MRC is currently in the process of regrouping with vendors to try to work with them and reduce these kinds of inconsistencies in measurement.

back to top