BARB Explained: The viewability question 18 June 2018 It’s natural advertisers only want to pay for a commercial that has had the opportunity to be seen by a person. This expectation creates a frame of reference for any television or video measurement currency. BARB recently conducted an industry consultation that reinforced this need; advertisers and agencies expect the measurement and reporting of online viewing to follow TV industry conventions. More specifically, the consultation rejected the online industry’s position that a video ad is viewable if at least 50% of the pixels are visible on a digital device for at least two seconds. When we report what has been viewed on a TV set, BARB’s principal convention is that a television programme or commercial has been seen when it is watched at normal speed, with at least one panel member registered as being in the room. Importantly, we deliver duration-based metrics; our television ratings (TVRs) are based on a technique that reports the average number of people present across the duration of the ad. These conventions guide us, as we start to report viewing of online TV delivered on TV sets. They also provide a starting point for how we define what has been seen on tablets, PCs and smartphones. Our reporting of online TV viewing relies on two data sources. Firstly, we collect data from software meters that are installed on tablets and PCs in our nationally-representative panel of UK homes; this provides evidence of who is in front of the screen. Secondly, we collect census-level data from BVOD services that have integrated software code to our specification; this delivers duration-based information on viewing sessions. This census-level dataset enables us to report the number of people watching programmes across four screens; it’s now being developed to deliver an equivalent measure for ads. Combining the census data with our panel measurement will enable delivery of online TV commercial ratings that are based on the average number of people present across the duration of the ad; so that’s just like TVRs. We also need to consider how much of the online TV content is visible on the screen. This isn’t an issue for tablets and smartphones because we don’t collect viewing data if the BVOD service isn’t completely visible on the screen. There is, however, a potential issue when online TV is viewed on PCs; we might collect data when someone opens another application or tab in front of the online TV content. By way of context, PCs accounted in 2017 for 35.5% of online TV viewing on devices other than the TV set. This proportion is declining as people increasingly use tablets and smartphones when they are not watching on a TV set. To deal with this issue, BARB is planning a methodology that incorporates PC visibility factors for each BVOD service that we report. We expect to source these from industry-ratified providers of viewability services; these companies are working with commercial broadcasters to confirm when PC-based viewing of television programmes and ads is in the top window. We are currently developing a timeline for the delivery of multiple-screen viewing data for commercials and will provide an update later in the year.