STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 MEASURING ONLINE RATINGS PROGRAMME CONTENT Everything in this brochure may be duplicated and/or made public, but only with the explicit mention of the source: Stichting KijkOnderzoek. STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 INTRODUCTION Ratings research is changing, alongside the Thanks to our new project – integration of the This brochure provides an outline of the first viewing behaviour of the Dutch audience. SKO Online Panel and SKO census data – we phase of our SKO-VIM project. It explains the Video content is increasingly viewed online, can now provide an answer to these questions. value of the online programme content data either live or on demand. Both programmes For the first time, it is possible to measure and that SKO will supply to the market, starting and video commercials are reaching their report the reach and the profiles of programmes from January 1 2016. In the course of 2016, target groups through the digital domain, on and video commercials that are viewed online. we will also start supplying data on online a wide range of devices. Such developments Initially, the project incorporates three video commercials. increase the need for accurate reporting of publishers: NPO, RTL and SBS. SKO is the first online behaviour, next to the existing offline party worldwide that is able to use a hybrid viewing behaviour research. method to provide high quality, detailed Bas de Vos viewing data on a daily basis. Managing Director Stichting KijkOnderzoek Since 2008, SKO has been offering daily reports regarding the number of online This project is the first phase of a larger whole: stream starts of NPO, RTL and (later on) SBS the SKO Video Data Integration Model (SKO- programmes. These stream starts (also know VIM). In the course of 2016, SKO-VIM will result as census data) shed light on total viewing in a new standard: the Video Total. volume, but not on the number of persons watching, nor on the profile of the viewers. 2 STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 SKO-VIM SKO-VIM consists of 2 phases: Goal of the Video Data Integration PHASE 1: PHASE 2: Model (SKO-VIM) is to shed light on MEASURING ONLINE INTEGRATION OF ONLINE total viewing behaviour, including VIEWING BEHAVIOUR NEXT VIEWING BEHAVIOUR AND behaviour on TV sets and internet. TO TV-SET BEHAVIOUR TV-SET BEHAVIOUR In order to reach this goal, we measure both TV set viewing In order to realise a reliable online video By means of fusion, the online video data will be behaviour (in our current TV panel) currency, SKO Online Panel data and SKO integrated with existing ratings. This results in a and online viewing behaviour on Census Data need to be combined. This is cross media Video Total number, allowing us to every kind of device, through an the first step of data integration. report on the total reach of programmes and video online panel. We also use census commercials and to identify the share of ‘offline’ and ‘online’ in total reach. Of course, it will be data of online programmes and possible to make analyses per target group, just online video commercials. like in phase 1. Data integration is conducted by Kantar Below, we offer a short explanation of Phase 1. Media, on behalf of SKO. The SKO website offers more detailed information about SKO-VIM. 3 STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 SKO ONLINE PANEL SKO CENSUS DATA DATA-INTEGRATION: CENSUS DATA AND PANEL DATA In order to measure online viewing Due to the ‘longtail’ that is part and parcel Census data does not concern persons; behaviour, we have established the SKO of the internet, it can be hard to accurately it does not tell us who started a stream. Online Panel. This panel consists of 5000 measure the reach of all online video and We only know that a stream was started on persons of 6 years of age and over. Within online campaigns in the panel. a device and for how long it was allowed to this panel, online viewing behaviour is Programmes with few viewers and online play. By combining census data with panel measured on every online device used video campaigns with few starts, will be data, however, we are able to gain accurate within the household; a total of 9500 hardly visible in the panel. In such situa- insight into the reach and into the profiles devices. Panel management is conduc- tions, census data can be helpful. Census of viewers. This data integration is an ted by TNS Nipo and Kantar Media. data consists of measurement totals of important element in producing online events on an online device; as such, it ratings. sheds light on total volume. In our project, we use census data concerning streams of programmes and video commercials viewed online. We use comScore’s and Kantar Media’s measurement systems to measure census data (and possible other, SKO-certified suppliers). 4 STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 PARTICIPANTS WHAT IS NEW? In Phase 1, we will report on the online video • Online ratings for programmes and (in due course) video programmes and commercials offered by commercials: who is watching online? NPO, RTL and SBS. Later on, other partici- pants will join the effort, such as Discovery • By combining panel data and census data, we are able to more Networks, Disney, VIACOM, FOX International accurately report on the reach and viewing volume in the SKO Online Channels Benelux and Sanoma. Panel. • The first effort worldwide to offer a hybrid method to supply detailed, high quality data on programmes and commercials on a daily basis. In the following, we offer a few examples of reports in order to illustrate the usefulness of the output. The reports are based on actual data, measured in October 2015. The results presented in the brochure are strictly meant for purposes of illustration. 5 STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 WHAT WILL YOU GET? EXAMPLES 6 STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 REACH AND VIEWING TIME Monthly reach October 2015 Currently, The Netherlands is inhabited by 15,8 Female viewers who used the internet to watch million people of 6 years of age and over. 21,5% on or more TV programmes offered by one of the REACH ABSOLUTE of these people watched an online NPO, RTL three publishers currently involved in our online % (000) and/or SBS programme in October 2015. Reach research, registered an average total viewing Total 6+ 21,5 3.408 of online programmes is higher among female time in October of 97,1 minutes. Male viewers Men 6+ 19,3 1.514 viewers than among male viewers; 23,7% realised an average total viewing time of 72,8 Women 6+ 23,7 1.894 Age 6-12 8,7 117 versus 19,3%. Reach within the target groups minutes. Age 13-19 24,4 343 13-19 years, 20-34 years and 35-49 years is Age 20-34 30,9 967 significantly higher than within the target The total average viewing time among online Age 35-49 24,7 912 groups 6-12 years and 50+. viewers of 6-12 years and 35+ is below that of Age 50+ 17,1 1.069 the 6+ viewing total. On average, the online viewer of 6 years of age and over spent a total of 86,3 minutes watching Online viewers of 13-19 years of age realised an TV programmes online. Again, we see a larger average total viewing time in October of 109,4 share in viewing time for female viewers. minutes. Theirs is the highest total average viewing time. Next highest was the average total viewing time of online viewers of 20-34 years of age; they reached a viewing time of 100 minutes. 7 STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 Viewing time in minutes per month (online viewers) Monthly reach (*1.000) online programmes per channel 6+ Total 6+ Total RTL Z Men 6+ SBS 9 Veronica Women 6+ Net 5 Age 6-12 SBS 6 RTL 8 Age 13-19 RTL 7 RTL 5 Age 20-34 RTL 4 Age 35-49 NPO 3 NPO 2 Age 50+ NPO 1 0 20 40 60 80 100 120 0 500 1.000 1.500 2.000 2.500 3.000 3.500 4.000 8 STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 REACH PER PUBLISHER AND DEVICE The total monthly reach in October 2015 among Reach can be differentiated between types of Reach per publisher and device the 6+ audience for online TV programmes offered device used to view content: computer, tablet or by NPO, RTL and SBS was 21,5%. If we zoom in on smart phone (laptops are regarded as ‘computer’.) 100% the individual publishers, we see that NPO had the In over 50% of all cases, a computer was used to highest reach (14,4%), followed by RTL (13,9%) watch programme content online, followed by the 80% and SBS (5,6%). tablet and the smartphone. The computer appears to have been used relatively 60% Reach per publisher more often to watch NPO programmes than RTL or SBS programmes. This may be caused by the fact that NPO did not yet have iOS measurements in 40% October. 5,6% 20% 13,9% NPO RTL SBS 14,4% 0% Total NPO RTL SBS Computer Tablet Smartphone 9 STICHTING KIJKONDERZOEK SKO-VIM PHASE 1 RANKING PER PUBLISHER RTL 6+ CHANNEL DATE LENGTH TITLE KDH% KDH000 KDH000 KDH000 SHARE ONLINE (MIN) ONLINE ONLINE TV ONLINE+TV OF TOTALE KDH Which programmes were most often watched RTL 5 01-10-15 61 Expeditie robinson 2015 1,2 185 1.347 1.532 12% RTL 4 02-10-15 23 Goede tijden slechte tijden 1,0 157 1.792 1.949 8% online in October 2015? Per publisher, we have RTL 5 12-10-15 44 Hollands next top model 0,9 137 602 739 19% made a ranking based upon online ratings.