MRG Tools of the Trade: RADIO Paul Kennedy, RAJAR Donna Burns, Radiocentre
TV viewing Reading Cinema Outdooring
Radio Listening
measures live audiences for the UK radio market
at a glance stands for RAdio Joint Audience Research established 1992 is the UK radio industry s official audience measurement system is jointly owned by the RadioCentre and the BBC includes representation from agencies (IPA) and advertisers (ISBA) contractors are Ipsos MORI and RSMB
some key facts measures over 320 UK radio stations reports on 100,000+ participants per year (c.25,000 quarterly) results published monthly (internal) & quarterly (public) data used by all stations for programming and by commercial radio for advertising sales only captures radio listening that is broadcast or streamed live
RAJAR s MIDAS measures ALL forms of audio & video because we need to understand how radio fits into people s lifestyles and media choices
MIDAS Measurement of Internet-Delivered Audio Services Conducted by Ipsos online completion what people listen to or watch every day for a week when are they doing it? which device(s) are they using? where are they? who are they with? and what else are they doing at the time? reports quarterly 2,500 sample per wave
jamjar measures the audio habits of 9-14 s conducted by Other Lines of Enquiry (OLE) online survey 600 9-14 s (equal splits by gender and age) what devices do they own/use what do they listen to how important are music and radio to them
RAJAR helps advertisers at all stages of the media process 1. Strategy 2. Implementation 3. Review
1. STRATEGY why consider radio?
Communication objectives Brand building Long term Activation Short term Source: The Long and the Short of It IPA Databank
Communication objectives Reach Emotion Fame Brand building Long term Activation Short term Source: The Long and the Short of It IPA Databank
Communication objectives Brand building Long term Targeting Proximity Cost/lead time Activation Short term Source: The Long and the Short of It IPA Databank
Which media? Personal media habits Influencers of media choice
Radio s challenge = adland is an unrepresentative sample
This means that radio suffers from Low salience Poor perceptions Inherited (bad) practice
Which media? Personal media habits Influencers of media choice Data
Record commercial listening: 35m adults tune in every week 67% weekly reach of 15-24s
Live radio is the most consumed audio format Online clips, 3% Catch up radio & podcasts, 4% Music streaming, 8% Owned audio, 12% Live Radio, 73% Source- RAJAR MIDAS, Winter 17
Live Radio remains stable, whilst streaming grows at expense of owned audio Share of adult total weekly hours (%) Spring 15 Spring 16 Spring 17 Live Radio Owned audio Streamed music services Online music/audio clips Catch up radio & podcasts Source: RAJAR MIDAS
Source- RAJAR MIDAS, Winter 17, Commercial radio, vs non BBC podcasts & catch up radio and non premium subscriptions to Streaming services (e.g. inc ads) Radio dominates commercial pure-play audio All adults Live commercial Radio, 88% On demand commercial audio, 12%
4.00-4.30am 5.00-5.30am 6.00-6.15am 6.30-6.45am 7.00-7.15am 7.30-7.45am 8.00-8.15am 8.30-8.45am 9.00-9.15am 9.30-9.45am 10.00-10.15am 10.30-10.45am 11.00-11.15am 11.30-11.45am 12.00-12.15pm 12.30-12.45pm 1.00-1.15pm 1.30-1.45pm 2.00-2.15pm 2.30-2.45pm 3.00-3.15pm 3.30-3.45pm 4.00-4.15pm 4.30-4.45pm 5.00-5.15pm 5.30-5.45pm 6.00-6.15pm 6.30-6.45pm 7.00-7.15pm 7.30-7.45pm 8.00-8.15pm 8.30-8.45pm 9.00-9.15pm 9.30-9.45pm 10.00-10.15pm 10.30-10.45pm 11.00-11.15pm 11.30-11.45pm 12.00-12.30am 1.00-1.30am 2.00-2.30am 3.00-3.30am Radio has highest reach when the shops are open 25000 20000 15000 10000 5000 0 Live Radio TV Digital Tracks CDs Listen Again Podcasts OMS Source- RAJAR MIDAS
RAJAR helps advertisers at all stages of the media process 1. Strategy 2. Implementation 3. Review
How the survey works - sampling framework Currently 320 stations on the Survey BBC National BBC Regional BBC Local National Commercial Local Commercial Reporting periods Q Quarterly (National Stations except Digital only) H Half Yearly ( Regional, National Digital, some local Stations) Y Yearly (Local Stations) Sample sizes vary* smallest sample = 500 adults 15+ (55 thousand adults per year) largest sample = 25,000 adults 15+ (55 million adults per quarter) Survey Approximately 100,000 adult respondents per year 15+ Continuous survey (excluding Xmas break) Survey in field 50 weeks/year = 2,000+ respondents/week *Sample design is the best possible to accommodate small and big stations on the same measurement system
Choosing the sample points 25% 25% 25% 25%
Sample points chosen to represent the station s area in terms of population density geographical dispersion overlaps with other stations socio-demographic profile ethnic profile The number of sample points is proportional to the number of people in the area and the number of stations available locally
Next we select the households random start address if unsuccessful try next door if successful skip 3 households keep going until quota achieved
Then we choose the people Quotas on gender age household size working status ethnic origin All designed to reflect the profile of the sampled area
five metrics for radio measurement what station? when? how long for? where were you? which platform did you use? (AM/FM, DAB, DTV, online/app)
The participants choose how to record their listening Mobile app Online (desktop/laptop) Paper booklet
They select their stations
They enter their listening
The key statistics Reach = the number in thousands or as a percentage of the UK/area adult population (15+) who listen to a station for at least 5 minutes in an average week Total Hours The overall number of of hours of adult (15+) listening to a station in the UK/area in an average week Market Share = Share in TSA (Total Survey Area) The percentage of total listening time accounted for by a station in the UK/area in an average week Average Hours per Head = The total hours of listening averaged across the total adult population of the UK/area Average Hours per Listener = The total Hours of listening averaged across all of those listening to the station in the UK/area for at least 5 minutes
Publication
Using RAJAR to construct the schedule
Station selection - region
Station selection - reach All Women 25 44 (Midlands) Station Population 000s Reach 000s Reach Percent Average Hours Per Listener Total Hours 000s Hours Index Free Radio FM (Shropshire) (was Beacon) 52.3 23.3 44.5 8.9 206.8 119.4 Heart Cambridgeshire 141.2 63.7 45.1 7.9 503.2 113.9 Capital East Midlands 348.2 130.9 37.6 7.0 912.4 108.2 Gem 106 (East Midlands) 369.1 98.7 26.7 8.6 848.3 104.9 Touch FM Staffs 40.3 7.5 18.5 9.0 66.9 104.3 Capital Birmingham 370.9 118.3 31.9 5.5 654.7 98.8 Free Radio FM (Herefordshire & Worcestershire) 64.9 24.4 37.6 6.9 167.9 94.5 Signal 107 152.8 11.9 7.8 6.5 77.2 93.7 106 JACKfm (Oxford) 83.5 15.4 18.5 4.9 75.4 92.6 Trax FM 55.7 13.0 23.4 8.3 108.6 92.2 Rugby FM 12.7 3.9 30.5 7.7 29.6 88.9 Signal One 115.8 62.2 53.7 9.6 595.4 88.1 Oak FM 47.7 4.7 9.8 2.2 10.3 85.5 Heart West Midlands 602.4 179.4 29.8 6.0 1080.6 85.4 Connect FM (was Connect FM and Lite 106.8FM) 75.6 11.0 14.5 5.4 59.5 74.1 Peak 107 FM 59.5 19.1 32.0 7.1 134.5 72.3 Free Radio FM (Birmingham & Black Country) 423.8 76.2 18.0 5.0 381.0 70.1 Free Radio FM (Coventry & Warwickshire) 109.7 22.9 20.9 4.2 96.4 68.7 Free Radio 80s (Coventry & Warwickshire) 109.7 5.1 4.7 4.5 22.9 63.5 Smooth Radio West Midlands 602.4 57.8 9.6 6.6 379.6 61.3 Source: RAJAR
Station selection - efficiency All Women 25 44 (Midlands) Station Population 000s Reach 000s Reach Percent Average Hours Per Listener Total Hours 000s Hours Index Free Radio FM (Shropshire) (was Beacon) 52.3 23.3 44.5 8.9 206.8 119.4 Heart Cambridgeshire 141.2 63.7 45.1 7.9 503.2 113.9 Capital East Midlands 348.2 130.9 37.6 7.0 912.4 108.2 Gem 106 (East Midlands) 369.1 98.7 26.7 8.6 848.3 104.9 Touch FM Staffs 40.3 7.5 18.5 9.0 66.9 104.3 Capital Birmingham 370.9 118.3 31.9 5.5 654.7 98.8 Free Radio FM (Herefordshire & Worcestershire) 64.9 24.4 37.6 6.9 167.9 94.5 Signal 107 152.8 11.9 7.8 6.5 77.2 93.7 106 JACKfm (Oxford) 83.5 15.4 18.5 4.9 75.4 92.6 Trax FM 55.7 13.0 23.4 8.3 108.6 92.2 Rugby FM 12.7 3.9 30.5 7.7 29.6 88.9 Signal One 115.8 62.2 53.7 9.6 595.4 88.1 Oak FM 47.7 4.7 9.8 2.2 10.3 85.5 Heart West Midlands 602.4 179.4 29.8 6.0 1080.6 85.4 Connect FM (was Connect FM and Lite 106.8FM) 75.6 11.0 14.5 5.4 59.5 74.1 Peak 107 FM 59.5 19.1 32.0 7.1 134.5 72.3 Free Radio FM (Birmingham & Black Country) 423.8 76.2 18.0 5.0 381.0 70.1 Free Radio FM (Coventry & Warwickshire) 109.7 22.9 20.9 4.2 96.4 68.7 Free Radio 80s (Coventry & Warwickshire) 109.7 5.1 4.7 4.5 22.9 63.5 Smooth Radio West Midlands 602.4 57.8 9.6 6.6 379.6 61.3 Source: RAJAR
Daypart selection - efficiency Illustration: all commercial radio 25-44 women 150 140 Sunday Saturday Weekday (av) 130 120 110 100 90 80 70 60 50 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 23.00 24.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Source: RAJAR
RAJAR helps advertisers at all stages of the media process 1. Strategy 2. Implementation 3. Review
Post campaign analysis
Radio ad effectiveness analysis Source: radiogauge
RAJAR helps advertisers at all stages of the media process 1. Strategy 2. Implementation 3. Review
Radiocentre effectiveness research +15% Ad Awareness +9% Sales +52% Online Response 7.70 ROI +5% Mental Availability
Best practice learning Weekly Reach Share of Budget Consistent Creativity Use of Brand Music
Helpful resource: Radiocentre planning tools
The future of radio measurement Transmission (radio) Distribution (audio)
Options for Measuring Radio Audiences? -Diary (UK, most of US, Japan, Mexico, Australia, South Africa, Netherlands, Belgium, Finland ) -Telephone Recall (Germany, Italy, Austria, Sweden, Poland) -Electronic Measurement (EM) (partially used in US, Scandinavia, Canada 100% in Switzerland, Iceland) -Streaming
Electronic Measurement (also known as EM) Audiometers hear what the ear hears. 2 different ways Encoding - a watermarked signal/inaudible code the human ear cannot discern but the audiometer picks up. Electronic personal meters or software smartphone apps (e.g Mediacell, PPM) Audio Matching records exposed sounds including radio, several times a minute, then matched to a computer which determines what you are probably listening to (e.g MediaWatch) Potential Issues Compliance remembering to carry it at all times. Expense hardware, software, encoders, panel churn
When RAJAR tested EM, some of the data was accurate, but UNIVERSE SAMPLE LISTENING
Problem with Diary Problem with EM
Streaming/Web Analytics Computers good at measuring computers, but not so good for people a stream doesn t mean anyone is listening or it could mean lots of people are listening one person can be several streams (and IP addresses) across a day but we don t know who they are Online people are seriously different from offline people, so we cannot use IP-based data to model traditional over-air listening behaviour
2 critical work-streams for RAJAR We can t ignore the fact that Electronic Measurement and IP- based data offer advantages. EM gives granularity + data stability if used in a panel format Latest EM systems are software-based - people already own their own hardware (makes it affordable and potentially delivers huge samples) IP data is immediate and comprehensive If RAJAR can harness these advantages we can enhance our measurement service significantly
2 challenges for RAJAR in 2018 1. Uncover the systemic differences between the current system and EM to determine whether they can be reconciled 2. Work out how to convert streams into people (ideally real people) Both projects currently underway
For more information For RAJAR information on the go, including top-line data, trend analysis, maps and more. www.rajar.co.uk For more information on the benefits of advertising on radio, including case studies and research. www.radiocentre.org
Thank you for listening