During this particular webinar we're
going to be learning about the latest trends, we're also going to be looking at technological
advancements and best practices in the field of music programming optimization with data. Let me go around the room by introducing our panelists. First up we have Antoine Baduel, who is the CEO
and owner of Radio FG in Paris, and then we have from Hyperworld, the Head of Research, Alexandre Houget, followed up by Lionel Guiffant, who is the General Manager of RCS
Europe. And we also are joined by
Robert Johansson who is the Head of Programming and Music at Best Radio Programming. We really needed at this time to be able to make the difference between an underground track and a high
potential track. Of course we had discovered many famous French artists with a an international
wide career such as Daft Punk, David Guetta, Bob Sinclair but of course if we identify those
artists it's because we made datas because those datas gave us the conviction that
all those tracks had a much bigger potential than other tracks with the same quality but
for other reasons, maybe because they were released too early, too late, or maybe they were too
underground, they didn't fit with the market. So far at the time when we played the track so
we always had a very close relationship to datas. And of course the situation has changed
a lot because till uh the Y2K or 2010. The only way to have datas around music was the
radio research whereas since the DSP, th
e Digital Streaming Platforms, have been launched,
the music industry now is focusing on data, so we have other datas and thanks [to] the partnership we have with Hyper world we changed a lot the way we ask for all the criteria and the way we do the interpretation change a lot as well so yes since the beginning we've
been working with datas and I think those datas contributed a lot to the marvelous
and international careers that Radio FG co-built with the artists. The datas are very useful fo
r us because it helps us to decide, first of all, for an exclusive track, that the people don't get bored or have the feeling that even if that the track
is not burned that the station is always the same. So it's a criteria of originality and DNA. And on
the other hand of course the lifespan of a track depends on its exposure everywhere especially
on the other stations and the DSPs. And regarding our listening habits we have, with
that format of course, very faithful listeners. Of course, as
faithful listeners, they
expect a lot from the station. That means a regular renewal of the playlist, that means to
risk a lot, to take risk every time when you're playing a track, that means again to do tests, and
panels, and research and datas, so that you optimize your chances to combine
audience, risks and originality. Wonderful and I think that very much encompasses
your your formatting strategy. You are quite exclusive but you've got also your finger on the pulse of what your audienc
e wants by continuously conducting research.
Yes absolutely Andy, our formatting strategy consists in a hand having very precise and serious results
and datas, but on the other hand, sometimes and Hyperworld know how we work sometimes
not follow the datas. Because the datas are tools but they are not our mind, I mean it's like the
AI, it's a tool, it's an assistance, it's support, but it's never your mind and your
conviction. And your conviction that sometimes you can have high speed tracks
but
you can also have diesel tracks. They start slowly but when they take off they are very
strong and that's the purpose because you need also to have an artistic consideration.
If you play hot tracks of your format, that high format tracks, they're in your
central format, it's much easier to have them tested quickly rather than other
tracks who will sound more underground. So our formatting strategy consists in respecting those two-thirds of exclusive tracks and two-thirds of new releases,
. That's the concept. But on the other
hand it consists in really forcing our convictions letting time to several tracks even if the tests
are very difficult. Because it's, I mean, you have successes but you also have tracks that are not
in the blockbusters but we call it in French "succès d'estime". "Succès d'estime" means it's not a success but the professionals, the audience, the fans, the followers, fully respect the track, and it's the first step for
for a follow-up that could be very s
uccessful. I love that because that still speaks to the
human element right still being necessary in combination with the data. So that's quite
an important consideration especially when it comes to music and really engaging and
retaining fans. It's sometimes too mainstream. So it's the obstacle first thing
is when it sounds too commercial the datas are here to give you the conviction that no
your fan base loves the track even if it sounds a little bit too commercial because when you're
wi
th an underground DNA you always need to keep your image but also following the tracks and the
artists you've been always supporting. That's the first point. The other point is always make a
balance between your own convictions, even if sometimes you're right. but when you right too
early, you're wrong. The datas are an amazing tool, an opportunity for us to have us play the right tracks, at the right time, at the right schedule, and clocks, in the day. It's an amazing way to confirm with int
erviews, and the presence of the artists, our convictions and
our decisions in the playlist. I mean there's no playlist if there's no editorial environments.
And it's very important for the listeners, for the station, but also for the relationship we
establish with the artists, to have that talk show, that conversation, that makes the
thing easier. And sometimes. I need to tell you that, some tracks that have very difficult
results in the datas when the artist comes to the station and prese
nts his tracks, does a live mix, and when I interview him and I see the conviction, sometimes I retest the track
just because, I mean there's something, and sometimes after an interview the
test can have changed a little bit or have changed significantly because the
interview and the presence of the artist has the thing go quicker and that's pretty important. And Hyperworld is a media marketing research institute that is specializing in audio studies
for radio operators such as Radio FG an
d the motto at Hyperworld is "smart research for your audio". This is the expertise, the belief, but also the way you feel music, that will make the difference. The
way you read the datas will help. But of course, this is still talent, and human which is the main
added value in the process. The whole process is online, from data collection to delivery. We recruit
respondents into major international access panels We have agreements with that panels and
we ask them, first, to interrogate som
e people who get some specific profiles and that radio station
wants to interrogate so we qualify them through an online questionnaire, and if they fit the profiles
we are looking for, then we ask them to listen every songs not the whole songs, but maybe 15 seconds of each songs, and then to evaluate it. Do they like it? a lot? Have they heard it too much? or maybe they never liked it. So we want to have informations, some feedbacks from them and at the end of the process ee gather all the dat
as together and we deliver datas on online platforms with Excel exportation
and also possible integration to RCS Selector or MusicMasters. We are the "guardians of quality"
when it comes to musical research. Since you always have to apply some golden rules like neutrality, consistency, security. Of course you can ask directly people what they
think about music, but you have to make sure that they involve into the study, that they can listen
to your material, that they take time to sincerely
answer your question - that sometimes they lie
to you, so you have to be able to detect it - and you have also you have to pay attention that this
is not always the same people who give their opinion because otherwise you will make a decisions based on only few people. You have to control the datas through all the process,
making sure that you are building something solid and strong, because the main point is to deliver
valuable insights to your clients because this is your clients have to t
rust 100 percent the datas
is reading, because it will take decision based on it, so it has to be something very strong. And
this is what we're intending to do at Hyperworld. Valuable quality insights, I love that. The main interest of musical research, and specifically when we talk about call outs, call online, is to
follow the life cycle of one title week after week in order to see how different indicators evolve.
Is passion, for example, passion is the percentage of respondents who will t
ell you that they really
like the song, for example, is passion growing up? week after week? is passion growing up along with
burnout? is burnout stable? is rejection very high at the beginning and very
low at the end? So this is the type of indicators we deliver to our to our clients and with this
data they can adjust their musical programming But of course we can deliver many other
indicators, but the basic ones are passion, rejection and burnout, which are the easiest
indicators to unde
rstand. It's a combination of speed, efficiency and I guess a legacy so something to look back on and best make sound results for your music programming strategy then. The metadata is crucial for programming. So once again that speed, that efficiency, that
legacy and of course I think the word that ties this all together is Quality. You can now import data from so many things you know, as research, but also, of course, on the stream, how much people you have at the beginning of the song, how mu
ch people you have at the end of
the song. Of course it's not very qualified but just gives you some idea about what happened on
your station. So you know the quality of the data, how they are produced, and it's very important. We know the story, how the song lives during two months, three months, six months...
But the decision is which song you put into that so it's where data is really key because it's giving
you the decision, the final decision and it's really why it's so important now. Whe
n you talked about the good and the bad datas, they are the good and the bad questions.
Sometimes the questions you ask but you know the answer, so this is a bad question. And all the other datas also need to be considered Because, at the end, this is the balance, is the
one who made by the program director, and it's very important. Because if you only
focus on one kind of research, one kind of data, of course you will really consider focus on the
target group. It's a good blend of data, musi
c, and artistic intelligence, that are the markers of
the industrialization at scale of customizable programming and the hyper distribution
across multiple platforms.
Comments