Are the people who take those tests representative of a broader population? I'm curious. I've taken those tests and I'm sure my tastes are more eclectic than a lot of the population.
-
Get involved.
We want your input!
Apply for Membership and join the conversations about everything related to broadcasting.
After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
If you use a disposable or false email address, your registration will be rejected.
After your membership is approved, please take a minute to tell us a little bit about yourself.
https://www.radiodiscussions.com/forums/introduce-yourself.1088/
Thanks in advance and have fun!
RadioDiscussions Administrators
KROQ
- Thread starter CassChopper96
- Start date
- Status
Music tests usually involve the target demographic. They randomly select people in the demo. Offer to pay them something to participate. They then have them rate songs by listening to samples. They used to do these in auditoriums. Someone from the station would be gathering the data in the background. Probably done online now since it's much easier to do. As far as I know, they don't always tell the participants what station it's for but some can probably guess based on the music being tested. At least that's how they used to do it.
Some stations also do the online music surveys that anybody can sign up for. They send out surveys every so often. Same general thing as far as how it works. Sample of a song, ask for input. Not sure how much of that data influences the programming. I know some used to test the entire library once a year and enter people into a drawing if they took the entire survey of 100 or more songs. Some might still do it.
What I see on Mediabase: Four songs in heavy, which is 50 spins a week, 5 in medium, which is 25 spins a week, and then 7 in light, which is 15 a week. There are maybe ten songs that get 1 spin a day, mostly in overnights. The evening show seems to be the place to play currents. Outside of the four heavies, the rest of the currents get next to no spins in morning drive. So I can see how you reach those conclusions.
The "recruit specs" for a test are determined based on a careful analysis of a station's needs.
In the case of a sick or failing station, the recruit might be based on both current listeners, past listeners and listeners to the direct competitor(s). Or, in lie of listening habits, they might play hook pods of different styles within a format, and to be invited to participate in the test, they would have to "like" two out of three or three out of five sample pods.
Such pods might represent eras, styles or some other aspect of a broad view of a format.
Management will determine the age spread (generally a 15 to 20 year wide "supercore"), gender percentages, weekly radio usage and other criteria.
A recruiting firm will be hired by the research company to find participants. They are paid per acceptable recruit. And the participants are paid, on average, about $100 (but that varies by market).
In the past, such a test might take about 2 1/2 hours at a meeting room or other venue. Today, most are done, after qualification, online with a credential and password. The usual is 500 to 600 songs, but in some formats it can be over 1000.
It's not random. Each participant is qualified on age, gender and radio usage and may have to "like" sample music pods that determine if they are partisans of the general music genre of a station or a proposed new format. The recruiter generally gets upwards of $100 per successful recruit, and more if the recruit is difficult or complicated.
I know of very few instances where the station does the test. It's either a research company or a division of a large group. Station personnel are kept away from any contact with the participants, although in in-person tests sometimes a station personality will come in afterwards and thank the listeners for helping make "out station" better and, maybe, have a prize drawing or give away T-shirts or stuff like that.
$100 and up is generally given to participants, whether in person or online.
They almost never tell the participant what station is doing the test. I've done hundreds of tests... close to 1,000 in fact, and never, ever revealed who the test was for until, in a few cases, the end of the test.
Online testing for currents is generally based on a similar recruit system where particpants are pre-qualified and then asked to score 25 to 30 hooks. Often a successful recruit is asked to "join a panel" and called back a number of times over perhaps a two to three month period.
Call-in tests are not valid. Some stations do them for PR value as "that's my station. I vote for the songs" but they don't use the tainted results.
Again, no valid research is call-in; it is tainted, biased and not representative. Listeners or potential listeners are recruited to a tight set of specifications and the name of the station is not revealed.
Stations may email a listener database member group to "sample a song and vote" but that is PR, not research. It helps bond listeners with the station. Once I did a parallel music test on the air (the real test was 100 people in purpose) and we ran all the hooks on a Sunday evening and listeners got the scorecard in a local newspaper that had about 1.1 million circulation or on forms given at a convenience store chain. Over 100,000 were received back. We tabulated about 500 at random, and they matched the real test. We did not use that "promotional test" data for anything.
Callout might be done for currents on 10 to 15 day cycles using a rolling average and it is ongoing. Library tests can be done several times a year, quarterly if the station can afford it (about $20 k to $35 k per test) and are 500-600 songs, some perceptual questions, radio and Internet usage questions, and even some morning show condensed bits to score. Usually there are about 100 persons in library test.
Last edited:
I've done that before and several stations have that on their website.
The online listener "Music Panel" survey is two fold, testing new music that is coming out (AAA's and Alternative are big on this) and creating listener involvement with the station, making listeners feel like they have a say in what music will be played. It builds listener loyalty.
If anyone remembers the Jelli radio experiment in Las Vegas ( all requested music done online) you can see why you need quality controls in research, or one listener can hijack the whole thing.
You can't test "new music" until the average listener has heard it about 7 or 8 times minimum. That means that, given general new music rotations, you can't start testing for about two weeks and won't have a complete test for over 3 weeks after first playing a new song.
Very, very true.
And those "website" invitations do not produce valid results. I know of no successful station that even tabulates such results. It's strictly done for listener involvement and loyalty building.
K.M. Richards
Program Director, The Eighties Channel™
Right here, watching you all fall all over yourselves trying to prove you know more than the radio professionals you derided in the post I replied to originally.
The Stones still hang on, though it helps that they never stopped recording. They had a minor AAA hit in 2020…
So it sounds like KROQ is basically returning to its pre-2020 playlist of a heavily gold-based mix of Rage Against the Machine, Red Hot Chili Peppers, Nirvana, Foo Fighters, Muse, etc. The kind of burnt-to-a-crisp lineup of the same old/same old that seemingly got them into trouble. Yet the attempt to freshen things up and get more adventurous didn't work either... so I guess, better the devil you know?
Why do you feel that you’re being personally attacked when radio pros, who have actually worked in this business, point out fallacies and note how things actually work in this business?
Whenever I’ve been proven wrong and corrected on here, I don’t take it as a personal attack. I learn and grow from it. Even more so, it helps when you work in radio to see how their points are proven true.
Each listener knows what they, individually, like and dislike. Only trained researchers and radio veterans can talk to a group of such people and find the common likes and dislikes.
If the listeners don't know the "new music surveys" are only done to give them a sense of influence or "belonging," then they'll never suspect that they're being treated with disdain. Most "normal" listeners don't care how the sausage is made, anyway; they just want to enjoy the meal -- which, in this case, is the music their favorite stations are playing. And if they happen to hear in the near future some of those songs that were on the "survey" they just took -- which they definitely will, since those stations always planned to add some of them anyway -- they'll never have reason to suspect that they've been had at all.
Asking listeners for their opinions isn't "phony and dishonest." But using those polls as real music research is statistically invalid. Why? Because you can't get accurate demographic results for the people taking part. Research is only useful if it polls your target demo. If in fact it reaches the wrong people, then the research is fake. People can and often do lie on online polls. They use them to drive up the results on certain songs or groups they like. So they are statistically invalid. It's like using listener requests as an indication of what the audience likes. All the requests measure is the people who call, it ignores the much larger group who don't.
This is not to say stations don't look at the results of those polls, and then post them so people can see the results. That's a good use for that information.
No way they would be competing against themselves if they did a jackish type station.
Back in the 70's, Top 40 stations KRIZ and KRUX in Phoenix used to put out printed handouts every week at the local record stores. On one side was the billboard top 40 list, and the other side was the local Phoenix top 40 hits that week, which were the same songs but in a slightly different order. That supposedly was based on station listener requests, and local record sales.
I realized later this was just an advertising ploy for the competing stations. No one ever seriously tabulated those up every week. If they did, it would be inaccurate, using call in listener requests.
I realized later this was just an advertising ploy for the competing stations. No one ever seriously tabulated those up every week. If they did, it would be inaccurate, using call in listener requests.
It depends. WABC used to call record stores. Of course now there aren't any record stores. But they were once good barometers of what was selling locally.
There is a lot of music research going on. Record labels do their own research to convince radio stations to add their songs. Publicists use research to build word of mouth. Promoters use research to get people to attend concerts. Everyone wants to know what the listeners like. It's not just radio research.
- Status