No. When I say YOU, I specifically mean YOU. If you don't like it, don't listen. You don't pay anything for radio.
I don't control what the papers publish. They publish a lot of things people don't believe.
You don't believe ratings? I don't care.
-
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
The old KRTH
- Thread starter David Shapiro
- Start date
- Status
No. When I say YOU, I specifically mean YOU. If you don't like it, don't listen. You don't pay anything for radio.
I don't control what the papers publish. They publish a lot of things people don't believe.
You don't believe ratings? I don't care.
It helps to understand some basic facts, first, Vinnie...the most important of which is that there is a minimum for incidental listening to be converted into a number. Five minutes. So unless you're walking very slowly past those cubicles, they won't count.
What does count? If your workplace plays, let's say KOST and it's audible at your desk to the point that the portable people meter can get the code, then that counts. Ditto the taqueria where you grab lunch twice a week for 40 minutes that plays KLVE. Ditto the barber shop that has KKGO on. Maybe you didn't tune the radio to those stations, but you were exposed (subliminally or otherwise) to what was coming out of the speaker.
Is it a perfect system? No. How flawed is it? I'm not qualified to say. But is it, as you put it, "capable of measuring anything with any modicum of accuracy, compared to the various Arbitron methods of old"?
Sure.
The Arbitron method was to issue a diary, in which you were supposed to write down all your listening for a week. What almost universally happened was that the day it was supposed to go back, the diary keeper tried to remember what they listened to...."Let's see....I usually listen to KBIG on the way to work, so I'll put that down for my 40-minute morning commute from 7:20 to 8:00 a.m. And I listen to John and Ken on KFI when I drive home from 5:00 p.m. to 5:40 p.m. So that's Monday through Friday." What that leaves out (because the diarykeeper honestly didn't remember) was punching out of KBIG to get the traffic on KNX and sticking around for the headlines and a couple of stories, which added up to 10 minutes, followed by a punch of the button to JACK-FM, which kept him for 15 minutes, then The Sound for 10.
KBIG gets credited for 40 minutes, and the other guys none...when the reality was JACK-FM for 15, KNX and The Sound for 10 each and KBIG for five.
And that's just Monday. There's six more days of fuzzy memory and get-it-done-and-get-it-in-the-mail oversimplification and guesswork to go.
Actually, with the way the diary method rewarded name or call letter recognition, it was skewed in favor of big-name jocks and big ad budgets. If anything, the little guys get an equal playing field under PPM.
And KRTH is still #3.
Facts won't matter. We've been through this many times before, as this 31 page thread proves.
No, it was very clear that BigA referred to you, specifically. We know that in an era of pull media, push media can not please everyone. You are obviously part of that 5% to 6% of folks who don't use radio regularly, and we, as an industry, learned decades and decades ago that it is futile to try to please much of that group.
Few stations use ratings in general public advertising any more. That practice came from the "It's all about us" era of radio. We are now in an "all about you" era, where advertising has to stress what's in it for the listener. Being 100 thousand watts or being #1 has no listener benefit.
And the ratings published in the press are of no value or use to radio stations or advertisers, which is why those beauty contest numbers are given away for free.
Ratings are used to sell advertising. As long as the agencies that use them find them accurate as a metric for pricing, we are all satisfied with the system.
Agencies don't automatically buy the #1 station. They buy the stations that deliver the specific audience they want at the target cost per point (CPP). So an overpriced #1 may be "bought around".
And the differences in most markets between the audience delivery of the top rated stations (and agencies buy rating, not share) is so small that meandering between #1 and #4 or #6 and #9 is actually quite inconsequential at the buying level. In fact, most agencies buy with multi-book averages on their target demographic... which is not 12+.
Any survey or poll can be improved and enhanced. Generally, that means increasing sample or frequency of measurement. And that means added cost for the subscribing stations.
Advertisers wanted, asked for and got radio to pay for exactly what you are describing. They wanted a measure of impressions and they wanted it faster an in a more granular fashion.
Advertisers don't care if you tuned to a station. They care whether you heard their ads.
The diary measured TSL, cume and memory. The PPM measures only cume and TSL. By getting rid of the memory element, we got more precise listening times, and many stations that were secondary choices never got written in the diary.
As to that occasional listening, it generally amounts to one or two quarter hours a week. It contributes nearly nothing to a station's total listening. About 50% of a station's cume accounts for over 90% of the time spent listening, and TSL is one of the two ingredients in share, rating and AQH listening levels. |
Over a 57-year career, I have seen many times over when stations have tried traveling a picturesque back road instead of the superhighway. They always lose to a focused competitor.
Indeed, I have tried it myself, and gotten whacked on the head for it.
When you try to please the small percentage of extreme variety seekers you burn off the huge mass of "comfortable familiarity" seekers. It's a bad trade.
Try building a fantasy football team with rookie unproven players and see how you finish the season.
And imagine the customer who is allergic to strawberry. If they have to eat a scoop of strawberry to get that desired scoop of pralines and cream, they would never go in the store as it would result in them getting sick.
OK, here is where I hit the gong, hard.
WRONG!!!!
I've conducted and participated in nearly 2,000 auditorium music tests over the last several decades. They were professionally recruited by specialized research recruiters in two dozen different markets, and conducted with the most advanced technology (the same dials you see on the cable news networks).
When properly recruited to include listeners to your station, its direct competitors and within the age groups you want to attract, outliers are rare. That is because they are likely not satisfied by either your station or your competitor, so they listen to none of them. But when one does get in because they listen for lack of alternatives, we remove them because they do not reflect the tighter range of opinions voiced by the vastly larger consensus group you need to survive.
Having seen a few hundred outliers out of nearly 200,000 AMT attendees over the years, I can tell you that, in general, they score very low the songs that 99% of the group likes and rank very high some of the "what if" songs that are tested but which the mainstream dislikes or detests.
Remember, it only takes a drop or two of arsenic in the pitcher to poison a whole roomful of people.
Songs have to be tested to see their appeal with your target audience. The only problem I see is momentum. If you have ever dj'ed in a live setting where you see the reactions firsthand, you know that momentum is real. Songs are going to test better if they are part of a cluster of songs that is slowly building that momentum, in fact an ok song that is sandwiched into a set of strong performers will get a much more favorable result that if it were playing among a bunch of other mediocre songs. Same is true for the greatest songs, if you hammer a long set of nothing but the hottest uptempo songs eventually there will be fatigue and that will result in a more negative reaction to a song truly liked.
I've looked over the KRTH playlist for quite some time, 6 hour rotations in this format is a fast turn over but there is no doubt that they are the right songs. I don't know that one could really do the music flow of this station any better than what they are doing at this moment, I really just can't find any major flaws at all.
Now, I do think they could spike a few of the older songs that have that mass appeal among people who weren't even alive when they came out. I'm talking just once or twice an hour, anything more than that and the station will sound dated.
My complaints are with chr, it is too slow at moving hits in and out. If they knew the hits when the passionate part of the demos were starting to like them they could more along the passive listeners, the followers if you will, on a schedule that is closer to that of the passionate music fans. You could bridge that gap so that there would be more of a consensus of when a song is burned. At the very least you have now at least 2 stations doing current hit music in a market, would it hurt to have one that is more aggressive on currents? Now you need to find that diamond in the ruff who can spot the hits early, and more importantly is keen to the first wiff of a new song that is not going to become a true hit. I've always told my pd, yeah so and so isn't going to work out, they say well it is still gaining spins.... then 3 to 4 weeks later they drop the song, that's way too many weeks of playing a dud when you didn't have to.
I've looked over the KRTH playlist for quite some time, 6 hour rotations in this format is a fast turn over but there is no doubt that they are the right songs. I don't know that one could really do the music flow of this station any better than what they are doing at this moment, I really just can't find any major flaws at all.
Now, I do think they could spike a few of the older songs that have that mass appeal among people who weren't even alive when they came out. I'm talking just once or twice an hour, anything more than that and the station will sound dated.
My complaints are with chr, it is too slow at moving hits in and out. If they knew the hits when the passionate part of the demos were starting to like them they could more along the passive listeners, the followers if you will, on a schedule that is closer to that of the passionate music fans. You could bridge that gap so that there would be more of a consensus of when a song is burned. At the very least you have now at least 2 stations doing current hit music in a market, would it hurt to have one that is more aggressive on currents? Now you need to find that diamond in the ruff who can spot the hits early, and more importantly is keen to the first wiff of a new song that is not going to become a true hit. I've always told my pd, yeah so and so isn't going to work out, they say well it is still gaining spins.... then 3 to 4 weeks later they drop the song, that's way too many weeks of playing a dud when you didn't have to.
Let me just add that I heard cbs-fm play Wilson Philips "hold on". NO ONE WANTS TO HEAR THAT SONG!!!!! I get it, they want to move into the 90s and this is a very safe song, but the reality is that once it comes on it is going to make just about everyone turn the station off. They would be so much better off, even with their youngest demos, playing a liked song from the 80s even 70s, yes you need to modernize and move up the average date of your music but playing a hold on says you just went with a date and safety, not thinking of what matters the most, does a majority of my listeners really want to hear this song in 2016?
And that is where the art of programming enters in. Properly setting up MusicMaster or Selector and then massaging every music sweep for the best flow is part of programming. The sequencing is critical so as to create a great pattern considering all the issues of era, tempo, texture and more.
But in radio, if you serve up hateful songs you will get tune out no matter how they are placed.
Hated, not hateful, right? Many a song with mean-spirited lyrics has found and kept an audience over the years.
I thought I was clear, sorry. Yes what you said here really goes without saying. You can set a billion parameters beyond the basic tempo and era and still a good programmer will message the playlist beyond what selector can do with those said parameters. I've been involved in the game long enough to remember the days when it was taboo to play two females back to back, which wasn't as long ago as we would like to admit.
My comments was based on the music test itself and how you can taint the favorability of a song with the songs that it is layered around. For example, and yes a simplistic one for clarity, if you fire off a music test with high scoring favorites one after another, then you play the snipit of a a song that is good not great, it might get lower scores than it deserves as you have set it up to fail by getting the participants in an expectation of hearing songs that are all among their favorites.
I hope that makes sense, if not let me know, I thought it was clear I was talking about testing, the balancing of the music log itself I would have though was such a no brainer that it be obvious that I wasn't talking about that and obviously the testing. I mean is there really anyone posting here that doesn't know and understand that every 15 minute segment of your music log must give a full representation of what you do and offer(outside of specially programming in mostly throw away hours)?
My comments was based on the music test itself and how you can taint the favorability of a song with the songs that it is layered around. For example, and yes a simplistic one for clarity, if you fire off a music test with high scoring favorites one after another, then you play the snipit of a a song that is good not great, it might get lower scores than it deserves as you have set it up to fail by getting the participants in an expectation of hearing songs that are all among their favorites.
I hope that makes sense, if not let me know, I thought it was clear I was talking about testing, the balancing of the music log itself I would have though was such a no brainer that it be obvious that I wasn't talking about that and obviously the testing. I mean is there really anyone posting here that doesn't know and understand that every 15 minute segment of your music log must give a full representation of what you do and offer(outside of specially programming in mostly throw away hours)?
"Hated, not hateful, right? Many a song with mean-spirited lyrics has found and kept an audience over the years."
This is a good point too, beyond the music log, which you would be careful not to play too many downers in a row, I would think that the tone you set by the cluster of songs you sample would also taint the results. If you played a handful of slow tempo or negative sounding songs, then fired off an uptempo fun party song, that fun uptempo party song would get a score much better than it probably deserves and vice versa.
This is a good point too, beyond the music log, which you would be careful not to play too many downers in a row, I would think that the tone you set by the cluster of songs you sample would also taint the results. If you played a handful of slow tempo or negative sounding songs, then fired off an uptempo fun party song, that fun uptempo party song would get a score much better than it probably deserves and vice versa.
By "hateful" I use the meaning based on music test results where the song becomes something like the fabled turd in the punchbowl... truly a hateful object.
Listening is sustained by the overall feel of the music blend.
But individual songs will test in a very narrow range no matter the environment when you ask listeners to score each one on its own. In studies, conducted by placing the same song in several different positions in a test, songs score the same whether next to songs that are better or worse, the same tempo or different, etc.
Also, when the same song is tested as part of a "blend example" pod of 8 to 10 songs it will individually test about the same whether in a high or low testing pod.
It is up to the PD to determine how to blend songs and to set the scheduler rules.
As mentioned before, I've done considerable testing about "position bias" by placing the same song several times in a test surrounded by different songs. The scores invariably are within 1% on a scale of 1-100, no matter where they are. And often they are off by less than 0.1%.
As long as respondents understand they are scoring each song on its own merits, the results are quite valid.
I also use "ringers" which are either out of format songs or totally unknown songs to make sure respondents are paying attention.
A procedure called a replication study is another technique I have used... it is doing two separate tests with identical recruits but with the songs in different order. As long as the sample is big enough, the results will also be very, very consistent.
If I am remembering correctly that was the Los Angeles station that would air the old Elvis Only radio show that was hosted by Jay Gordon until he finally retired the show after a couple of years of repeats.
There's absolutely no such thing as pleasing everyone. You aren't hurt by what you don't play. No non-geek person drives to work and says "I haven't heard Donny Osmond today. Donny Osmond might please somebody and send everyone else running for their presets. There's no need for a "resident geek" to suggest songs no one remembers.
- Status