Who's Doing Good Radio In Boston?

[davideduardo]

Looking at the artist list from WBOS, I don't think anyone is testing the music. It's a hodge podge of fringe hair bands and other very marginal music. No wonder they're doing so poorly. They need a playlist doctor to go in and clean things up.

No, they need to call a research company and test the music against the perceived target audience.

On the other hand, they may not even know what their target is, so they are throwing turds against a wall to see if any of them stick. They may need to do some perceptual research first to see where there is a "need" for a certain music blend that is powerful enough to get listeners to change stations or stop streaming and listening.

 

[davideduardo]

The big change with Michael came with his move to Columbia and his work with Quincy Jones. They came up with music that appealed to everyone and every format. Songs like Beat It appealed to rock stations because of the Eddie Van Halen guitar solo. Rock With Me was pure pop. Don't Stop Til You Get Enough was post-disco dance music. Brilliantly produced records.

Frank Dileo was head of promotion at Columbia, and he went on to manage Michael. Frank was a brilliant music strategist.

And that reinforces your excellent post about how the labels, A&R and promotional activities, determine how a song is going to be introduced to the public.

 

[TheBigA]

And that reinforces your excellent post about how the labels, A&R and promotional activities, determine how a song is going to be introduced to the public.

It's also what is missing today from alternative rock and active rock formats, and could be why those formats are floundering right now. All of the best music people are concentrating on other genres, and the radio in those formats are more concise and therefore more successful.

 

[Least]

False argument. You can't compare the situation 10 years ago with today because what we play today is what people what to hear today. How the song charted a decade or more ago is irrelevant.

You misunderstood my statement. I was making the same point, just said differently. I was saying that your group is speaking to the trend at moment. Your going off the current data, not that you are comparing past to current.

In other words, if I took yours and Big A's posts from even 2016 and compared them to your posts today, an uninformed reader would say that you are (to whatever little to big way) contradicting your past statements. Again, that is uninformed, because you are speaking of surveyed listener interest, which is subjective. Subjectivity is always temporary as opinions and taste changes over time. That was my point.

 

[Least]

There is a lot of mythology about radio ownership before 1996. When you look at radio owners pre-90s, no current radio-only company is bigger than international conglomerates such as GE or RCA or Nationwide Insurance. The biggest group owners in the 70s and 80s were far more top-down in their programming strategies than companies such as Clear Channel. ABC was the biggest group owner before 1996, and they clearly made a decision not to buy more radio stations even though they were allowed to. GE and NBC sold all of their radio properties in 1988. So there really was a change in the type of company that owned radio after 1996.



As I said in my last post, it doesn't appear that WBOS is testing their music. I doubt any kind of science would approve of their current music list. WZLX is far more scientific, and is thus getting better ratings. I don't know anyone at Beasley, but perhaps they're still programming WBOS as a satellite of their station in Detroit. That station is a big success there. Not so much in Boston.

My understanding was that there was cap, which allowed for more owners and supported some of the smaller owners in big markets. But, I don't profess to be the guru of radio history. My observation is based off the various formats that have come and gone from the Boston market, recognizing that there has been a narrowing of most music types since 1996. We seem to mostly be a few forms of Rock, Hip-Hop, Pop, Country, and Sports Talk. However, what I failed to recognize was the independent variable of change in listener preference. That's on me, combined with not knowing some of the information that you shared.

Going to David's post and my reply. Allow me to add some context, your and my back and forth about 90s music being what tests well was from when WBOS was still an alternative station. My point was to highlight that people aren't accounting for change in listener preference as you all do when your statements change over time.

 

[davideduardo]

...why those formats are floundering right now.

When I hear the term "floundering" I think of a dead fish washed up on a beach. In this case, appropriate.

 

[davideduardo]

Subjectivity is always temporary as opinions and taste changes over time.

Excellent statement.

 

[TheBigA]

My understanding was that there was cap, which allowed for more owners and supported some of the smaller owners in big markets.

There WERE more owners, because of the caps, but the big change came in the 80s, starting with Docket 80-90 from the FCC, adding more FM stations to the band, and then the sale of radio by GE, RCA, National Life, and other heritage owners. That left a vacuum in the radio ownership ranks. Everyone wondered who would fill the void, and the answer came in the early 90s.

My observation is based off the various formats that have come and gone from the Boston market, recognizing that there has been a narrowing of most music types since 1996.

Once again, the changes in music are separate from radio. I'd suggest looking beyond Boston, because the major cities are the financial centers. The thought at the FCC that the money from the big cities would help support the less profitable stations in the smaller markets. That may have been the thinking 25 years ago, but most of the big companies are selling off their smaller markets now. iHeart and Cumulus have other ways of making money in small markets without actually owning stations there. If you go to small markets now, you'll see mostly small owners. That doesn't mean you see better music choice or programming in general.

 

[Retro]

The reason is likely that it tests very marginally against one station's target and not at all against the other station's target audience.

It does not matter how "big" a hit a song was when it was fresh. What matters is "How much do you want to hear that song on the radio today?".

Exactly, I do know as much!

 

[Retro]

When we did auditorium library music tests where a group of people all heard the same songs at the same time, we used 8 second hooks. That was proven to do two things: it gave time to identify the song and time to register on a dial or a paper form the score. Those tests could do up to about 600 songs in a session before fatigue set in.

Depending on the market, an incentive of between $60 and $150 would be given; the amount depended on the format and the economy of the market. Albuquerque got the lower amount and Long Island or Boston got the highest.

Recruiting does not identify the station. It is "blind" and people are invited, by phone in the past and by phone or email now, if they meet certain criteria such as age, gender, ethnicity and total use of radio. Then, often, they are played 3 to 5 sets of 3 hooks of songs that are within the scope of the station's target. For example, for a classic hits station, one set of three might include a borderline rock tune and two mainstream ones. Another would have a very light ballad that was very AC. And another would represent the core of the station. To be invited to the test, the candidate had to like at least two of the three (or 3 of the five).

For in-person tests in the past, a local recruiter was used... the same kind of recruiting recruiter gets samples for consumer research on cars and laundry detergents, too.

Callout for currents used about 25 to 30 hooks of the same length and the listener scored with the buttons on their phone. The same hook length was employed. Fatigue set in at about 35 songs, so most of us who did research stayed on the safe side, particularly if there were 18-24's in the sample.

Now we do the tests online, although the recruiting is still done personally and with the same specific targets. People are recruited to participate and some kind money or gift card reward is given.

The current song research uses up to 35 or 40 songs in a session. The recurrent / gold research does as many as 600 songs, but in several sessions, perhaps over a week's time; each research company has their own strategy or "feel" for library testing.

The interesting thing is that as soon as a listener scores a song, they get another hook to judge. The hoods can be as long as one wants, but 90% are scored in 6 seconds or less and 95% are scored in 8 seconds. When research is done on the few cases where over 8 seconds is taken, we find that recruits say, "I love that song and I just wanted to hear more of it".

Nobody needs more than 8 seconds to accurately say whether they know the song, used to like it and are burnt out on it or pick a number on a scale to show how much they like it. Each research company has its own scale. I like a "slider" that is labeled from "I hate it" to "I dislike it" to "neither like nor dislike" to "I like it" and "I love it".

I've done well over 1000 music tests with about 100,000 total persons back in the day when we did them in person at meeting rooms, wedding reception centers, church assembly rooms and even school auditoriums and community centers. Many of those I actually moderated as I was training staff. I've spoken with hundreds of people after the test and at breaks when they wanted to ask questions or make comments.

And that is just for music tests. Callout is bit different, and so are one-one-one or focus group perceptuals as well as "big book" format searches and studies about a whole station and competitors.

That's because national media, even going back to music shows by the Big Bands and crooners in the 30's and 40's were done on networks... radio back then and TV now. The big hits in every format have been national going back to the days of sheet music over 100 years ago.

The mistake here is saying "pick". The radio groups don't "pick" the songs. The listeners do. And generally, they like the same songs. Local differences are influenced by the competition, burn on individual songs due to excessive play in the market and other local differences in competitors, the station´s target ages and the like.

And when we do research, we don't play favorites. We just want to see how much listeners in a very specific target group want to hear songs on the radio today.

I do recall a couple of coworkers tell me that they very specifically participate in a music test from WFNX back in the day, mentioning using the matrix that you have referred to, and eluded to.

While I know that y'all do research, I really do have to question if radio programmer's already somewhat of a preconceived bias, where certain songs by artists are simply never tested, because they have already decided in their mind that a song will not test well at all.

 

[davideduardo]

While I know that y'all do research, I really do have to question if radio programmer's already somewhat of a preconceived bias, where certain songs by artists are simply never tested, because they have already decided in their mind that a song will not test well at all.

Our experience tells us that if we test a library song several times, and it does not pass, it will not improve as years go by.

When preparing a list for a music test, we usually are able to include several hundred "what if" titles. An AC with 250 to 350 library songs might test 500 to 600 titles, trying to see if songs that had burn have recovered or whether songs from popular artists have grown in appeal or for some other reason look better now than they did a while back.

We also look at tests in other markets within our company, check airplay for similar stations in other markets and keep trying to find playable songs. A lot of songs that get played as currents and recurrents just don't endure, so there has to be an ongoing search for usable songs.

 

[davideduardo]

The thought at the FCC that the money from the big cities would help support the less profitable stations in the smaller markets. That may have been the thinking 25 years ago, but most of the big companies are selling off their smaller markets now. iHeart and Cumulus have other ways of making money in small markets without actually owning stations there. If you go to small markets now, you'll see mostly small owners. That doesn't mean you see better music choice or programming in general.

What many people don't know is that 30% of all radio revenue is billed in just the top 10 markets.

A comparison that is even more striking: there are about 30 U.S. stations that each bill more than all 28 stations in market #100, Syracuse, do in total!

 

[TheBigA]

I really do have to question if radio programmer's already somewhat of a preconceived bias, where certain songs by artists are simply never tested, because they have already decided in their mind that a song will not test well at all.

I don't know about anybody else, but I don't get paid for playing my personal favorites on the radio. My personal biases don't matter. At the end of the day, the music is meant to attract an audience that we can sell to advertisers. That's the whole reason for music in the first place.

 

[TheBigA]

What many people don't know is that 30% of all radio revenue is billed in just the top 10 markets.

The other bit of mythology I read is that big corporate radio killed small local owners. Historically that's not correct. Small local owners met up with big investment companies and became big corporate owners. That's the story of Capital Cities. It was a small owner of TV stations in Raleigh. Then it bought ABC. That was the David buying Goliath story. Same with Clear Channel. It was a small family company that met up with Texas investment companies. They disliked all the big New York media companies. So they loaned the Mays family hundreds of millions of dollars to buy radio stations. The small Texas company became a national player, and now owns the highest rated stations in NY and LA.

 

[davideduardo]

I don't know about anybody else, but I don't get paid for playing my personal favorites on the radio. My personal biases don't matter.

The first time I did a music test back in the 70's I decided to give my own scores to the first 100 songs to compare with the results. I was off by more than 25% on over half the songs, and there were quite a few that I liked that did not pass at all.

The station was #1 in its market against more than 30 competitors. We realized we could solidify our position by playing what the listeners liked, not what we liked. The station increased its share and stayed #1 for more than two decades because I learned to divorce my own taste from programming.

 

[TheBigA]

We realized we could solidify our position by playing what the listeners liked, not what we liked. The station increased its share and stayed #1 for more than two decades because I learned to divorce my own taste from programming.

Country star Buck Owens owned radio stations in Phoenix and Bakersfield. He was once asked if he played his own music on his stations. He said, "Hell no! I want to make money with those stations. I make more money playing Garth Brooks than I do playing my own music." BTW Buck's family still owns and programs the station in Bakersfield KUZZ. It's the #1 station in town with an 8 share! Buck taught his son very well.

 

[Retro]

Our experience tells us that if we test a library song several times, and it does not pass, it will not improve as years go by.

When preparing a list for a music test, we usually are able to include several hundred "what if" titles. An AC with 250 to 350 library songs might test 500 to 600 titles, trying to see if songs that had burn have recovered or whether songs from popular artists have grown in appeal or for some other reason look better now than they did a while back.

We also look at tests in other markets within our company, check airplay for similar stations in other markets and keep trying to find playable songs. A lot of songs that get played as currents and recurrents just don't endure, so there has to be an ongoing search for usable songs.

It is the good 'Ole perverbial chicken and the egg. If you do not play it, people will not know it, if people do not know it, then it will not test very well then either.. Side note, do any individual radio stations below either the Top 3 or Top 5 markets actually break any new music or artists anymore at all?

 

[TheBigA]

Side note, do any individual radio stations below either the Top 3 or Top 5 markets actually break any new music or artists anymore at all?

It depends on the format. In the country format, radio stations usually break about 20 or so new artists a year. This year's big discovery is Lainey Wilson. The thing is it takes a lot of work. To get that hit, an artist has to visit every market, and shake a lot of hands. Artists in some genres simply don't want to do the work any more. They just put the music on Spotify and walk away.

Looking at WKLB they're playing a new artist named Frank Ray once a day. They're also playing Callista Clark 3 times a day.

 

[davideduardo]

It is the good 'Ole perverbial chicken and the egg. If you do not play it, people will not know it, if people do not know it, then it will not test very well then either.. Side note, do any individual radio stations below either the Top 3 or Top 5 markets actually break any new music or artists anymore at all?

The fallacy of that point is that you can't test songs until they have been played lots of times. New music is uncertain; were it predictable labels would only issue hits.

After a station plays a song around 100 to 125 times (my estimate based on average TSL times), one can start testing it. By that time there is enough familiarity and the average listener has heard it 5 to about 8 times, overcoming the novelty value of a brand new unfamiliar song and bringing into the realm of possible hits.

Sometimes, after perhaps two weeks of play we will test a new song to get a benchmark, but we won't change rotations or kill a song based on the first test, which is intended to allow further tracking to see if it grows or stagnates... or tanks.

Yes, some stations in smaller markets will try new songs. And stations in larger markets will often the longest while they watch smaller markets test for them. Particularly, if you are the only station in a format in a market, it is a lot safer to wait and see and let others take the first step.

 

[CTListener]

It depends on the format. In the country format, radio stations usually break about 20 or so new artists a year. This year's big discovery is Lainey Wilson. The thing is it takes a lot of work.

And she kept at it even after her first single didn't do much. I'm curious to see how radio reacts to her follow-up to "Things a Man Oughta Know," but that might not come for a while, as most stations are still playing the hell out of her big summer hit here in October.