Tuesday, January 29, 2008

An Answer Worth Reflecting On

Earlier this month, marketer Seth Godin conducted a seminar about trends in marketing. A great summary was posted on the "10e20" blog, including some of the questions addressed to Mr. Godin. During the Q&A session, the following question was asked:

"With all the channels - how do we get noticed?"

I can imagine a program director asking this question.

Mr. Godin's answer:

"That's a pretty selfish question. It's not how do we get noticed. How do we make a product or service that people choose to talk about? Because it helps their life/planet etc. If we make a remarkable product, we will be talked about in tons of different places."

What are we, as broadcasters, doing to make our stations worth being noticed and talked about?

Something to reflect on?

Tuesday, January 15, 2008

It's the Environment!

There is a widespread misconception about what music radio is.

This misconception is pointing us in the wrong direction. I think:

· Music radio is not music discovery.

· Music radio is not music choice.

· Music radio is not music variety.

· Music radio is not hearing songs in the order you would like to hear them in.

Music radio is about creating an environment.

We can’t be an iPod. Guess what. We couldn’t be an 8 track tape, either.

But we’ve got some kind of groupthink thing going on that is turning many music stations into bad iPods. And let us be perfectly clear: We can’t be a good one. Get over it, please.

If we’re not an iPod, then what is the role of music radio?

The role of music radio is to provide an environment.

It is Not About This

It is not about 10,000 choices. It isn’t about the long tail or the short tail, because it isn’t about the tail at all. It is an ecosystem. It is about the tribe. Music is a large part of the tribe’s identity.

It is not about them programming their own songs. People who have time to program their own music, like people who called into radio stations to make requests, just aren’t right. The aliens that abducted them did some damage before they were returned.

It is About This

People want to belong. They want to know what other people know. They want to participate in the human experience.

They share music with their friends. That includes the chance to sing along, be inspired by the lyrics, learn the dances, copy the clothes, go to the concerts, listen to music together, and more. They belong to the tribe.

They learn what is #1 with the tribe. Our challenge is to know what is #1 with the tribe, too. We need to understand the tribe, and maintain credibility.

They learn what the tribe thinks is important. We must know that too, if we are to join the conversation in progress.

People don’t pick their friends because they are exactly like them. No, they are only similar, and the tribe members challenge and change each other. People change their values in order to belong. If we are part of the tribe, we get to be change agents too. They want this human interaction, and they can welcome it from us. Jocks who speak with authority? Yes. But we have to know what we’re doing.

People want to hear the songs that matter. For whatever reason they matter. It is our job to know what they are.

People want to hear from credible people who live in the same world they do. They can’t get that from an iPod. But sadly, they can’t get that from today’s “standard” music station. There is no listener environment.

What We Are Doing Today

We try to eliminate negatives. Shut up the jocks. Honestly, that is understandable. To actually develop an air staff that can stay in close touch with the environment that the listeners crave is very hard. And you can get, to a point, a boost from simply eliminating bad content. After all, the music itself represents untold millions of dollars of effort by teams of professionals.

But you can get that on an iPod.

But we don’t’ really eliminate the negatives.

We interrupt the only product we have left (the negative-free songs) with insanely long commercial breaks. And we air promos that sound like we never got past Star Wars lasers or the guy that voices Monday Night Football.

And sadly, we often put on a contest only because [insert advertiser here] wants to sell them some stuff. We should not be surprised when things like this stretch our credibility, then break it. It used to be that a program director could veto a contest that violated the expectations of the listeners. Now, we act as though the tribe we want to belong to is a tribe of one. The “one” is the advertiser we’re supposed to plan a promotion for.

Jack, when it works, is an environment. And the people for whom it works believe that it is part of their tribe. The rest of the music stations are often so bad that for just that little bit of environment they will put up with all kinds of music that they don’t like. I’ve done the research. Trust me. It isn’t pretty.

But we don’t really work hard even on the music.

And that is the other thing. We don’t even get the “iPod” idea right. We are doing far less research than ever before. The quality of that research is at an all time low. There is no real analysis (I excuse myself on this one).

We learned in 1990 that looking at test scores without any thought to how songs relate to each other is simply wrong. But today, I new generation of programmers are doing it again, because it is easier than discovering the musical part of the station’s essence and then training young program directors so they learn to manage the environment that is being created on-air.

So we’ve given up on human communication, personality, entertainment (beyond the song), information, education, expressing honest emotion, listener involvement, understanding the conversation and participating in it, being extremely topical, accessibility, and training.

And we want to blame it on the iPod.

Two Paths Forward

We had, and I think we still have, two paths.

1. Keep cutting back. Shrink our way to greatness. Tom Peters is wrong.

2. Create an environment for our listeners. This will take talent, training, head count, research, creativity, and it will cost so very much more than what we are doing today. And maybe we can’t add enough value for 50% profit margins. Maybe we can only do 40%.

I’m kidding about the 40%. Because I think Seth Godin is right. Your radio station is the Trojan horse. We suck them into the club (or more truthfully we join their club) and only make 40% doing it. But then we move the 100,000 local listeners off into 100 different directions that appeal to 1000 people each, and connect them to our local community and advertisers in a way that nobody knows how to do well – yet. And we make an extra 40%, and the 80% total makes what is going on in radio today look silly. Then, the on-air product will just be the tip of a big iceberg.

What is ironic is that large radio companies could be in the best position to locate and develop the best ideas. They can elevate the average station’s quality through training and mentoring. They can invest in the future by developing programming ides on HD channels, experimental Internet stations, and tap into their large talent pool for other experiments in audience interaction. Large stations could settle for 40%, and with the other 10% raise the barrier to entry so high that nothing can compete with radio as a companion and a portal to the rest of the listener’s tribe, and a link between that tribe and the local advertisers.

We can be an environment, a way for people to participate in the human experience. By playing the music of the tribe. By joining the conversation of the tribe. But first, we need to understand the tribe.

Saturday, January 05, 2008

The Station that Listened

I punch in the station. I hear Billy Joel's River of Dreams. Then, Boogie Oogie Oogie, by Taste of Honey.

Sound like a lame station?

Well, also know that despite both of those songs having been played so many times, both songs were back announced with title and artist. Then, the DJ talked.

They identify every song.

Sound even goofier?

Okay, one more interesting fact: The number one complaint from listeners is that we don't announce the title and artist of the song. Trust me on that one. I do this for a living.

Yeah, but what do they know? We're the professionals!

Punch line:

This station is WBEB, Philadelphia. They have a 15.8 share. The number two station has a 6.9.

Imagine that. Listen to what people want. Give them what they want. And win. Huge.

It warms my heart and brings a little tear to my eye.

Want to hear the station that wins? I thought so! Here you go:

http://tuner1.dc1.sonixtream.com/solon/media/tuner/Tuner?aff=b101radio&useSame=true&type=IE

Friday, January 04, 2008

Music Test Basics Series: Test One Song Twice

Here's a helpful tip.

At least one song should appear two times in your AMT. Place it in the first half of the test and again in the second.

Why?

First, you'll see how different the two results are. Hopefully, they are similar (be sure and use the same hook).

But the second reason may be even more important.

Doing this can allow us to verify that the data was handled properly. Virtually without fail, a song that is tested twice will cluster most closely with its second test.

I always check. If it doesn't, we know to backtrack and triple check that we didn't make a processing error.

It is a simple and effective internal check.

Music Discovery versus Rediscovering Favorites

I’m spending some more time with Pandora.

This is the custom radio station built just for you using the music genome project.

To me, it seems pretty much silly. They have a huge bias away from familiarity. If there is a hit version of a song, they'll find it recorded by another artist as an album track.

So it is all about new music discovery.

I'm thinking that new music discovery is overrated.

Pandora isn’t about music memories, familiarity and comfort.

Great, for age 12-24. Well, for 30% of 12-24 anyway.

Most of us want to use music to reconnect to something, I think. We like what we like. We want what we like. In small doses, a little challenge might be nice. But every song?

I'm thinking that the role of radio is not usually music discovery. Our role is to create an entertaining and informative companion. For most people, that means familiarity.

Forget My Opinion. What Do Listeners Want?

This raises a question. In all the research we do, have we ever found and then used a measurement that tells us the relative interest in music discovery and exploration versus returning to familiar, comfortable songs?

The only thing I can think of at this moment is the often included,“On a scale from 1-10, where 1 means “PLAYS ALL NEW MUSIC” and 10 means “PLAYS ALL OLD MUSIC”, where would you like your favorite station to be? "

And then we usually ask, "Where on this same scale is [name of station here]?"

Sounds like a pretty weak attempt, I hate to say. How about you? Have you found a survey question or other way to keep your station properly balanced between "discovery" and "favorites"?

If This is Their Music, They Should Love You, Right?

When you do music research you obtain opinions from (usually) between 100 and 200 people.

You don't know in advance, but once the test is done you know:

  • Which people feel similarly about the music you tested.
  • Which people are really a mismatch with the room.

This involves a simple recipe:

  • 1. Identify the person whose taste in music is least like everybody else. This involves a simple comparison of their opinion about each song with that of all the other respondents.
  • 2. Remove that person.
  • 3. Recalculate the test.
  • 4. Repeat until you have a smaller group. Those still in the sample think alike (relative to the people you removed) about music.

There are a number of ways we can use that information, but I'd like to focus on one. And I'll do that by posing a question.

Within the total sample, a certain percentage are your P1 listeners - those who say they listen to you most. Typically in a library music test this will be 50% to 60%.

Think about the people who represent the common thinking about the music, not the independent thinkers we just tossed out of the sample. Consider that you chose these people because you want their input about your music. You chose songs that you feel give you the best body of music from which to construct a playlist.

Should you be more popular with this group?

For example, if your P1s were 50% of the total sample, once you focus in on the people whose taste is defining your format, should you P1 percentage increase, say to 55%?

Yes.

Yes it should. Every time. This is your test, your music, your people. If the people whose taste doesn't match the format actually are more likely to be a P1 to your station, then something is seriously wrong.

We call the people who are left in the sample after we clean out the odd people "Pure Core Format Fans". We call this Pure Core analysis. Only Steve Casey Research does it.

I want you to have this information. Because it works to well as an indicator of your station's health. At least it works well to show whether you are aligned with the format you think you're aligned with.

An Excellent Early Warning Indicator

In a shockingly high 35% to 40% of music tests we see a decrease in P1 conversion rate among the Pure Core format fans. And that is a very dangerous result. Knowing that is happening completely justifies the time and expense of the music test all by itself!

Responses, when the station isn't working better with the Format Fans:

  • In most cases, you'll need to accept the result as a learning experience. You have not centered the station where the true center of the format is. Doing so will cause you to make much more sense to the listeners. Ratings will increase.
  • Modify your demographics. Maybe what you want to do isn't valid for the demographic coalition you're trying to build. For example, the "River" format tries to meld 30-35 females, 36-40 females, and 30-40 males. Finding a combination that works for these demos has proven very challenging. Several research projects have shown the need to eliminate some of these people in order to focus the station enough for it to make sense.
  • Change your music. This doesn't happen often. But it does happen. I've seen AC stations with inexperienced programmers create a music test list that was better suited to a genuine CHR. The result was a list that wasn't basically compelling. The thing that people came together on wasn't the music so much as it was an agreement that the music was wrong. It is almost always possible to avoid this. Show your planned test list to a number of experienced broadcasters. Ask them if it "makes sense" given your format and respondent demographic groups.

Remember that all listeners are not equal. Some really support your format. Proper analysis will tell you whether those people also really support your station. And getting that information from your AMT is something you should insist on. It will give the information you need to increase ratings, have more fun and get a raise.

Thursday, January 03, 2008

A Focus Group With Something to Say

 

In reviewing some notes, I was reminded of a fantastic capability that was added to MusicVISTA some time ago, but which you may have either not heard about, or forgotten:

Identifying “perfect” candidates for focus groups

The problem:

In too many cases, focus groups frustrate us, when the respondents say something like “well, I just don’t like the music”. And we are stuck.

The solution:

Once a person has participated in an AMT, we know whether they “just don’t like the music” or not.

We also know whether you were their favorite radio station.

With that information, we can do the following:

1. We can identify the listeners who were most  “format friendly”. They are people who had very typical responses to the big hits. There are not music issues. They can’t say “I just don’t like the music”. We already know that they do.

2. We can further refine this list to only those people who are not P1 to the station.

3. Start at the top of the list and invite these people to participate in a focus. Since they enjoyed their time and being paid at the AMT, they will almost agree to participate. And yes, you will have to pay them again.

4. Now, we have the “perfect” focus group.

You now have a group of people who because of their tastes in music should be listening most to us, but who have identified themselves as P1 to another station.

Those people have things to say that will be very interesting to you.

Nobody else in the history of our industry has created this kind of AMT respondent analysis. It is unique, it is powerful, and it could help you get the most powerful, ratings increasing listener feedback of your career.

If your station considers doing both library music tests and focus groups in 2008, I hope you’ll let me know, so I can provide you with even better tools.

Ask Them Their Age

Helpful research design hint: When screening a library music test, many stations will ask potential respondents their age in a general way. "Are you 25-29, 30-34 or 35-39 years old, or older?"

They are afraid that having a stranger ask them what their exact age makes some people uncomfortable.

And there is some evidence that this can happen.

But, once they are at the test, filling out the personal information sheet, don't be shy. It is pretty clear you are doing a legitimate survey at that point.

Have them write down their age as an exact number of years. do not ask them to choose from a number of ranges.

Why?

  • Research types like me will use that information in a variety of ways.
  • We may shift the age breaks a bit to better balance the sample.
  • We might divide the ages up in more than one way: For example: 25-29, 30-34 and 35-39, but also 25-32 and 33-39.
  • Some programmers want us to find what might be called the "sweet spot". For example, what is the 5 year group out of a 15 year wide sample that gives the highest scores to the songs?

The value to you is: Research analysts can give you better analysis with precise age data. And once the respondents have begun the test, there is no problem in asking.