listeners are bored

[davideduardo]

Re: A Simple "Yes" would of been sufficent... (eom)

> I am.
>
> But even in a your strong defense of Arbitron, you have to
> revert to statements like "wobble" & "wander".

Wobble is a statistical term. In fact, statistics is the only science where "error" is not a dirty word.

As long as radio can not do a census of every citizen every day, we will use some form of a probability sample (trying to make it a random sample at the same time). Whenever you do a sample of a highly variable universe, you will have a margin of error. Such margin of error is easily calculated for the degree of confidence one wishes in the sample.

Sample size is almost always determined by a trade off of cost and margin of error; what is the biggest one-standard-error-range that is tolerable within the range of meoney we are willing to spend.

Sample error and sample to sample wobble will affect any listening poll, irrespective of the methodology used.

Wobble is not a diary issue. Wobble is not a people meter issue. Wobble is not a telephone coincidental issue. It is a "law" of statistics.
>
> Doesn't it make you the least bit uneasy that our
> "customers" (ie advertisers) are so approving of the system?
> Maybe morning drive spots should be selling for 10x what
> they're going for. At this point, we simply don't know

We do know within a very tight range. Bigger stations seldom wobble by more than around 10% or so, and much of that is due to the competitve array int he market also changing, not the sample. Smaller stations (in terms of share and cume) wobble more as there are few "evidences" of listening, so one or two diaries can affect the results more.

Keep in mind that advertisers, as I said, use multi-book averages to flatten the waves, and they know that any sample has inherent error.

One proof of the validity of Arbitron is that many radio companies do thier onw in house tracking, and find that they see the same numbers (even thougth they only track themselves and direct competitors).

The ability to replicate results proves a sample to be valid. For example, the reason we use 100 person AMTs is because you can replicate exactly 9 out of 10 times with a sample of 80 persons done over and over, so 100 guarantees that the variables of show rate by demo will not ruin the sample. That is righ... you can test your music with 100 people or 1000 and the changes will be insignificant as you make the sample larger and larger.
>
> The similiar results produced by the PPM shouldn't come as
> any surprise to anybody. It's basically the same system,
> produced by the same company, with a diffenent method of
> gathering of data.

No, it is a totally different system.

The diary is a new sample every week, consisting of 8.5% of the total book sample. The PPM has a panel, and it is the same sample every minute, hour, day, week and month. The diary is active, and depends on recall. The PPM is passive, and automatically tracks listening... even phantom listening. The PPM has a near-perfect panel which is proportional on every one of about 20 stratification variables, while the diary has a sample that can vary weekly in make-up, and has to be adjusted for by placement balancing as the first diaries come in, and weighted in each cell to some extent to achieve proportionality.

You could not find tow more different research techniques, no matter who does them.

> It would be pretty embarrasing for them
> to come up with a set of radically different answers.

They knew this was statistically impossible going in. In fact, both methods are MRC approved... and no other method is. The MRC is not going to approve a method that wastes the money of the advertisers and agencies that support it.

> If
> you compare poor results with more poor results that doesn't
> confirm the previous poor results... It only means you got
> more poor results.

You are building a straw man. Nobody questions Arbitron's systems. They do question the delivery time of the diary book, the declines in response rates, the difficulties involving cell phone homes, 18-24 male sampling, etc. But any company using any method is going to have the same problems doing a poll.

Again, there is no "other" method that costs within the same range and that would get MRC approval. Without MRC approval, the users of the data would not emply it... meaning the advertisers.
>
> And just by saying that statistical results are accepted
> does not mean they are necessarily accurate... It only means
> they are accepted.

See above. Any poll will have the same margins of error based on universe and sample size within the universe. Unless you increase the sample by perhaps 20 to 30 times, you will still get the kind of wobble you see today IN ANY SYSTEM of measurement.
>
> I contend that radio will continue to sell itself short up
> until the day they can accurately pin point your listener
> and get inside their head. Unfortunately, I don't have that
> system developed yet and if I did, you can bet I wouldn't be
> posting it on a message board.

Other than having every US resident implanted with a computer chip, there is no other method besides a sample of the universe. And all samples can be seen to have error in proportion to size and the universe being measured.
>
> On the other hand, I guess the possiblity exists that radio
> is over-selling itself...

When I can put a dozen or so people and some software on a computer to work calling a market for a couple of days and get the same ranking of stations Arbitron does, I think we can say that the results are very close to reality... close enough for advertisers not to waste a second of time debating the system.

 

[KMRichards]

Re: A Simple "Yes" would of been sufficent... (eom)

If you're not going to put your message in the message body, at least delete the quoted text.

Not everyone knows that EOM means End Of Message.<P ID="signature">______________


</P>

 

[Seven_Costanza]

Maybe we're arguing apples & oranges here...

I'm not interested in a system that counts "ears". I will agree that Arbitron is probably as good as that type of system will ever get. But I want a system that accurately reflects "effectiveness". I have witnessed too many cases of stations that are very similar in ratings have very different results in getting their listeners to participate in a charity, a promotion or other event.

Ultimately, the ability to get the listener to buy into the messagage is what drives the business, not selling to the advertiser. That should be the industry standard, not how many 25-54 year old males make up those listeners. Billboards and busboards may make a station #1 because they increase station recall, but does that mean you are better broadcaster or simply a better marketer?

In the recent case of the phantom sextuplets, people claimed to have seen the mother breastfeeding the babies in public. Are these the same people that fill out arbitron diaries? You bet they are. These are the same people we rely upon to be the only base of a multi-billion dollar industry.

I simply want a scorecard of station effectiveness. I want to know what works and what doesn't, I want to know why and I want it now. That's not too much to ask for is it?

 

[argytunes]

Re: Maybe we're arguing apples & oranges here...

> Ultimately, the ability to get the listener to buy into the
> messagage is what drives the business, not selling to the
> advertiser. That should be the industry standard, not how
> many 25-54 year old males make up those listeners.
> Billboards and busboards may make a station #1 because they
> increase station recall, but does that mean you are better
> broadcaster or simply a better marketer?

What I'd like to know is HOW MANY ADS & PROMOS can a listener tolerate before he or she tunes the radio station out entirely? You could have the best format in the entire world with the perfect mix of music, good announcers and jingles, terrific contests....but the moment a 3-5 minute spot cluster comes on the radio, how many listeners are ACTUALLY LISTENING TO EVERY PROMO, EVERY SPOT, EVERY ("We've got another 10 in a row coming up in 60 seconds")?

The arbitron diaries have always been a crap shoot. Simply because they're supposed to be filled out by a cross-section of individuals who have no connection with any aspect of radio. The diaries are supposed to represent all types of listeners and all age groups...but they don't!

What would happen if the diaries were taken away and 300 people were paid to spend an afternoon listening to a variety of radio stations or formats. These people (who are being paid to listen) could then offer their opinions and views about the radio stations they heard? How many promos and commercials they actually listened to and remembered? Whether the approach of the format pleased them or ticked them off? Whether 12 songs in a row or a 90 minute music sweep (or non-stop talk) was worth the number of ads and promos to support it?

I'll bet the responses would be VERY SURPRISING? Particularly to a lot of the consultants who are convinced that every single radio market can and should be programmed COMPLETELY THE SAME?

argytunes

 

[ricoontheradio]

Re: Maybe we're arguing apples & oranges here...

> Ultimately, the ability to get the listener to buy into the
> messagage is what drives the business, not selling to the
> advertiser.

The problem is radio is not a direct response medium, so that doesn't really matter. Radio is typically used as a support medium. It's the mass marketing in multiple mediums that create strong branding.

If you want direct response, mail post cards.

 

[davideduardo]

Re: Maybe we're arguing apples & oranges here...

> I'm not interested in a system that counts "ears". I will
> agree that Arbitron is probably as good as that type of
> system will ever get. But I want a system that accurately
> reflects "effectiveness". I have witnessed too many cases
> of stations that are very similar in ratings have very
> different results in getting their listeners to participate
> in a charity, a promotion or other event.

I can only imagine the cost of this. It would be more than radio bills a year. It would require a personality profile of every listener surveyed, and require tracking them all through the day.

Very often it is impossible to research something with too many variables. In this case, advertising is effective only partially due to the medium that transmits or conveys it. Much has to do with the product or service itself, the price for the product, the creative approach, the number of spots, the competitive array in the product's field, etc.
>
> Ultimately, the ability to get the listener to buy into the
> messagage is what drives the business, not selling to the
> advertiser. That should be the industry standard, not how
> many 25-54 year old males make up those listeners.

"Media" is plural for 2medium." And that word has the same root as "intermediary." Radio is a go-between, and there are many, many other consideratins beyond the medium that affect much more the ability of a product to generate sales.

Or, as often said, "the best way to get a bad product off the market is to advertise it." Lack of sales, in this case, indicates a bad product even if it is well promoted.

> Billboards and busboards may make a station #1 because they
> increase station recall, but does that mean you are better
> broadcaster or simply a better marketer?

Outdoor is used to build cume, not recall. Recall comes from listening and remembering to write it down, usually later in the day. Billboards may make it easier to rember, but the product itself generates the diary entries.

I had a station in market 13, where 75 to 79 stations make the book (highest of any market) which never used outdoor, never used TV and never did money contests. It was #1 for 20 years, with the core demos being 18-34 and 25-44. The programming creadted the recall, not promotion.
>
> In the recent case of the phantom sextuplets, people claimed
> to have seen the mother breastfeeding the babies in public.
> Are these the same people that fill out arbitron diaries?
> You bet they are. These are the same people we rely upon to
> be the only base of a multi-billion dollar industry.

If you want to criticize the measurement, look at how newspapers sell: circulation. No idea who read all the pages, no idea if the ads were noticed, no idea if they were acted on. Occasionnally a study is done, but not on an ongoing basis. It is too costly.

Some of the things you ask for are possible, but at a cost that far exceeds the billing of the entire industry. Radio requires frequent measurement... the PPM will issue monthly, in fact. To do a census of listeners in the roughly 250 Arbitron markets 4 times a year would cost $30 billion a year, give or take ten billion.
>
> I simply want a scorecard of station effectiveness. I want
> to know what works and what doesn't, I want to know why and
> I want it now. That's not too much to ask for is it?
>

No, it is not too much. But it is literally impossible. (And I have not even considered the fact that most ad campaigns run in multiple media, making the contribution of each very hard to evaluate in a vacuum as a vacuum does not exist. In vact, radio is usually a support medium for campaigns, not the principal one... ad agencies use radio, tv and others in combination... and go by ratings. Only direct accounts may buy one staiton, but those are not sold on ratings anyway.)

 

[davideduardo]

Re: Maybe we're arguing apples & oranges here...

>
> What I'd like to know is HOW MANY ADS & PROMOS can a
> listener tolerate before he or she tunes the radio station
> out entirely? You could have the best format in the entire
> world with the perfect mix of music, good announcers and
> jingles, terrific contests....but the moment a 3-5 minute
> spot cluster comes on the radio, how many listeners are
> ACTUALLY LISTENING TO EVERY PROMO, EVERY SPOT, EVERY ("We've
> got another 10 in a row coming up in 60 seconds")?

There have been PPM studies done on this. All commercials cause attrition, and I believe there is a presentation under PPM on the Arbitron site about this.

Radio is a commercial medium for 11,500 or the 13,500 stations in the US. Every PD will have slightly different theories on stops and length, and each market's rates will determine hwo many spots have to be sold.
>
> The arbitron diaries have always been a crap shoot. Simply
> because they're supposed to be filled out by a cross-section
> of individuals who have no connection with any aspect of
> radio. The diaries are supposed to represent all types of
> listeners and all age groups...but they don't!

There is no proof of this. In fact, there is proof that the survey is witnin a few percent of being proportional in every demo that we, as subscribers, have collectively asked for over the years.
>
> What would happen if the diaries were taken away and 300
> people were paid to spend an afternoon listening to a
> variety of radio stations or formats. These people (who are
> being paid to listen) could then offer their opinions and
> views about the radio stations they heard?

Would not work, as you have taken the listening out of the normal environment and declared each particpant to be an expert. All kinds of biases enter in.

>
> I'll bet the responses would be VERY SURPRISING?

And useless when measured in a controlled environment.

> Particularly to a lot of the consultants who are convinced
> that every single radio market can and should be programmed
> COMPLETELY THE SAME?

I do not know of a single consultant who believes that.

 

[argytunes]

Re: Maybe we're arguing apples & oranges here...

> >
> > What I'd like to know is HOW MANY ADS & PROMOS can a
> > listener tolerate before he or she tunes the radio station
>
> > out entirely? You could have the best format in the
> entire
> > world with the perfect mix of music, good announcers and
> > jingles, terrific contests....but the moment a 3-5 minute
> > spot cluster comes on the radio, how many listeners are

> Radio is a commercial medium for 11,500 or the 13,500
> stations in the US. Every PD will have slightly different
> theories on stops and length, and each market's rates will
> determine hwo many spots have to be sold.

Maybe the rules need to be changed? Instead of basing a spot break theory on the size of the market....focus on "audience attention and retention" instead! It's the listeners who determine the success or failure of an advertiser...not a program director.
> >
> > The arbitron diaries have always been a crap shoot.
> Simply
> > because they're supposed to be filled out by a
> cross-section
> > of individuals who have no connection with any aspect of
> > radio. The diaries are supposed to represent all types of
>
> > listeners and all age groups...but they don't!
>
> There is no proof of this. In fact, there is proof that the
> survey is witnin a few percent of being proportional in
> every demo that we, as subscribers, have collectively asked
> for over the years.

But they often go to households who have NO CONTACT (or LIMITED CONTACT) with radio stations. A 'non-listener' can just check a column or two...and mail the diary back! How scientific is that???
> >
> > What would happen if the diaries were taken away and 300
> > people were paid to spend an afternoon listening to a
> > variety of radio stations or formats. These people (who
> are
> > being paid to listen) could then offer their opinions and
> > views about the radio stations they heard?
>
> Would not work, as you have taken the listening out of the
> normal environment and declared each particpant to be an
> expert. All kinds of biases enter in.

There's no NORMAL ENVIRONMENT for any age group...simply because the "attention span" is so limited! Hence my suggestion for "gathering a bunch of listeners and paying them for their time!" The responses would be more honest...and indicative of what people are willing to tolerate.
>
> >
> > I'll bet the responses would be VERY SURPRISING?
>
> And useless when measured in a controlled environment.

ONCE AGAIN...THERE IS NO "CONTROLLED ENVIRONMENT!" You're assuming that listeners are "devoted to filling out the diaries they receive?" AND THIS JUST ISN'T TRUE!
>
> > Particularly to a lot of the consultants who are convinced
>
> > that every single radio market can and should be
> programmed
> > COMPLETELY THE SAME?
>
> I do not know of a single consultant who believes that.

Then maybe YOU should start doing a little more research? Every consultant I've spoken with has a tendency to compare similar size markets. Based on the age, the number of people and income, a consultant will automatically assume a a certain format will work? So what every audience in similar markets gets is "the same old promises!" This includes the high-energy sweepers and jingles, and the usual nonsense that consultants ASSUME a listener will enjoy a format. I guess this is easier than ASKING US WHAT WE REALLY WANT TO COME OUT OF OUR RADIO SPEAKERS? I'm not implying that it's possible for a radio station to 'program their music or format' to each and every person who listens to it. But there's a lot less 'person-to-person' present on the air right now.

WHICH IS THE REASON RADIO HAS ALWAYS STOOD APART FROM TELEVISION, NEWSPAPERS & OTHER SOURCES OF COMMUNICATION.

Arbitron used to be a better (and more accurate) system when there were FEWER "alternative listening choices!" But with the number of frequencies and fewer people listening, the ARB findings are based on the way BUSINESS (on-air product) is currently being presented. In many markets, that procedure is sloppy. Simply because there are listeners who (in search of a better product) have more options in which to find it!

So the feedback Arbitron is receiving DOESN'T NECESSARILY REPRESENT what most of us are listening for!

argytunes

 

[davideduardo]

Re: Maybe we're arguing apples & oranges here...

>
> > Radio is a commercial medium for 11,500 or the 13,500
> > stations in the US. Every PD will have slightly different
> > theories on stops and length, and each market's rates will
>
> > determine hwo many spots have to be sold.
>
> Maybe the rules need to be changed? Instead of basing a
> spot break theory on the size of the market....focus on
> "audience attention and retention" instead! It's the
> listeners who determine the success or failure of an
> advertiser...not a program director.

Rates are a function of audience size. When we deal with agency business in rated markets, there is no other criterion applied.

Radio stations will try to balance "pushing the rates" with what advertisers will pay per listener (actually, per percent of the local universe in the buy demo) but with te number of stations in each market, generally rates only grwo in proportion to established metrics of audience size (a listener has about the same value whether in NY or in Traverse City... it is just that the NY station gets more of them by default).

Advertisers do not look to the media for the success factors unless you are talking about singel staiton buy, local, direct accounts. Since we are talking ratings, we are talking multimedia buys. The success depends here on the product, the price, the creative, the total budget, the media mix and the lenght of the campaign as well as competitive campaigns, share of voice, etc. Very little depends on the stations and other media individually, as that part is quantifiable into cost per impression.
> >
> > There is no proof of this. In fact, there is proof that
> the
> > survey is witnin a few percent of being proportional in
> > every demo that we, as subscribers, have collectively
> asked
> > for over the years.
>
> But they often go to households who have NO CONTACT (or
> LIMITED CONTACT) with radio stations.

Arbitron does not "go to" any households. In fact, Arbitron uses a random interval sort of SSI data to make placement calls, not visits. This is the key to a random probability sample... where EVERY person should have an equal opportunity to participate.

In fact, Arbitron will not place diaries with people with media affiliations. they do not want people who have contacts with radio stations, whether they be ad agency or station affiliations. They want a random sample of people, and listening or not is not part of the criteria.

> A 'non-listener' can
> just check a column or two...and mail the diary back! How
> scientific is that???

That is perfect. We WANT to know the behaviour of every segment of the population. Arbitron is not measuring "radio listeners" but "people." Since there is a percentage of people who do not use radio, we need to know that, also, or we are not measuring the 12+ universe.

> > Would not work, as you have taken the listening out of the
> > normal environment and declared each particpant to be an
> > expert. All kinds of biases enter in.
>
> There's no NORMAL ENVIRONMENT for any age group...simply
> because the "attention span" is so limited!

The normal environment is where the individual listens to the radio in thier daily life You can not move "daily life" into a laboratory, as it cesses to be normal at that point.

> Hence my
> suggestion for "gathering a bunch of listeners and paying
> them for their time!" The responses would be more
> honest...and indicative of what people are willing to
> tolerate.

That would be a perceptual project, not ratings. Ratings are used mostly to sell advertising, and determine how many listeners there are so the stations can set prices and sell ads.

Many stations do perceptual research for themselves, secretly. They talk to, sometimes, hundreds of people. Sometimes they have listeners rate segments (morning shows, music mix, etc.) using a dial like we saw in the last presidential elections on CNN. This is to determine how much or little different aspects of a station or its competitor are liked. These can be done on the phone, in person, in groups, or one by one. You will never see the results of these, but millions are spent doing them.
> >
> > And useless when measured in a controlled environment.
>
> ONCE AGAIN...THERE IS NO "CONTROLLED ENVIRONMENT!" You're
> assuming that listeners are "devoted to filling out the
> diaries they receive?" AND THIS JUST ISN'T TRUE!

They are not devoted, but one of the things the Philly PPM test proved was that the diary does a terrific job. The shares and AQH persons (the metric for pricing) were almost the same in the PPM and the diary.

Most people who have diaries do a reasonable job of filling them out, especially since Arbitron calls each household 4 times during the course of the survey week.

(The only difference in the PPM is the registering of very short, very occaional listening to additional staitons... causing the cume to increase and the TSL to go down, but AQH persons stays the same...)
> >
> > I do not know of a single consultant who believes that.
>
> Then maybe YOU should start doing a little more research?
> Every consultant I've spoken with has a tendency to compare
> similar size markets.

That is not "Putting the same programming on in every market." What is is doing is comparing similar situations so that the market can be evaluated. Nearly every city in the US is a candidate for most of the major formats, and, depending on ethnic compostion, for all the rest.

Smaller towns can not support niche formats (Smoothe Jazz gets a 3 share in Billings or in LA... but you need a 10 share to survive in Billings). So we have a limited palette of formats that people in the USA accept. You start there, looking at vable formats for your market size and then comparing to the competitive array. At that point, one researches the existing staitons to see the strengths and weaknesses and makes a decision.

And, of course we compare similar size markets. I can not put a $2 million a year LA morning talent in Austin, as the market will not support that kind of costs. So I also need to know what I can do given the market billings, and compare with sucess stories from comparable markets.

> Based on the age, the number of
> people and income, a consultant will automatically assume a
> a certain format will work?

No, they will look at the above factors and then do a format search study, to confirm a need.

> So what every audience in
> similar markets gets is "the same old promises!"

This has nothing to do with consultants and everything to do with the fact that the tastes of Americans across the country is very, very similar. And, within certain formats, nearly identical. An AC in seattle will sound not much different from one in Cleveland or Atlanta.

There are no hidden formats. If there were, our research would find them very fast and they would be gtried all over within a year or so. Look at Jammin' Oldies, 70's formats, 80's formats, Jack, etc. Some new ideas work, some die off, and some stay.

> This
> includes the high-energy sweepers and jingles, and the usual
> nonsense that consultants ASSUME a listener will enjoy a
> format. I guess this is easier than ASKING US WHAT WE
> REALLY WANT TO COME OUT OF OUR RADIO SPEAKERS?

You have obviously never participated in the planning and implementation of a perceptual. All this stuff is asked in one way or another. However, the sweepers and jingles and stuff is the "art" of radio, whether done by a PD or assisted by a consultant. Good programmers understand how their staiton should sound... we often say we can hear the station in our mind before it goes on the air... and we compare the product with our vision and tweek till they match.

Some of radio, and in many cases, the larger part, is using the skills of the programmers involved to create a great product form research format concepts and ssong lists. Putting it together, massaging each hour's music, and such are what the local PD does. The consultant does not run Selector, does not rored the liners, etc., etc. The consultant helps develop strategy, but does not program the station.

> I'm not
> implying that it's possible for a radio station to 'program
> their music or format' to each and every person who listens
> to it. But there's a lot less 'person-to-person' present on
> the air right now.

And you know this how? You definitely don't know that stations spend more time and money talking to listeners today than they ever did before. So how would you know this?
>
> WHICH IS THE REASON RADIO HAS ALWAYS STOOD APART FROM
> TELEVISION, NEWSPAPERS & OTHER SOURCES OF COMMUNICATION.

No, that is not true. Radio's glory years were based on networked, national shows and very little local programming at all... from the early 20's to the mid 50's in fact... radio's first 30 years!
>
> Arbitron used to be a better (and more accurate) system when
> there were FEWER "alternative listening choices!" But with
> the number of frequencies and fewer people listening,

There are 2% fewer people listening (as a percentage of population) today than in 1965. In other words, the erosion of radio listeners has been at the rate of 0.05% per year on the average. Hardly notable.

> the
> ARB findings are based on the way BUSINESS (on-air product)
> is currently being presented. In many markets, that
> procedure is sloppy. Simply because there are listeners who
> (in search of a better product) have more options in which
> to find it!

So? Whatever they listen to is written down. the number of stations does not define the way people are surveyed, it defines how thin the slices of the pie are going to be.
>
> So the feedback Arbitron is receiving DOESN'T NECESSARILY
> REPRESENT what most of us are listening for!

Arbitron is supposed to measure what people listen TO, not what they listen FOR. Arbitron is a radio advertising sales tool. It is not programming research.

 

[Seven_Costanza]

Re: Maybe we're arguing apples & oranges here...

> I can only imagine the cost of this. It would be more than
> radio bills a year. It would require a personality profile
> of every listener surveyed, and require tracking them all
> through the day.

You're stuck in "survey" mode. I'm looking for observational results. "How much did you raise for charity last year?" Is a simple question. Come up with 25 to 100 other simple questions with quantifiable, auditible, answers, tab the answers, produce the results. Cost: Almost nil. Value: Priceless. You get a scorecard of station's effectiveness within a given market. Convince the ad community that this is the way to do business and you have created a massive paradigm shift.


> I had a station in market 13, where 75 to 79 stations make
> the book (highest of any market) which never used outdoor,
> never used TV and never did money contests. It was #1 for 20
> years, with the core demos being 18-34 and 25-44. The
> programming creadted the recall, not promotion.
>

That's all well and good. But how effective was the station at selling itself to the listener? Could they sell out a theater in 10 minutes? That's could be a measure of effectiveness (I'm guessing they could, but just for arguments sake here). Did you ever get excited because a song you're exculsive on charted high in Sound Scan? Another possible measure of effectiveness.

> If you want to criticize the measurement, look at how
> newspapers sell: circulation. No idea who read all the
> pages, no idea if the ads were noticed, no idea if they were
> acted on. Occasionnally a study is done, but not on an
> ongoing basis. It is too costly.

At least they know EXACTLY who the majority of their readers (subscribers) are. I can't pick up my phone and ask Joe "what did you think of our morning show this morning?" I have no idea who Joe is or if he bothered to listen this morning. That's an extreme example, but it just underlines the fact that we have no idea who is listening and when and why, etc. At least newspapers have some kind of baseline to work with.


> Some of the things you ask for are possible, but at a cost
> that far exceeds the billing of the entire industry. Radio
> requires frequent measurement... the PPM will issue monthly,
> in fact. To do a census of listeners in the roughly 250
> Arbitron markets 4 times a year would cost $30 billion a
> year, give or take ten billion.

Only if you stay in "survey" mode. The answers should come observationally and andetoctally.

> No, it is not too much. But it is literally impossible.

Nothing's impossible buddy. Now go have a Happy Easter if that's the way you go.

 

[davideduardo]

Re: Maybe we're arguing apples & oranges here...

>
> You're stuck in "survey" mode. I'm looking for
> observational results. "How much did you raise for charity
> last year?" Is a simple question.

The only results a station wants are those that relate to related behaviour... not things like charitible giving which would be hard to separate as to how it might be related to a radio station choice vs. family upbringing (a custom of giving) or income (the ability to give). There are too many variables there to identify the radio component, if there even is one.

> Come up with 25 to 100
> other simple questions with quantifiable, auditible,
> answers, tab the answers, produce the results. Cost: Almost
> nil. Value: Priceless. You get a scorecard of station's
> effectiveness within a given market.

As far as lifestyle, that is why we have Tapscan from Arbitron. A separate follow up survey quantifies hundreds of behavioural items, from frequncy fo sporting events attended to car purchases and ownership, income levels, degree of education, travel and leisure activities, etc., etc.

A sample size of 25 or even 100 is not adequate to create a user profile. Let's take one of the oldies stations I work with as an example. These are the distinguishable varibles I would want data on:

M & F

25-34
35-44
45-54

Each age set by sex.

Income level distribution based on a pilot study to determine the basis for the quota.

educational level.

Family size.

National origin

Language dominance

Time in US / Born in US.

Stratification based on P1 and P2 relationships.

Anything less, and I can not understand the audience I have right now... that takes about 300 interviews, and would cost somewhere in the vicinity of $45,000 based on a 25 minute interview with limited verbatims.

> Convince the ad
> community that this is the way to do business and you have
> created a massive paradigm shift.

Radio is a secondary medium for ad agencies and they buy it to fill in reach and frequency for TV, print, Cable, etc. There is no way to "convince" the ad community to buy for different and highly subjective reasons when they don't want to be convinced.
>
>
> > I had a station in market 13, where 75 to 79 stations make
>
> > the book (highest of any market) which never used outdoor,
>
> > never used TV and never did money contests. It was #1 for
> 20
> > years, with the core demos being 18-34 and 25-44. The
> > programming creadted the recall, not promotion.
> >
>
> That's all well and good. But how effective was the station
> at selling itself to the listener?

It was #1 for 20 years. As an ad medium in a very congested market (132 staitons, 8 or 9 TVs, cable, satellite, etc) there is no way of measuring it alone. I don't know of any agency buy (95% of business was agency) that bought only one station.

> Could they sell out a
> theater in 10 minutes? That's could be a measure of
> effectiveness (I'm guessing they could, but just for
> arguments sake here). Did you ever get excited because a
> song you're exculsive on charted high in Sound Scan?
> Another possible measure of effectiveness.

We did not play exclusive songs. We waited until other stations took the risk and then played them. In fact, in the PD office there was a sign that said, "The record promoter is not your friend." It was intended to be seen by record promoters.

Radio, in making hits, is only a medium that gets exposure. The quality of the song to the audience makes it a hit. A stiff can be played over and over, and it is still a stiff. You can not polish a turd.

The station wa based on personality. Still is.
>
> > If you want to criticize the measurement, look at how
> > newspapers sell: circulation. No idea who read all the
> > pages, no idea if the ads were noticed, no idea if they
> were
> > acted on. Occasionnally a study is done, but not on an
> > ongoing basis. It is too costly.
>
> At least they know EXACTLY who the majority of their readers
> (subscribers) are.

So does radio, on the ZIP code and Census track level. We can quantify how many listeners we have within, let's say, a 10 minute drive of a certain supermarket. We do not need to know the names, just the density by geographic unit.

> I can't pick up my phone and ask Joe
> "what did you think of our morning show this morning?" I
> have no idea who Joe is or if he bothered to listen this
> morning. That's an extreme example, but it just underlines
> the fact that we have no idea who is listening and when and
> why, etc. At least newspapers have some kind of baseline to
> work with.

I can have people "pick up the phone" and with a simple screener, find at random a number of listeners and ask them how they used my staitons and others in the last day or few hours or week... and I do. Hundreds of thousands of times a year.
>
>
> > Some of the things you ask for are possible, but at a cost
>
> > that far exceeds the billing of the entire industry. Radio
>
> > requires frequent measurement... the PPM will issue
> monthly,
> > in fact. To do a census of listeners in the roughly 250
> > Arbitron markets 4 times a year would cost $30 billion a
> > year, give or take ten billion.
>
> Only if you stay in "survey" mode. The answers should come
> observationally and andetoctally.

Then they are anecdotal, not replicable and valueless. A survey is a survey because it is desigend to have a specific margin of error and to be replicable in results no matter howm many times you repeat it with new samples.
>
> > No, it is not too much. But it is literally impossible.
>
> Nothing's impossible buddy. Now go have a Happy Easter if
> that's the way you go.

Actually, certaint hings are impossible. To spend $30 billion on research in a $20 billion industry is not possible unless Uncle Midas comes and turns BS into gold...

 

[KMRichards]

Re: Maybe we're arguing apples & oranges here...

> ONCE AGAIN...THERE IS NO "CONTROLLED ENVIRONMENT!"

> I guess this is easier than ASKING US WHAT WE
> REALLY WANT TO COME OUT OF OUR RADIO SPEAKERS?

> WHICH IS THE REASON RADIO HAS ALWAYS STOOD APART FROM
> TELEVISION, NEWSPAPERS & OTHER SOURCES OF COMMUNICATION.

> So the feedback Arbitron is receiving DOESN'T NECESSARILY
> REPRESENT what most of us are listening for!

Shouting does not make you right, only annoying.

Arbitron is not in the business of telling stations what listeners want to hear. They are in the business of telling advertising agencies how many people listen to the radio stations in a market where the agency wants to buy time.

I am afraid you will continue to lose this argument, because your posts show a marked lack of knowledge about the process, as well as an emotional -- but unrealistic -- demand that the process change.<P ID="signature">______________


</P>

 

[argytunes]

Re: Maybe we're arguing apples & oranges here...

> Shouting does not make you right, only annoying.

I'M SORRY YOU FEEL THIS WAY. I use CAPS because they separate my comments from yours. "THE SHOUTING THEORY" that seems to upset a few people doesn't bother everybody? [Whoever came up with that rule probably had some "personal control issues" that were never resolved]!
>
> Arbitron is not in the business of telling stations what
> listeners want to hear. They are in the business of telling
> advertising agencies how many people listen to the radio
> stations in a market where the agency wants to buy time.

I guess that's my point? Are the advertising agencies you've referenced actually getting an accurate picture of the age group who are supposedly listening? [If I use your "screaming theory" when it comes to posting on this board, perhaps this is the reason why there are so many bad screaming announcers??? They feel the only way to get our attention is to yell at us???].
>
> I am afraid you will continue to lose this argument, because
> your posts show a marked lack of knowledge about the
> process, as well as an emotional -- but unrealistic --
> demand that the process change.

I always thought the boards were supposed to be intended for "a variety of opinions?" If you want everybody to think the same way you do...why bother having them? I ask nobody to support or agree with anything I write.

However...unless the "way business is currently done" goes through a transformation or two, many radio stations are already writing their own epitaph!

I apologize for disagreeing with you, KM...but "the process" (which is your term) STINKS!

argytunes

 

[davideduardo]

Re: If radio has a problem... it is not Arbitron.

>
> I'M SORRY YOU FEEL THIS WAY. I use CAPS because they
> separate my comments from yours. "THE SHOUTING THEORY" that
> seems to upset a few people doesn't bother everybody?
> [Whoever came up with that rule probably had some "personal
> control issues" that were never resolved]!

Actually, the reason books and newspapers are printed in mixed case for everything except hedlines and titles is that mixed case is vastly easier to read, proveen by countless studies.

Any of us who had to deal with mechanical teletypes for news remeber how much harder it was to read, because some of the visiual cues, like capitalization of proper names, were lost. It was a developed skill to read all caps, and even though I learned, all caps sure annoys me.

Netiquette, going back to dial up bbs systems (like The Well) and text based systems like CIS, says that all caps is SHOUTING and rude.
> >
> > Arbitron is not in the business of telling stations what
> > listeners want to hear. They are in the business of
> telling
> > advertising agencies how many people listen to the radio
> > stations in a market where the agency wants to buy time.
>
> I guess that's my point? Are the advertising agencies you've
> referenced actually getting an accurate picture of the age
> group who are supposedly listening?

Yes, they are. As multiple methodologies have revealed about the same results over the years, it is reasonable to assume that today's numbers are getting an accurate count subject to the variable of any poll (in other words, there is no Arbitron issue, just the variances of doilg a poll and not a census).

Arbitron does a good job of getting very clolse proportionality in all cells it breaks out (stratification variables) and uses minimal weighting to give proporitonality in the reposts. The PPM has near-perfect proportionality since it is a permanent panel, recruited for proportionality.

> [If I use your
> "screaming theory" when it comes to posting on this board,
> perhaps this is the reason why there are so many bad
> screaming announcers??? They feel the only way to get our
> attention is to yell at us???].

No, skilled PD have determined that stations are more successful if they have a mood created by the imaging and jocks. Each PD will have a different approach, but in many formats, high energy is the difference between winning and losing, while in others being minimally intrusive is the key.

I can go to an art supply place and by oils and canbvasses and an easel... and read up on art. But no amount of research and preparation will make me into a van Gogh or a Picasso unless I have the basic talent to combine paint in an artistic way. Same with radio. Research is a tool. Great equipment is agreat tool. But only a good PD can win.
> >
> > I am afraid you will continue to lose this argument,
>
> I always thought the boards were supposed to be intended for
> "a variety of opinions?" If you want everybody to think the
> same way you do...why bother having them? I ask nobody to
> support or agree with anything I write.

I believe the point being made is that you are like donQuijote, charging windmills. You are questioning things that have simple answers.

To wit: the Arbitron argument has played out over the last 40 years, and agencies accept it as it is accredited by the MRC, and stations buy it as the price for the current sample size is affordable. The main impediment to PPM, in fact, is cost, not methodology. Radio has limited resources and will pay for the sample that gets the job done... and not one diary more.
>
> However...unless the "way business is currently done" goes
> through a transformation or two, many radio stations are
> already writing their own epitaph!

I would say things like the loss of entertainment and creativity in some stations or companies is the issue, not the way we measure audiences. The measurement is just fine, and does not affect the way listeners perceive stations.
>
> I apologize for disagreeing with you, KM...but "the process"
> (which is your term) STINKS!

Actually, the ratings process is highly refined and is at a perfect cost vs. benefits point today. Ratings do not materially affect what a listener hears day to day (although if a station gets no ratings, it changes format... or gets all 55+, it changes format, etc.) so the issue is content, not metrics.

 

[KMRichards]

Re: If radio has a problem... it is not Arbitron.

> > I'M SORRY YOU FEEL THIS WAY. I use CAPS because they
> > separate my comments from yours. "THE SHOUTING THEORY"
> that
> > seems to upset a few people doesn't bother everybody?
> > [Whoever came up with that rule probably had some
> "personal
> > control issues" that were never resolved]!
>
> Actually, the reason books and newspapers are printed in
> mixed case for everything except hedlines and titles is that
> mixed case is vastly easier to read, proveen by countless
> studies.
>
> Any of us who had to deal with mechanical teletypes for news
> remeber how much harder it was to read, because some of the
> visiual cues, like capitalization of proper names, were
> lost. It was a developed skill to read all caps, and even
> though I learned, all caps sure annoys me.
>
> Netiquette, going back to dial up bbs systems (like The
> Well) and text based systems like CIS, says that all caps is
> SHOUTING and rude.

I suppose everyone who has used the Net over the past 20+ years all have "personal control issues" in the eyes of the original poster.

The reality is that the Moderator, having been on the Net since the late 1980s, learned the NO SHOUTING rule many, many years ago, and has incorporated same into "personal moderating style".

You don't like it, then don't post.<P ID="signature">______________


</P>

 

[Seven_Costanza]

Re: Maybe we're arguing apples & oranges here...

I'm refering to surveying the stations themselves, not the audience. For example, ask "How much was your station able to raise for charity last year?", and if your station ran any form of charity event the answer is as close as an adding machine. (If you can convince your listeners to give generously to a charity, you can probably convince them to buy more than they fare share of cornflakes).

The more of these types of questions that can be quantified, the clearer the picture you'll get of your effectiveness as a broadcaster.

But you're correct in assuming that selling the buyers on a new method would be the hard part. Coming up with a market by market scorecard would be a simple and inexpensive part.

Because of the rise of the Internet with its tracking cookies and other "new media", radio is at a point that it needs to come up with a new way to sell itself.

And PPM ain't it.

 

[davideduardo]

Re: Maybe we're arguing apples & oranges here...

> I'm refering to surveying the stations themselves, not the
> audience. For example, ask "How much was your station able
> to raise for charity last year?",

In most markets, multiple stations from multiple owners participate in charity functions. As such, the results of any fund raiser would be comingled with the other stations and not distinguishable.

Single station activities also have too many variables.

A campaign for a less "popular" charity would get less results than a more visible one. For example, Katrina relief right after the hurricane would pull more than a campaign for the local animal shelter.

The support from the charity itself, such as providing phone operators, training, orientation, etc., is critical. St. Judes Hospital does so well because they help the stations (In LA, we raised $2 million in a weekend, a record for them) while other groups have limited experience withthe media.

In other words, you want to quantify based on a quality that is 80% in the hands of people and organizations outside radio and where the result would be meaningless. Your idea is also dangerous, as it would make stations favor the big charities and ignore the ones that would not get them points.

 

[Seven_Costanza]

Re: Maybe we're arguing apples & oranges here...

I'm painting with a broad brush here and you're completely missing the point. Sorry.

Let's try this again...

Conventional wisdom had Howard Stern's listenership at somewhere around 5 to 10 million. Arbitron has spend years and years and millions of dollars to come to this figure. Everybody has taken it as gospel and CBS (nee Viacom) sold alot of spots to Check Cashing businesses and Strip Clubs based upon these numbers.

Sirius Radio gave him $200 million in equity based upon the possiblity of attracting these 5 to 10 million listeners. Yet, after an entire year of Howard's five hour advertisemnt for Sirius, how many listeners signed up? I'm seeing reports of 2 to 3 million.

The question is where did those other 3 to 7 million listeners go? The answer is nowhere... Maybe they were never there to begin with! Sure, people may have written his name in a diary, but was he effective? It's that simple. Howard was given a year to convince people to continue to listen to him - the only quantity he can really sell - and he sold it to 2 or 3 million people. In other words his effectiveness is 2 or 3 million people.

The only part that remains debatable is whether 2 or 3 million people is a good result and what the value is of 2 or 3 million listeners (people that are willing to make a sizable finacial commiment to hear him) versus the 5 to 10 million that were the basis of his advertising.

 

[KMRichards]

Re: Maybe we're arguing apples & oranges here...

> I'm painting with a broad brush here and you're completely
> missing the point. Sorry.

Well, whatever you are painting with, this thread is quickly devolving into a Seven vs. David debate, and I am getting weary of it.

> Let's try this again...

No, let's not. <P ID="signature">______________


</P>

 

[davideduardo]

Re: Show me the money

> Conventional wisdom had Howard Stern's listenership at
> somewhere around 5 to 10 million.

Wrong. Arbitron, which is the only ratings company covering the markets where Howard was on terrestrial radio, measured precisely* the number of listeners, right down to the hundreds, in both AQH and Cume.

* "precisely" in sttistics or a poll means "within the margin of error inherent in the sample and methodology. In this case, it is a couple of percent.

> Arbitron has spend years
> and years and millions of dollars to come to this figure.

Wrong. Wrong. Oh so wrong.

Arbitron did not spend any "extra" money or time to meansure the listenership to Howard. They measure every station in every market, and if anyone was listening to Howard, it showed in the morning drive daypart.

Arbitron measures radio listening 24/7. It does not conduct special polls for individual stations or shows.

> Everybody has taken it as gospel and CBS (nee Viacom) sold
> alot of spots to Check Cashing businesses and Strip Clubs
> based upon these numbers.

And your point is? The numbers were valid, and the advertisers found value for the investment. This is the way all advertising, for strip clubs or Coca Cola, is bought. The point of evaluation is audience delivery vs. cost.

> Sirius Radio gave him $200 million in equity based upon the
> possiblity of attracting these 5 to 10 million listeners.
> Yet, after an entire year of Howard's five hour advertisemnt
> for Sirius, how many listeners signed up? I'm seeing
> reports of 2 to 3 million.

So? It is a paid service. On terrestrial radio, it was free. A large percnetage of listeners apparently did not find the cost of installing a Sirius radio, plus themonthly fee, to be worth it.

> The question is where did those other 3 to 7 million
> listeners go? The answer is nowhere...

Your batting average today is .000 because what happened is taht those listeners found other free radio alternatives. Radio listening in the "Stern Markets" is not down since he moved to satellite, based on the first 2 months of the Winter book. So, Joe Sixpack who listened to Howard in NY now may be listening to some other station.

> Maybe they were
> never there to begin with!

Wrong again. No comment.

> Sure, people may have written
> his name in a diary, but was he effective? It's that
> simple. Howard was given a year to convince people to
> continue to listen to him - the only quantity he can really
> sell - and he sold it to 2 or 3 million people. In other
> words his effectiveness is 2 or 3 million people.

What a strange leap of faith.

I like movies. However, I do not have the paid movie channels on Direct TV as I do not think I would get any value from the cost.

In other words, I made a decision not to pay extra, not a decision not to watch movies.

Many of Stern's listeners looked at cost, obviously, and decided not to buy now. Some may later buy. SOme may find other things to listen to. But they were listening to Stern in the past and they are not in the present.
>
> The only part that remains debatable is whether 2 or 3
> million people is a good result and what the value is of 2
> or 3 million listeners (people that are willing to make a
> sizable finacial commiment to hear him) versus the 5 to 10
> million that were the basis of his advertising.

Actually, as I said, the cume figure for Howard was very precise, and not a range.

Heck, if soccer games were free, I would go to more of them. As it is, I do not go. It's about money, not radio listening, in this case.
>