Author Archive

In-Car Listening: the Black Hole of Radio Ratings

February 11, 2012 Leave a comment

With all the hand-wringing about new in-dash audio alternatives to radio, it’s amazing that we don’t really know how much time people spend listening to radio in the car.  

PPM doesn’t separate in-car listening from other out-of-home listening. And while diaries do break out in-car listening, the recorded “reality” of diary measurement neither matches listener perceptions nor what PPM seems to show, albeit from behind the curtain. 

If you ask them, most North Americans will tell you that at least 50% of the time they spend listening to radio is done in the car. But diaries typically show that just over a third of all listening takes place in the car.

Meanwhile, although PPM generally shows the same overall split as the diary for in-home and out-of-home listening, it actually shows higher out-of-home listening on evenings and weekends than the diary. It’s safe to assume that most of this evening and weekend listening is not happening at work, but rather in the car—during the trip to the grocery store or while you take the kids to soccer—those occasions that you’re not likely to recall a week later when you’re filling out a diary. This makes you wonder just how much out-of-home tuning that PPM records during the workday is actually at work listening—and how much represents in-car listening when you’re running errands or going to meetings out of the office that again wouldn’t be recalled or in turn recorded in a diary.

All of this suggests that we should probably be crediting more radio listening to the car than what our ratings systems—or, more specifically, the diary system— have traditionally led us to believe. And, as commuting times grow ever longer, this means the car becomes an even more important competitive arena for radio listening.

Should PPM-rated stations design and target their programming to the in-car listener, and promote usage of listening to their station in the car—much like Lite Rock stations were built for the workplace and promoted to reach the at work listener? And just how should radio react when it comes to sizing up and dealing with the potential impact of Pandora and other new audio alternatives in the car?

We need to add an accurate accounting of in-car listening to the ‘to do’ list of our ratings services—ideally somewhere near the very top.

So, How Many People Really Tune Out When the Spots Come On?

December 19, 2011 1 comment

The headline sure is encouraging.  According to the Coleman Insights/Arbitron/Media Monitors study, What Happens When the Spots Come On, radio stations hold 93% of their lead-in audience through the average stop set.

But, before you start loading up on long stop sets, there’s something of a devil lying in the details here. When they presented the research, Coleman and Arbitron were careful to note that the 93% doesn’t refer to how many listeners stay tuned throughout the stop set but rather to the overall audience levels prior to and immediately following the break. The actual proportion of listeners who stay tuned through the average stop set is lower than 93%.  The problem is we don’t know exactly how much lower.Image

Radio stations experience a natural audience churn that happens independent of any tune-out triggers.  Listeners get in or out of their car, enter or leave the room where the radio is on, or do any number of other activities that begin or end their listening session and have nothing to do with their reaction to what they just heard.

This natural churn is also captured in the results of the Coleman/Arbitron/Media Monitors study.  As a result, it effectively dilutes the impact of spots on listeners who tuned in at the beginning of the stop set.

Imagine that we have a radio station with ten listeners tuned in via PPM. After the first spot, one of the ten listeners switches to another radio station while another one ends their radio listening session (let’s say by leaving their car). Meanwhile, still another listener turns on their radio for the first time that day. We now have nine listeners, eight of whom were there before the first spot. (That’s 90% of the number of listeners we had when we started, while just 89% of the listeners still listening to radio are still with us.)  Once again, after the second spot, one listener starts their listening session with us, another one ends their radio listening session altogether, while a third listener switches to another station. But, this time, a listener also switches from another station to our station. (Maybe they were running spots too.).  All in all, two listeners have jumped on and two have jumped off after the second spot. We still have nine listeners (90% of the overall audience level of when we started), but our station audience now holds just six of the eight listeners—or 75% of those who are still listening to radio.

Of course, this just an example. The amount of natural audience churn and the number of listeners switching to and from one station to another during the stop set is not called out in the research. Jaye Albright, who drills into the granular detail of all things PPM for her clients in Canada using BBM’s minute-by-minute data, tells me she was recently looking at 25-54 switchers to a major market Country station and found it’s roughly a 4-8% “in” and 4-8% “out” in every single minute.

This is not to say the results of the Coleman/Arbitron/Media Monitors results aren’t meaningful. It’s good to know that, regardless of tune-out, overall audience levels stay quite high during the average stop set, even if we should also consider the net effect—how much the audience might have grown if we weren’t running any spots. It’s also helpful to note the differences in audience levels by demo, format and ethnicity—News/Talk, Urban and Hispanic stations for example should be taking these results to the bank.

Coleman, Arbitron, and Media Monitors also quite rightly point out that the study says nothing about the impact of high commercial loads on the brand.

The research partners promise further analyses in 2012. Hopefully, we’ll get more insight into how spots and other content affect not just audience levels but actual switching behavior.

The “No Research” Legacy of Steve Jobs

November 13, 2011 Leave a comment

As a researcher, it’s a little dispiriting to hear the constant refrain that Steve Jobs achieved his remarkable string of successes without any market research.   And it’s positively frightening to think that thousands of less-talented entrepreneurs will blindly follow Jobs’ example of working without a net, only to come crashing down to earth.

But it is worth setting the record straight.  Apple has used market research, though perhaps not in the traditional sense of the word (much to their credit). As recently as May of this year, they launched the Apple Customer Pulse panel among consumers who own Apple products.  And while they may not have used outside researchers to help them develop new products, they appear to have had a rigourous design process built around the needs of the consumer. In Jobs’ own words, “It’s not the consumers’ job to know what they want… we figure out what we want. And I think we’re pretty good at having the right discipline to think through whether a lot of other people are going to want it [emphasis mine].” And then there were those iPhone prototypes that Apple employees carelessly left at bars – presumably artifacts from internal product testing. All of this counts as market research, if only by another name and outside the box that market researchers typically use to confine their craft.

More accurately, Job’s legacy is that most traditional market research is useless when it comes to designing new products—after all, consumers can only comment on those things for which they have a frame of reference.  But it’s also true that research that digs beneath the surface of what consumers say, and focuses on understanding  their needs and motivations can inspire the development of innovative and successful products.  Maybe Jobs and Apple could have used more of that kind of research when they designed some of their less successful offerings… like Ping and Apple TV.

Categories: State of research Tags: ,

“Who cares if radio survives? Something else will happen.”

November 13, 2011 2 comments

This is one of my favourite new quotes. It’s from Ira Glass of public radio’s “This American Life” in a recent public forum.  Though I’m sure a few folks who own broadcast licenses would disagree, there’s an essential truth and an implicit hope here for the rest of us who are attached to the radio business one way or another.

A 1997 photo of Ira Glass: in 2011, might we suggest that radio can have pictures too?

Radio as we know it is changing in a way that we may not even recognize it a few years from now.  Radio that emanates from the broadcast tower may even fade away—though I personally suspect it will be around longer than some pundits predict.  But, even if those towers do go down, something similar will inevitably rise up from the ashes.

I’m convinced that listener research can play a big role in understanding these changes and identifying new opportunities. But it will take a vastly different approach than most of the radio research that’s done today.

The questions are no longer “what station do you think of first for [insert attribute here]” but more like “given all the other media alternatives available today, what is it about [insert type of radio here – AM/FM radio, Pandora, podcasts etc.] that gives it its value to you.”

How, and to whom, you ask the questions also has to change. You can’t expect to get the insights from the dinner-hour interruption of a random landline telephone call that you can get out of a two-way conversation with an engaged consumer.  And that conversation could take place where the consumer listens to whatever they listen to, in an online/mobile community of which they are a member, or any other mutually agreed upon setting.

The key is staying open to all the possibilities, with the resolve to be on the side of the survivors, wherever that might lead you.

Is Radio Ready to Manage the Data Deluge?

October 15, 2011 Leave a comment

IDC estimates that mankind will create 1.8 zettabytes of data in 2011. That’s 1.8 trillion gigabytes, some 9 times greater than the amount of data we generated as recently as 5 years ago.

The sheer volume of data, and the race to harness its value, has potential to be a game-changer for  all sorts of industries. And that includes radio. We now have more decision-making information at our disposal than ever—PPM, cost effective online survey options, chatter on social networks, and transactional data from radio’s digital extensions (not just the station website or Facebook page, but also apps and complementary web services to broadcast such as Radio DNS).

The whole notion of “Big Data” holds huge promise, but also raises big questions. How can we keep up—capturing and storing all that data? How do we take data from different sources and combine them into a coherent whole? How do we filter, process and communicate the data to provide timely, relevant information for decision-makers? And maybe the biggest question of all—is radio willing and able to face these challenges?  Other sectors, including the digital media, are already well down that road.

Starting the Conversation with a New Launch

August 21, 2011 Leave a comment

Lost in all the buzz surrounding Merlin Media’s new FM News stations in New York and Chicago have been the listener polls they have placed on their (still rudimentary) websites.

There’s a web page on each station site where their morning news anchor asks listeners to respond to a handful of questions to get their input on the station. Interestingly, the questions are presented in a poll format where you can “vote” and view results to date. (Sidenote: despite what appears to be a female-based format, at I write this, 79.2% of those responding to the FM News 101.9 poll were men.)

I suspect that the main purpose of this “research” is marketing. Both web pages boldly announce that “We’re building FM News 101.1/101.9 just for YOU!” And, with all the brains and resources behind these launches, you would think they could do a bit better job if research was their job one. (And maybe they are, somewhere behind the curtain.)

The idea is intriguing. New stations like these that venture into uncharted terrain could clearly benefit from listener feedback to help them refine their programming. There a number of ways they could do this: online bulletin boards conducted over a period of several days, a larger panel selected from the target to give a quantitative assessment, etc.  With all the “Format Finder” studies that have conducted over the past 30 years or so, how come you don’t hear much about “Format Fixer” studies before or immediately after a station launches?

Closing the Engagement Gap

August 7, 2011 7 comments

Fred Jacobs invoked the 80:20 rule this past week to blog about the value of conducting research on loyal, committed  listeners.  He makes some great points. It also struck me that he might be on to something that could help close the PPM engagement gap.

PPM is re-writing the rulebook for North American radio by placing more emphasis on exposure, and less on engagement. Listeners no longer have to be sufficiently engaged in a station to remember listening to it, they only need to be exposed to the signal. And so it is that low engagement, mass appeal music machines tend to do better in PPM than in diary.

That’s not all bad of course. Radio and its advertisers get a better measure of how many people actually hear the station and the ads it runs.

But there’s also a problem: as PPM leads radio away from engagement and towards exposure, more and more advertisers are heading in the opposite direction. They’re upping their spend on digital and social media precisely because these media specialize in engagement vs. exposure. By putting all of its apples in the PPM basket, radio risks falling even farther behind in losing those engagement dollars. Arbitron appeared to be on the way to closing the engagement gap with their “Radio Affinity” research project, then abruptly shelved it late last year.  

That leaves it to stations to reach out to their loyal audience base, and not only find out what would get them to listen longer but also what would encourage them to engage with advertisers.  Doable? I think so.

Researching Why People Tune In vs. Managing Tune-Out

August 1, 2011 Leave a comment

I was intrigued by one of the findings released last week from the Alan Burns survey on women radio listeners in the U.S. Fewer than 4 in 10 women (39%) who listen to morning radio have a morning radio feature or game they look forward to.  I couldn’t help but wonder what that figure might have looked liked before PPM, and the subsequent gutting of morning show talk and features.

PPM does a fine job of helping stations manage tune-out.  But is anyone tracking those things that get people to tune in to radio in the first place?

Pandora’s Black Box?

July 30, 2011 2 comments

In case you skipped the trades for some July sun this week, Edison Research issued a release Thursday (07/28) that generated some heat. An audience analysis by Edison Research indicates that 18-34 listening to Pandora has reached between .7 and .9 ratings points in the top US radio markets. This would effectively rank Pandora among each market’s leading 18-34 FM stations (see RAIN’s analysis).

This triggered considerable hand-wringing and denial from radio, but also some questions about the data which, at least so far, are unanswered. Mary Beth Garber of Katz Radio (who admittedly is backing its own horse in this race) questioned  the market geography of the study, and whether it provides an “apples-to-apples comparison.”  In the release, Edison says they converted data provided by Pandora into the kind of AQH estimates used by Arbitron.  Beyond that, few details on the methodology or data source are provided, with nothing on either the Edison or the Pandora website.

It’s easy to see why Pandora would commission this analysis. Placing their numbers on traditional radio yardsticks reinforces the legitimacy of the service to media buyers and shareholders—even if it’s open to debate whether Pandora constitutes “radio” or just another way of listening to music. And there’s no reason to suspect that the numbers have been goosed; Pandora’s dominance in the space is well documented.

But a little more transparency in the methodology would be that much more convincing.

Why CPO Households Matter

July 23, 2011 2 comments

What you don’t know about cellphone-only households in your market can hurt you. Especially if you don’t know how many of these households your ratings company is sampling.

We just completed a strategic study in a major Canadian market where 31% of 18-49 year-old radio listeners said they lived in a cellphone-only household (sample size: 1,000, using a market representative online sample).  And, despite BBM claims to the contrary, CPO households varied sharply by format. From 43% among the cume for an alternative-leaning station to 35% for the cume of the market-leading CHR  station and 29% for that of a local Hot AC.  Yes, age plays a role, but notably, only 3 years separated the average age of the cume listeners to the three stations. (Note: the proportion of cellphone-only households in Canada is generally considered lower than it is in most Western countries.)

%d bloggers like this: