Where is the Outrage Over PPM?

This week I went down the PPM rabbit hole and it is worse than I ever imagined.

After talking at length with researcher Richard Harker (hear the interview here), watching this 25 minute video on the science of watermarking audio, reading blogs and articles and then comparing it all to my personal experiences with PPM data, I believe the issues with PPM are nearing DEFCON1 for our industry.

Some things all radio broadcasters should know about PPM

The PPM tones are encoded and masked by other audio. If there is no audio on your radio station, there is no PPM encoding. If you are a spoken word radio station every time the host stops talking, takes a breath or a dramatic pause – the PPM tone stops encoding.

The PPM tones encode at certain frequencies (1 to 3 kilohertz), much higher frequencies than a typical male radio announcer, meaning higher pitched voices and music actually be decoded more consistently.

There has been no test results, at least released to the radio industry, how loud the radio station must be playing or how close to the radio the PPM device needs to be in order for the masked tone to be recognized and decoded. Though it is noteworthy that background radio station formats, like smooth jazz, have suffered greatly in the PPM era.

Audio watermarking technology can be wobbly leaving gaps (some small, some giant) in decoding and unknown amounts of unreported listening.

PPM encoding on internet streams is even less reliable. Just like a .jpg or .mp3 is compressed to make smaller files, your internet stream is compressed too, which means there is even less audio to mask the tone behind.

Because of these factors, some radio stations may only be encoding 50% of the time or sometimes even less and receiving greatly reduced credit in listening compared to what is actually happening.

My Conclusions

Radio should be mad as hell. This is costing people jobs, livelihoods, and impacting radio families across the country. Programmers, myself included, have made “strategic” adjustments to shows, personalities, and formatics based on inaccurate PPM data.

If I’m Premiere Radio or really any big radio company I’m lawyering up. With the hit talk radio has taken in recent years (see: Rush) could it be that the audience likes it fine, but PPM doesn’t?

Fight back. The Voltair seems to be a worthy investment for some stations. It essentially makes your watermarked audio easier for the PPM to recognize and decode.

Also, and this goes against my better judgement, if you’re News/Talk or Sports I would seriously consider adding a music bed or crowd noise at all times so the encoding never stops.

Arbitron Panelist,”F— this!”

“The first couple days I’m wearing this thing and I’m turning into radio just trying to get these points and then finally I said, ‘Well, f— this.’” – Former PPM panelist

How many times have you wanted to say ‘f— it.’ to Arbitron? Those were the words of a caller I will call “Joe” to the Tom Leykis Show on New Normal Network.

Listen to the comments in context here.

This is a taboo topic of conversation for terrestrial radio and it would disqualify a radio station’s ratings faster than you can push “scan” when the Kars-for-Kids commercial starts up. However, it was the Top Story on Radio Stuff’s podcast. (editor’s note: Arbitron emailed their responses to our questions, so we had a British bloke voice the answers on the podcast. It’s worth a listen for that alone.)

DISCLAIMER: As a programmer, I have praised Arbitron for measuring my genius programming with precision when ratings are high and cursed them when they dropped. (Certainly there must be a sampling error, no?)

Joe v. Arbitronado

So, I sent the audio from “Joe” to Arbitron’s Director of Programming Services Jon Miller and asked him what he thought. He says it sounds like things are working the way they should.

 “Arbitron has safeguards in place to help ensure the integrity of our PPM ratings. In this case, the panelist’s comments are a demonstration of some of those safeguards, such as calling households if their compliance in carrying the meter falls off.”

Caller Joe complained to Leykis:

 “I started getting all these phone calls. I’m thinking you know who the hell is this calling me? We’ll it’s Arbitron. So apparently this meter has something that can tell when you’re moving and when you’re sitting still. And so when I’m not moving, they’re calling me – pretty much harassing me about why am I not wearing the meter and they can’t get accurate ratings and I need to be wearing my meter. Well, after about two weeks of this, they called me and said, “You’re not wearing the meter!” and I finally said, ‘alright, fuck this. Send me a box and I’m sending all this shit back.’”

Miller wouldn’t go so far as to characterize the company’s behavior as harassment, but the calls are part of the quality control.

“Arbitron monitors compliance with its instructions, contacts households who aren’t complying and works with them to improve their carry habits whether through coaching or other incentives.”

The 2010 Broadcast Architecture study on PPM panelists talked to one woman who clipped her meter to a ceiling fan, Joe plopped it down in front of a radio, and I’ve heard that others have attached them to pets. So, Jon Miller, how do you know when a panelist is cheating.

“The PPM has a motion detector built into it allowing us to capture both the motion records and media exposure from that day. There are thresholds for how many hours of motion a day we require for our panelists to be counted in the days ratings, and the more they wear the meter the more incentives they receive.”

And then there’s sample size. It wasn’t brought up by the caller, but it is commonly heard uttered in anger and echoing through the hallways outside PDs offices.

“PPM’s sample sizes are designed to deliver the same level of statistical reliability as the Diary survey, but with less total sample. We accomplish that by surveying listeners for a much longer period of time (28 days in just one PPM survey month) compared to the one-week diary timeline. This level of detail, thousands and thousands of days of measurement across a single month, allows us to see so many granular things with PPM data that we just can’t with the Diary.”

Parting Shots

“If Arbitron is the standard in terms of traditional AM and FM radio then it’s a flawed standard.” -Joe the caller

Joe’s point would be more valid if he wasn’t just complaining that he wanted to earn the money without doing the work.

Jon offers some real valuable insights here for everyone in radio.

“One of the most fascinating things I’ve seen over the 5+ years that PPM has been in use is how much and how fast listening habits are evolving and changing. Nothing stays static in PPM, and we’re finding that radio listening is dynamic. This continually motivates broadcasters to continue offering compelling content on stations with clear and strong brand images, so that they can cut through with listeners in an ever more crowded media world.”       – Jon Miller, Arbitron VP of Programming Services

Clear. Strong. Brands. Cut Through.

And I would add this:

“PPM isn’t perfect, but it’s all we’ve got.” – Larry Gifford 

Upon Further Review

We can’t force our ideal listeners to participate – it’s a roll-of-the-dice and sometimes you roll snake eyes.

Arbitron is weeding out at least some of the cheaters. Good.

We can’t know “true” listening behavior without NSA quality spy equipment and the violation of our listener’s constitutional rights.

Stations and panelists are both trying to game the system. Makes me wonder who Arbitron is gaming.

The sample size is what it is, unless stations want to spend even more ridiculous amounts of money to be told your station is still – awesome, sucky, irrelevant, vital – depending on the time of the month.

The success or failure of your station is in the hands of Caller Joe. Good luck.

Secrets of PPM

PPM is about four years old and we continue to learn more about how to use the insane amount of information it provides and what strategies to use to maximize ratings. Inside Radio and Research Director Inc. just released a new study on PPM based on the top 30 markets. Research Director Partner Charlie Sislen talked to LarryGifford.com about the study (click here for the full study) and he provides strategic tips for programmers who are struggling in this new world.

 One of the findings in the study which may be disturbing to programmers of spoken word formats is that CUME is more important than was first thought. “Everybody knows in the PPM world CUME is important, but when you look at top skewing radio stations they are all really CUME and not TSL driven. We really believe it’s the P1 CUME that’s doing it,” says Sislen.

Sislen notes that play-by-play can help the spoken word format draw in CUME, but it is not a magic bullet; the success of play-by-play depends on the team and the market, and in many cases it can be a detriment. “What a programmer has to recognize is that this event is an important launching point for your radio station to recycle this audience back in. Take them from being P4s and get them into regular listeners. Give them a reason to come in to the other day parts and give them a reason to come back the next day, the next week, some time outside of that play-by-play.”

So how can you win with PPM? Sislen offers the three “C’s” for talk hosts;

• Crisp
• Concise
• Compelling

“If you’re not crisp, concise and compelling and you’re spoken word, the listener is going to go away, and the moment the listener goes away the PPM knows it.”

For programmers, Sislen stresses the importance of building occasions. “For sports, to put it in perspective, the typical P1 spends 6:44 with the radio station (per week). You get four more occasions (of at least five minutes within a quarter-hour) from those people that 6:44 has gone to 7:44. That’s massive. We’re just talking about your P1s and we’re getting them to come in just four more times in an entire week.”

The other hot issue is clocks. How many spots are you running and where are you running them? Some suggest two breaks per hour straddling the top and bottom quarter-hours and Sislen doesn’t disagree. “In a vacuum, absolutely that’s true. However, we don’t live in a vacuum and the spoken word is different than a music format and you’ve got to know what your competitions clocks are. Study your clocks and make sure you understand the rules that you need to garnish to get credit for a quarter-hour.” Once you figure out your competitors clocks, Sislen says you should be going into content when they go into commercial break. It’s that simple.

Listen to our entire conversation, including the importance of marketing your show and station, how buyers and sellers still need more education and what’s next for PPM.

Charlie Sislen Interview Podcast

PPM Strategy Tip

You can have fewer ratings weeks negatively impacted by vacation if you encourage talent to make vacation weeks Thursday through Wednesday. That way you interrupt only one week of ratings, instead of two.