’It’s All I Ever Read!’ (Or ‘Making Better Listener Promos’)

It always makes me happy to hear audience members on the air — when done right, it legitimizes the radio station and reminds other listeners that they’re still part of a community. But when I listen to a station critically, I almost always end up making a note about the sweepers and promos that feature “listener” audio.

Often, the listener sweepers sound phony. Sometimes they’re too busy or overstuffed. Or they’re assembled in a way that undercuts their effectiveness. In recent years, it’s been common to come across a small-or-medium-market station, often unhosted, where sweeper after sweeper is a barrage of not-so-credible listener claims taken from an imaging package, including my oft-derided “It’s all I ever listen to.” That one just seems like cruel irony on a 2-share CHR station.

There was a long-ago time when listener drops were a big deal — a mark of imaging sophistication back when stations used less imaging outside of the :00 ID and contest promos. The first ones were used over intros, there was generally only one per station, and they were short statements: “13Q plays less commercials”; “CKLW plays the most music,” or on its Detroit rival, “WDRQ is a ball!” Having a listener declare that “fewer commercials means more music on WLS [Chicago]” doesn’t sound so profound now, but radio people used to talk about that one.

Later, there was a trend toward listener “comment line” promos on stations such as KPWR (Power 106) Los Angeles or KMEL San Francisco during their most phenomenal moments in the late ’80s. Often, programmers would throw in negative audience comments as well, either to add authenticity, or in hopes of prompting another listener to defend the station.

Eventually, the “comment line” feature faded, but similarly structured listener promos remained. They became a cliché of radio-station launches. By then, there was little attempt made to sound authentic. You would sometimes hear “you’re my favorite station” a half-hour after a station launched. Just as often, though, I heard stations who didn’t get the listeners on the air soon enough, creating the impression that the new station was merely sending its 10,000 songs in a row into the ether with nobody noticing.

KRTH (K-Earth 101) Los Angeles used short listener drops in an exciting, kinetic way that many stations copied, usually to less effect. As stations became more overproduced in the late ’10s, the listener drops became a word salad, random punctuation, not an actual statement about what was good about the radio station in a way that might help listeners articulate their fandom to others, or create fear of tuning out. I’ve always agreed with Alpha Libraries VP Steve Wall that “listener drops are just as important as the voiceover person.,” and it irked me to hear them become throwaways.

Listener sweepers bother me a little less than they did a few years ago, in part, oddly enough, because of AI. I haven’t knowingly heard great AI-driven jocks or newscasts, but I have heard surprisingly authentic AI-generated listener drops that sound less canned than the alternative. One PD notes that the services have become more sophisticated over the last year, with a better ability to mimic the creator, and have better inflection. (“Often, it’s about knowing what to capitalize or put in quotes,” he says.)

I’m not opposed to some use of AI listener drops. As a PD, it’s tempting to put your words in a listener’s mouth. That said, part of the goal in capturing listener audio is to figure out what they think your station’s value is. It is unlikely a listener would tell me “K-Sean plays just enough variety, but they always surround the ‘oh-wow’ songs with two smashes,” or “you play my favorite current songs 92 times a week, which is just about right, because at 109 times, I really do get sick of them.”

An AI-generated drop is easier to make sound local. Great Lakes Media CEO Tom Langmyer notes, “It’s hard to sound authentic when you are in Texas, and the person who says they ‘listen to your station all day’ has a New York accent or vice versa.” But AI listeners can’t hear themselves on the air, recommend you to a friend, or carry a PPM meter.

“AI is getting better,” says Wall. “The speech-to-speech process sounds best and most natural, but real people with personal stories and enthusiasm can’t be mimicked by AI.”

When I polled Facebook friends about listener drops, the majority were still in favor of trying to get them either at station events or from callers (or now, listener talkback features in radio-station streaming apps). “I found running tape on a listener who just won a bunch of money always provided good reaction bits,” says retiring KONO San Antonio APD Steve Casanova.

“We are heavy with callers at [CHBM] Boom [97.3 Toronto],” says PD Troy McCallum. “Over 10 years of harvesting great calls has provided us with an arsenal of authentic audio.”

“Listeners waiting in line at a concert or station event in your market are always willing to talk,” says WEZN (Star 99.9) Bridgeport, Conn., PD Kevin Begley. “Send the street team or somebody with a recorder.”

Depending on callers in the text-to-win era or members of a skeletal station staff is harder now. Consultant Dave Lange replied to Begley’s comments by asking, “Does anyone have a street team anymore?” That led WKCI (KC101) New Haven PD Adam Rivers to reply, “I can confirm, as the direct competitor, Kevin Begley’s stations have great street teams.”

“I’d strike gold [at listener events] when someone said something clever off the top of their head, sounding real and natural. I don’t think AI can replicate that,” says imaging voice/production veteran Jeff Berlin. Often “the ideal window was between the first and third alcoholic beverage.”

“Less is more. Don’t have them saying ‘radio speak,’” says Tom Lawler. “’That’s my station’ is the worst line anybody could ever use in an imaging piece,” adds Matt Mony. “I’ve never liked imaging where a listener repeated an awkward line like ‘they play all the hits’ or ‘they help get me going in the morning,’” says WILK Wilkes-Barre, Pa.’s Jason Barsky.

Barsky recommends “weaving in a couple of quick clips that share … a relatable message” and letting the imaging voice deliver the actual sell lines:

LISTENER: “My co-workers got mad cause I’d sing … ”

VO: “All the hits … ”

LISTENER: “ … telling me, ‘Use ear buds … ’”

VO: “ … with fewer commercials … ”

LISTENER: “ … but now they love it. I’ll bet we sing more than we work.”

VO: (Ends with the station name.)

There’s also one simple assembly issue that would make sweepers better. Listener drops frequently suffer from the same issue as those built around artist IDs from an imaging package. A major-market station can end with at least one celebrity actually saying the call letters. A smaller one can’t pretend that Dua Lipa ever visited the 1-800-LAWYER studio, when the station imaging voice has to be the one to actually end with the station name.

More than ever, the same dynamic is often at work in listener drops. K-Earth’s sweepers had the listeners deliver the call letters. I hear a lot of splitters where the imaging voice tags the audience comments in a way that sounds tacked on, or with a read that sounds desultory and kills the sweeper’s energy. Just making sure that the inflection and EQ match what came before would make a big difference.

Better listener drops are part of the larger “listeners on the air” issue. Often, it comes down to whether we think listeners can still be prompted to call a radio station. “Any great air talent will tell you, put a listener on the air and the phones light up with listeners wanting to be on the air,” says Wall. I hope that’s still true, if stations are willing to provide prompts.