KILLING IS MY BUSINESS: Chattin’ Charts with Billboard’s Keith Caulfield

Music sales and airplay charts document the most heard and purchased artists, albums and songs in music. Death metal and grindcore never get airplay and barely sell anything compared to less-niche genres. So the logic should be simple: charts are irrelevant to extreme metal bands and fans, right?

Not entirely. The fact that almost nothing is selling these days has actually helped bands chart that would normally have zero chance of making it into Billboard Magazine, the music industry’s favored source of info on what the nation is buying and listening to.

For the final print edition of Killing Is My Business [in Decibel issue #123, Mastodon cover], I spoke with Keith Caulfield, Associate Director of Charts / Sales at Billboard. According to Caulfield, when you drill down to Billboard’s genre charts (e.g. Hard Rock Albums or Hard Rock Digital Songs), there’s actually a decent opportunity for a metal band with a following to show up. And there’s more opportunity than ever for bands at any level to use the kind of data that Billboard uses to help make smart business decisions.


Can you explain exactly what your position entails at Billboard?

Primarily my job is to manage the Billboard 200 chart, which is our main top albums chart, do analysis on that chart, and trends in album sales. Talk about who’s #1, who’s moving up, who’s coming down, who’s selling, who’s not. My main gig is doing a lot of administrative stuff, ensuring the accuracy of our charts, working with Nielsen SoundScan to make sure the chart’s correct, making sure that albums are categorized correctly on different charts. I also manage some other charts, like the Catalog Albums chart, Heatseekers, Internet Albums, Digital Albums, Broadway Albums, Soundtracks, Compilations…and in addition to that, I do write. I have a column that runs in the magazine each week.

How exactly does Billboard determine what is hard rock vs. active rock vs. alternative rock?

We have a lot of different rock genres, and it depends on whether you’re looking at our airplay charts, or our sales charts. Specifically for our Hard Rock Albums chart, it’s basically a broad term that’s any sort of loud, guitar-based rock music. Which includes obvious metal acts, obvious punk rock acts.

It’s one of those things where you know hard rock when you hear it. You know Jack Johnson and Colbie Caillat are not hard rock, whereas Metallica and Pantera are very clearly hard rock. Sometimes an act can be multiple genres. Nirvana is both hard rock and alternative, and they’re also in the catchall “rock” category. Then someone like Sheryl Crow might just be rock, but not hard rock or alternative.

There are some other factors: how an album is being marketed and positioned in the marketplace. What radio stations are playing it, if any. If something’s not very clear cut, we have to go to other sources to see how it’s being treated, how the band identifies themselves, how are they being promoted. But most of the time you can quickly tell sonically – “Is this hard rock?” “Yes it is.” And then you move on.

Keith Caulfield

For a chart that’s sales-based, like Hard Rock Albums, there can be a lot of research that goes into it. Would I imagine there are some other charts that map more clearly onto radio formats?

We have a number of those. We have a Mainstream Rock chart, an Alternative Rock chart, there’s a Triple AAA radio chart, which is like adult alternative radio…it’s kind of like KCRW-ish, if you’re in Los Angeles. There’s also heritage rock stations, which are more old-school, old-fashioned, ‘70s type rock music, with a little bit of contemporary music. And those charts are basically directed by the stations that report to that chart.

There are a certain number of stations that are on a “panel” – the bucket of radio stations that report to a particular format. So there are X number of stations we consider “alternative” stations, and those stations each week are monitored by Nielsen BDS, part of the Nielsen company. Each week we tally up how many times all these songs are played on those stations, that are on the alternative panel. And that, in turn, produces the Alternative chart.

So sometimes, if KROQ in Los Angeles decides to go out on a limb and play a song that’s not traditionally rock, but they want to play them on KROQ, then you’ll see some weird anomalies pop onto the Alternative Airplay chart. Eminem charted Alternative back when he first came out, with “My Name Is.” I always thought that was so weird – like “Woah, what’s this white rapper dude doing on the Alternative chart? How does Billboard think he’s alternative?” Well, we don’t think he’s alternative, but the radio stations that we consider alternative decided to play him.

So even though you’re just reporting what airplay is happening, there are ways in which that act of reporting ends up defining how people think of a band or an artist.

Yeah. Certainly this past year, a lot of people blinked at the categorization of Lorde as a rock artist. She actually won a rock category at the Billboard Music Awards. It’s funny, because when a bunch of alternative stations play you, and they make “Royals” one of the biggest hits in alternative radio, well, by definition, you’re technically an alternative act, according to those stations that played you. And that’s where the categorization happens – those stations played you, and in turn that funnels into all the other things that happen afterwards. Award shows, where you get placed in magazines, where people categorize you on the web – “Oh, they’re alternative!”  It’s all because of what happened to this introductory single, before your public identity was ever established.

With metal, it’s very clear. You’re a hard rock band; you’re a metal band, there’s no denying that. Cannibal Corpse is not going to be turning up on Top 40 radio anytime soon.

What sort of data does Billboard use to determine positions on your various hard rock charts?

The two sales charts, Hard Rock Albums and Hard Rock Digital Songs, it’s sales from brick and mortar retailers, from chains like Target. It’s also internet sales from, digital sales from iTunes, Amazon MP3, Google Play… It’s just what Nielsen SoundScan collects in terms of sales each week. I think SoundScan captures like 90% of music sales in America. So the chart’s pretty close to what’s being tangibly sold in the US each week.

Same thing for Hard Rock Digital Songs, except it’s just songs that are paid downloads from digital services. Pretty much any significant seller of digital music in the United States is reporting to SoundScan each week. There are dozens of them.

I know you have separate streaming charts as well. Does Billboard have ways of looking at the overall picture of an album or song, regardless of the way it’s heard?

We have rock on-demand charts. On-demand is where you’re making a specific decision to listen to something – like I’m clicking “yes” on Spotify, whereas a passive listener is just having something streamed to them, without making a decision. We’ve got a chart for everything. But these are all separate entities. We don’t have one blended chart that blends all these things together, if that’s what you’re getting at.

[UPDATE: the month after this interview was conducted, Billboard announced that it would start integrating streaming statistics into its main Billboard 200 album charts beginning December 2014.]

How long have the Billboard Hard Rock Albums and Hard Rock Digital Songs charts been around?

The Hard Rock Albums chart actually started in July of 2007. Hard Rock Digital Songs we’ve had since January of 2011. SoundScan’s been tracking digital song sales since 2003 or 2004, but it took a very long time to get the process in place to divvy up all the different genres for digital songs. The album and genre categorization was in place before then – we’d been doing that for decades. But in terms of digital songs, it’s a lot harder.

What was the first #1 album or track on those two charts?

The first #1 hard rock album, in July of 2007, was Smashing Pumpkins’ Zeitgeist. The very next #1 album after Smashing Pumpkins was Linkin Park, Minutes to Midnight, if that makes you feel any better. Then Korn, untitled.

The first hard rock digital song #1, in January of 2011, was My Darkest Days featuring Zakk Wylde’s “Porn Star Dancing.” Sick Puppies was next, Aerosmith, Guns n’ Roses.

Which album has held its #1 position on the Hard Rock Album chart the longest?

You may laugh at this: Nickelback’s Dark Horse. It was #1 for 37 weeks on Hard Rock Albums.

I’m laughing, but I’m crying on the inside. I’d be curious to see the one that I would qualify as hard rock as metal – which of those actually lasted longest?

Certain really aggressive, sonically pummeling rock music doesn’t really sell incredibly well in general, period. So for any of those albums to even chart is kind of a big deal, and for them to hit #1 is an even more incredible deal.

The few charts where it might happen on for an extreme act are things like Heatseekers or Tastemakers. Might you have examples of extreme acts that have done well on those charts? I know Motionless in White was at the top of some chart just this past week.

Reincarnate was #1 on our Rock Albums chart, and the Hard Rock Albums chart, as well as #2 on the independent albums chart, and #8 on Internet Albums. Also #9 on the overall Billboard 200 chart, which is the overall all-the-genres-competing-against-one-another chart. [Motionless in White’s label] Fearless [Records] has actually generated a lot of high-charting records on the pop charts over the past year or two.

[Extreme] albums don’t generally stick around for very long. They’ll chart for a week or two on the Billboard 200 maybe, a little bit longer on the rock charts. But because so much emphasis is put on the first week of an album being on sale, and also because there’s a fairly limited fan base that’s going to buy this album, it adds up to a robust first week but then not a lot of sales afterwards. That’s why you’ll see someone have a big debut, and then fall off the chart.

That happens with a lot of niche acts, be it a hip-hop act or a classical act or a metal act. You have 300,000 fans, and you’re going to get maybe 20% of them to show up in your first week. And then the rest will eventually get the album, maybe they’ll stream it later, but the core fans that turn out in your first week, that’s going to be your biggest week. And then after that, it’s dwindling sales, and the album falls of the charts. So it’s not unique to metal. That’s just the nature of the beast.

When the self-titled Metallica album came out in 1991, it was a different era for metal. But that was extremely successful for a long, long time. What accounts for that long-tail for certain albums?

It could be said for any kind of album. Metallica at the time had built up such a large fan base from their work in the ‘80s. Then MTV started to embrace them with “One,” and they really embraced MTV with the self-titled album. It was the perfect blend of bringing your core fan base, and merging that with a mainstream sensibility to a degree. Sometimes artists and consumers meet at the right time and the right place. Your music resonates with people. That’s why they were able to debut at #1 with the black album, and they’ve been incredibly successful ever since then. Almost every album they put out debuts at #1 with a huge first week.

In terms of the long-tail, how it continued to sell so well, the thing had like six, seven different hits on it. MTV consistently played videos on it for, I swear, two years. You heard them every time you turned on rock radio. I’m sure the entire album was probably in rotation at many radio stations. They were on the road constantly.

I think literally for three years after it was released, they were supporting it on tour.

It was just constant hard work, constant promotion and constant great music that clearly resonated with people. That’s why that album continued to sell so well, and continues to sell so well today. It is the biggest-selling album of the SoundScan era –and when we talk about the SoundScan era, we mean when Nielsen SoundScan started tracking sales in 1991. It has sold more than 16 million copies in the US, by far the biggest-selling album in the US since 1991. So that says something!

That metric captures the success of Metallica pretty accurately, because SoundScan started tracking sales the same year the album came out.

Yeah, right around the same time. Metallica’s album debuted in August, and we started using SoundScan in May of that year. That same year, Skid Row’s album [Slave to the Grind] debuted at #1. That was a really big deal. And that was only a month after we started using SoundScan. Everyone was like “Whaaat?!”

The charts operated so differently before then. Before SoundScan, the charts were very driven by pure pop and adult contemporary music – just because that was the best kind of tracking capabilities we had, and that was what was reported to us by the retailers. Once SoundScan information came on board, we were able to see that these genre acts, like Garth Brooks, like Reba McEntire, like Metallica, like Skid Row, like Pantera, sold incredibly well. And that wasn’t truly reflected on the charts before SoundScan started. That’s why you see so many core rock acts, and core country acts, for example, show up high atop the charts after SoundScan started.

Pantera’s a unique situation. When Far Beyond Driven was released, there were so many more fans of metal at that time, and they were willing to go out and buy. Is that one of the only truly extreme albums, with screaming and loud guitars and heaviness, that has done that well on the Billboard 200?

I put together a list of the “hardest” rock #1s on the Billboard 200. I can quickly rattle them off for you, and you tell me if any of them match the sound of Pantera. I have a feeling they probably don’t: basically every Metallica album since Metallica. Skid Row’s Slave to the Grind. Van Halen, not so much…Limp Bizkit…Tool? They were #1 with Lateralus and 10,000 Days. Korn had a couple #1s. Staind was #1 a couple times. System of a Down. Disturbed…not really. I don’t think AFI counts, but they’re certainly a rock band. AC/DC, not exactly a screaming type band, but kind of screaming! Avenged Sevenfold, probably not “metal” metal. And Black Sabbath actually was #1 last year.

Wow, with 13!

Yeah, that was their first #1 album! It’s a long time coming.

None of those bands other than Sabbath approach the level of extremity or underground respect of Pantera.

Pantera continues to be this big “really?” at #1. They built up such a fervent fan base at that point, and it was the right time and the right place. Had it been any other particular week, they may have not been #1. They sold 186,000 copies in the first week with Far Beyond Driven.

Okay, so the Hard Rock Albums and Hard Rock Singles charts we’re talking about track physical and/or digital sales. Do you find that your sales charts like these two are pretty well correlated with the airplay or streaming charts?

Yeah. I think so. Sometimes you get an anomaly. Sometimes songs are huge on the radio, and no one buys them for whatever reason. Sometimes songs are big airplay hits, but that doesn’t translate into sales. It’s more difficult to convince people to buy something. As hard as it is to get on the radio, it’s harder to get anyone to buy something.

And vice-versa. Sometimes you have a great sales story, and radio just ignores you. So there generally is a lot of crossover between all the charts, and sometimes you have a curiosity that’ll do well – it’ll be a top 10 record on the airplay charts, but in terms of sales, it sells like 1000 downloads a week. And you’re just like “Really! How many millions of people heard this song, but no one wanted to buy it?”

How do companies use Billboard info to make business decisions? 

I think the general vibe is that our charts are used as a measurement tool for the industry – record labels, management, booking agents and so forth – to gauge what are the most popular acts, and what are the most popular songs and albums in the country. How can we use that to book talent on TV shows? How can we use that to book shows at concerts? How can we route our tours better?

Certainly SoundScan’s information and a lot of streaming information, you can whittle that down to zip codes and regions, so you know how well you’re doing in certain areas, so you can better gauge how to route your tour, for example. Our charts are national charts, so I don’t tell you on our chart how many copies sold in Poughkeepsie. But you can access that data through SoundScan.

Then a lot of our charts are used in a more consumery way. Consumers like to find out where their favorite acts are on the charts. It’s this huge range of “Hey I like to look at lists. Where’s my favorite artist? Oh, it’s at #6 this week. That’s fun!” to the far extreme of “Let’s look at super-granular data from Nielsen to see should we have a meet ‘n greet at this place, or should we save that opportunity for a different place, because there aren’t enough fans there?” 

I can imagine stories about a manager, or a label, that’s looking at the Heatseekers charts and they’re like “Wow, this person doesn’t have a label already! 

Yeah. Especially with our new Twitter charts, we have these new Billboard Twitter real-time charts which are pretty cool. They monitor the most-discussed and talked-about tracks and artists on Twitter. We have two different charts. One is just for the top tracks, period. So if a hugely popular artist on social networks drops a new song, it could [make it on] that Twitter chart. And it’s a real-time chart, so it may not be #1 15 minutes from now, ‘cause it’s constantly refreshing.

Then we have a chart that’s just for emerging artists on Twitter. It’s for artists that have less than a certain number of Twitter followers, so Twitter and Billboard define them as “still developing.” We hear stories from departments at different record labels and management people who look at these charts as guides to help them find new talent and to acquire for whatever the organization is. They look at these charts to help them narrow down “Alright, which of these acts are unsigned? Can we license this song that’s really popular on the charts and happens to be a European hit, but no one’s licensed it in America? Maybe we should use this chart to help find out which songs we should acquire.”

That same thing was done with all of our charts in the past, and continues to be done, but I think the Twitter real-time charts are a great example of how something that’s very now, very of the past hour, can be used.

Or you know, if you’re just a publication, if you’re a TV show and you’re looking for stories to tell, look to the charts and see “Who have we not talked about recently? Let’s scroll down and find someone. Oh look! They have a great story! They have an interesting look!“ That sounds really pedestrian, but that happens too.

It’s an interesting story in itself, those real-time Twitter charts. That’s kind of the holy grail for any data keeper – knowing exactly what’s happening NOW, and to be able to analyze it and deploy it.

Yeah. If everyone could be like “Right now, three copies have been sold in this area. One ticket has been bought in Anchorage, Alaska.” Yeah, that’s great.  But my god, we’d be just drowning in data. At least the Twitter real-time charts give you a snapshot of what’s happening right now.

And it captures a certain segment of the audience and a certain segment of the music audience that may not be captured in other ways. A lot of people on Twitter just don’t buy anything. They stream it, they play it on YouTube, they share links with people. And that’s how they interact with music.

Have you noticed any chart trends that apply to metal more than to other genres?

I don’t have a great thought on this one for you unfortunately. The general notion is that, because of the way our album sales charts work, everyone has a much better shot at charting than they ever did before. One, you don’t need to sell that much to get onto the charts, period, because fewer people purchase albums. And therefore if you can convince your fan base to buy a couple thousand albums in a week, or 10, 15, 20 thousand albums in a week – that’s a TON for some metal acts, but if you can convince them to buy that many in a week, you’ll chart. And you may not be around the week after that, but you’ll have your moment in the sun on the charts.

I mentioned Cannibal Corpse jokingly earlier, ‘cause I just love saying their name…I don’t listen to them, I just love the fact that they exist. They’re so extreme! But they just charted their highest-charting album ever, with A Skeletal Domain. It debuted at #32 on the Billboard 200 chart. Which is huge. Are they as big as they were in the ‘90s, early 2000s? I don’t know. But they’re able to have their moment. They can now say they’ve had two Top 40 albums in the United States on the Billboard 200 chart. Which is pretty cool for a very extreme act. And there’s lots of examples like that. They just happen to stick out because every time they pop on the charts, I go “Ooh boy! I get to write about Cannibal Corpse.”

There are so many sales in the metal world that happen either at shows, or some kind of distro that exists only online. It’s a single person, sending you a copy from his or her home, where it might not be reported to SoundScan.

[SoundScan] knows that they can’t capture every single transaction. So they try to extrapolate for what they’re missing. Sometimes sales are weighted. The number that we announce – you know, “Lady Gaga and Tony Bennett sold 131,000 copies, and they’re #1” – well it’s not really a pure number. That’s a weighted number. It’s a number that is based on actual sales, but it’s also weighted a little bit, to account for what’s missing in their capturing of sales. So when people say “I sold exactly 10,000 copies,” actually you probably sold more like 10,500. Or maybe 9,800-something. But SoundScan did some shake ‘n bake to the number to determine “Alright, we’re missing this percentage of the population. How do we estimate how many copies we missed?” If they’re doing their job correctly, which we think they are, they are going to capture the sales and project for the sales that they’re missing.

Now, for sales that are just some dude selling something on eBay, no, we’re not gonna be able to capture that. If it’s someone selling something at a show, actually those sales can be reported by SoundScan. We accept venue sales all the time from acts on the road. All you have to do is set up an account with SoundScan, and you report your venue sales each week. They have a process in place to collect the receipts, you show them the ledger of what’s been sold, blah blah blah.

Sometimes we see acts pop on to the Heatseekers chart primarily because of venue sales. You can look in SoundScan, and see each week, they don’t have that many [sales], but the sales you see each week, “Oh look, it’s in this region. The next week they didn’t sell any in that region, but now they’re over here, in San Diego. The next week they didn’t sell any in San Diego, but now they’re in LA.” It’s because they’re moving through the country, selling a handful of copies at all these different locations. Sometimes that can help you have a nice robust number, and that can attract attention, and maybe can get you a bigger label deal, etc. But yes, we do accept venue sales, and SoundScan does that all the time.

Are you aware of any charts other than Billboard’s that focus on extreme music?

I’m not a metalhead, so I don’t know. I know there are tons of publications and websites that deal specifically with metal and hard rock music, but I’m not immediately familiar with other charts that deal specifically with hard rock and metal. 

I used to subscribe to Rolling Stone, and there would be a snapshot of what’s selling well at one specific store in North Carolina that might be metal-focused. Or you go into Amoeba, and they have a list of the Top 10 metal records from that month. 

I guess that’s useful, but we’ve already said how old we are, because we’re talking about physical albums at a brick and mortar retailer. Ooh, how old-fashioned is that! There are hardly any traditional record stores left. It’s just so hard now to say “This is the definitive way that we should rank the top metal acts.”

I’m trying to think what would be most useful to a band. I would think it would be most useful to find who just clicked on my YouTube video? Who just clicked on my lyric video? I’d imagine a lot of these metal bands, if they’re young enough, they’re really in tune with social media and the internet, and digital stuff. What would be incredibly useful to me is using Google metrics, and YouTube metrics, to find out where my song as played, and who’s clicked on it, who’s watched it so many times, and how many different playlists have I been added to. Who’s subscribed to my channel. Maybe I can route our future climb-in-a-van tour towards those cities. Just in the same way you’d use SoundScan information to do that, this could be on a much more granular level, visually-speaking.

But that isn’t like a public chart. It’s not something just anyone can access on YouTube or Google Analytics. It’s just focused on your own material. If you say “Here are 10 other metal acts that are like me. I want to look at how they’re trending on YouTube and Twitter and Facebook across the US.”

How can a band get that kind of information?

We have great social charts that track the most social artists – the Social 50 is what it’s called – it’s a company called NextBigSound [that provides the data]. They scrape data, and they also have deals with different providers where they access Facebook information, Instagram information, YouTube, Vevo, Twitter…the really big, obvious ones. And they look to see who’s up, who’s down, who had more likes than normal this week, who had more fans subscribe to them and so forth.

NextBigSound is a really great resource for artists. If you subscribe to NextBigSound, you can access a lot of data that they collect for you. It’s kind of like a Nielsen SoundScan, it’s kind of like a Google, where they can provide you with a lot of information that can help you better plan your plan of attack when you roll out an album or a song. We use that information to make a lot of our social charts.

Are there obvious trends that Billboard charts can’t capture?

I’d like to say that we try to capture everything. If we notice something we’re missing, or that is a new trend in the industry, we try to get a chart going that would capture that. If you’ve seen how our charts have evolved over the decades, back in the ‘40s and ‘50s we had jukebox charts, we had charts that charted what DJs would report as being played, and what was the top-selling songs in retail stores.

Eventually, that jukebox chart went away, because jukeboxes didn’t exist so much anymore. But then we started tracking all these different genres of music, because the different genres started to grow and got bigger. We have more genre chats now than we’ve ever had before. Back in the early ‘90s, we had country and R&B, we had classical, we had jazz, but it wasn’t until later that we had blues charts and new age albums and world albums and all these different genres that became more and more popular, and we were able to track them.

The same thing happened with YouTube and streaming and Spotify, all these different ways that people are consuming music. We were able to create charts and work with the industry, work with the people that collect this data to present charts to the public. If we ever see something that is not being captured by us, we certainly try to gauge if that is something we should be capturing, and if so, then we try to make it happen.


Visit Billboard online at

Follow Keith Caulfield on Twitter yonder: @keith_caulfield

See everything he’s ever written for Billboard here.