Local Current Blog

40 years of album sales data in two handy charts

Earlier this week we asked you to guess the highest-selling singles and albums to come out of Minnesota—and the answers to those two big questions will be revealed this Sunday night on the Current Presents: The Best-Selling Minnesota Artists of All Time.

But before we get down to brass tacks, I wanted to share a few notes from my research. This kind of stuff can get pretty dry, so I’ll try to keep it as lively and interesting as possible with the help of some pretty fascinating graphics from the Recording Industry Association of America (RIAA). Everyone likes graphics.

Who tracks album sales?

When I first had the idea to compile a list of best-selling albums and singles from Minnesota artists, I thought, Why hasn’t anyone done this already? Surely this will be the easiest information-gathering of all time. And when I think back to that pre-research time now I laugh and laugh at how blissfully naive I was about the music industry and its data.

The truth is, aside from the very trackable sales happening in the modern age through outlets like iTunes, no one actually knows exactly how many copies of a particular album or single have sold. Let me repeat that: NO ONE ACTUALLY KNOWS exactly how many copies of a particular album or single have sold, not even the artists themselves. There are a few reasons for this.

Prior to the RIAA forming in 1952 and beginning their process of certifying gold albums in 1958, only the record companies themselves were keeping track of how many copies of each title had been manufactured and sold. And at that point in time there was really no way of knowing what happened to the albums once they left the manufacturer and were purchased wholesale by the individual record stores—maybe they all sold, maybe some of them ended up in dusty bins underneath the shelves. Also, not to say that record companies aren’t run by upstanding and truthful people, but there was nothing preventing the labels from inflating their own sales numbers to make themselves look good. As such, any information available about album sales prior to 1958 is uncertifiable.

Then along came the RIAA, whose original mission was to “administer recording copyright fees and problems, work with trade unions and do research.” After a few years on the market the RIAA decided to start rewarding singles and albums that performed especially well. In 1958 they began offering Gold Awards for singles and albums that had reached $1 million in sales, and after a decade and a half they changed the criteria to 500,000 units.

This was good thinking on the RIAA’s part—back in the 1950s, a 45rpm vinyl record could range anywhere from 50 cents to a dollar or two each, so it was difficult to compare artists in terms of dollars earned—and by the time we get to the mid-’70s we start to see some awesome, easily trackable data regarding the music industry at large. Which brings us to our first chart!

All data courtesy of the RIAA

All data courtesy of the RIAA

On the left we have a pie chart that shows the breakdown of all the different formats of music that were being sold in 1983, and on the right we can see how drastically things had already changed 10 years later. And this is where things start to get interesting.

Before we get too far, another note about tracking sales: In addition to the RIAA, Nielsen SoundScan began providing sales data to record companies and industry insiders in 1991. The two companies gather their data a little differently—the RIAA tracks the number of units shipped from the manufacturers, minus unsold units returned to the manufacturer by record stores, while SoundScan gathers the data directly from the cash registers at stores. Because of this, Nielsen SoundScan’s data is much more accurate and precise, but it is also more closely guarded. Nielsen charges a significant fee for researchers who wish to access their database.

Ok, now on to the good stuff.

40 Years of Data

Over the course of researching the best-selling artists to come out of Minnesota, I spent a lot of time digging into the decades of data that have been gathered by the RIAA. Much of that data was specific to individual artists (and will be revealed on Sunday night), but I also found some really intriguing charts and graphs that relate to the format music has been sold in over the years and how overall sales have fluctuated.

Are you ready for another chart? I know I am.

units-1998-2003Here, we see a trend starting the develop that builds on what we saw in the above 1983 and 1993 pie charts. What is that trend? That for a period of the 1990s and early 2000s, people were buying the highest proportion of a single format of music than had ever been tracked before. The compact disc was not only the most popular format, but for a brief period of time it was the only format people were purchasing.

And in the 2000s, things shifted dramatically again:


Because of this, we’ll see some surprisingly results pop up when we examine the best-selling singles from Minnesota vs. the best-selling albums. In some years it was easy for a single to sell a million or more copies because that’s what people were buying, where in other years it was virtually impossible to purchase a single song and fans were forced to buy the entire album. I’ll have more on that and how it relates to Minnesota artists’ sales very soon.

But while these individual pie charts are interesting and all, one of the most startling and compelling findings in the RIAA database was a graph that shows the rise and fall of all of the formats of music over the past 40 years. And this is where things get CRAZY.


On the left, the RIAA has tracked the total number of units sold in the U.S. each year, broken down by format, and on the right they have tracked the sales in terms of dollars. You’ll notice that the number of units purchased has increased steadily from the time they started collecting the data until now—including a huge leap in sales over the past seven years as people download more and more music online. But on the left, what we see is that while the music sales are currently at a 40-year high, the money brought in from these sales is the lowest its been since 1988.

Here’s that chart again, only this time adjusted for inflation:


Suddenly, all of the industry’s freaking out and melting down over the internet’s impact is starting to make a lot of sense.

Tune in this Sunday night at 10 p.m. for the Current Presents: The Best-Selling Minnesota Artists of All Time, and check back tomorrow for even more nerdy analysis of music industry data as it relates to local musicians.





  • Alfredo Rodriguez

    The precipitous drop in units and dollars from 1999ish coincides with the release of Napster. People decided to stop buying the cow when they could get the milk for free.

    • Evan Olcott

      Then why didn’t units rise when Napster closed in 2001?

      I think the important point to take here is that the distribution system that was to be “the way it should be done” was under massive resistance by the industry itself, and they missed the boat. It’s notable that a *computer company* had to step in and make things right.

  • Lisa Lendway

    Can you give a link to the actual data? Good article, but I’d love to improve the graphs and use it in a class I’m teaching … yeah, I’m a nerd.

  • lara c

    What’s more surprising is there are still people paying for music. Especially for songs that play 20 times a day on the radio and everyone has grown sick of.

  • Jonathan Hamlow

    I wonder first of all how encompassing the data is. I haven’t personally bought a new CD in probably over 5 years, but I purchase at least as much music as ever these days, so I fit the trend. I pay for my music. BUT – I’ve bought more music from non-RIAA member independent artists either directly via their website stores (artists like Brad Sucks or Andrew (“Songs to Wear Pants To” Huang) and I’m pretty certain nobody’s tracking that. I’ve also bought a ton through Bandcamp sites and I wonder, is this data tracking that? Often in these cases I’m paying only between 10 and 75% of the typical cost of an album download from the majors (Amazon, Google Play, iTunes, all of which I still use) but I know for a fact the artist is receiving nearly 100% of what I pay and often ten times what a label would “trickle down” to the artist.

    Maybe I’m not that typical but I wonder if anyone is really trying to measure the impact of this sort of commerce outside the mainstream of what the RIAA considers to be “the music business”. With some of the major artists of the day opting out of major labels (Bon Iver) or away from labels entirely (Macklemore) the RIAA model is only becoming more irrelevant. No matter what though the brief golden age of the CD, when people were willing to purchase a thousand million CDs at $14 a pop are gone forever and nothing could ever bring them back, even if you could somehow stuff the genie of digital distribution back into the bottle. Who knows how sustainable it ever was since a fair piece of it was likely people repurchasing their old tape/LP collections on CD – digital collections anyone savvy enough to rip to their hard drives and back up properly will never, ever buy again. There are many other demands for the entertainment dollar (video games, streaming services, expanded video options) and the CD is a relic of the media-centric past.

  • jon_howard

    Would offer up another thought, which says this is nothing to do with downloading per se or even whether people are paying, it’s about the disagregation of albums and a change in listen habits that is now only interested in individual tracks. In both value and volume terms, album sales (formats combined) had been increasing to the turn of the century. Then you have the Napster blip. But what has never recovered for the digital download generation is the sense that albums are a holistic artifact to be experienced in its entirety and in a programmed order. Once you lose that way of consuming music you see a steep rise in units (in the old world a single and lp were both one unit) and a parallel loss of all the value.