We are near the end of the 2019 S.P.IN. submission drive. Thanks to everyone who has contributed. Don’t forget to check out the total number of submissions, leaderboard, and recent submissions here. You can see the positive impact of S.P.IN. on driving more submissions below.
However, there have been fewer submissions in 2019 compared to last year, a first for Discogs.
Earlier this year, we examined this trend to see if we could identify where the growth is slowing down. Some suspicions for the decrease included the blocking of bootlegs in 2017, an ever-shrinking amount of releases not in the Database as it grows, and censorship in some countries. However, we learned quite a bit and uncovered new trends while investigating, and wanted to share an update here.
To start, we have not seen a dip in unique contributors. We have seen a 1.1% increase in unique contributors to Discogs in 2019 compared to 2018.
And those contributors are still adding to the Database, albeit many of those contributions are starting to shift from new release submissions to filling in missing information, like credits and images. In fact, Image Additions are up by 2.5% YOY.
While Release Edits are up by 16%!
These edits are incredibly valuable. Often times even more information is added during second or third edits than during the initial submission. Diognes_The_Fox was able to provide an example of when Release Edits provide more information than the Release submissions itself.
Releases by Year of Submission
We first looked at how the year of publication for releases was distributed. In the graph below, the X-axis is the year the submission was added to the DB. Each block of color is a decade. We can see here that there is a “cooling off” regarding submissions of catalog releases (defined as releases older than 18 months from initial release date) since 2017.
When we break down the same stats as the last graph but unstack the years, we can see that by far the largest percentage of submissions in a year are from that year. For example, in 2018, the highest share of submissions by year of release was 2018. In addition, each year of release decays to a lower level in following submission years.
It has never been easier to release music. With distribution services like Soundcloud and Bandcamp, affordable home recording options and audiences that are receptive to diverse sounds there is more music being released this year than any other year in history. We can see the culmination of this by looking at the share of releases in the Database by year. This trend appears to accelerate as early as 2010.
Sub-Pop Catalog Case Study
We also wanted to get a sense of how accurate and complete the Database is for newer releases. To inspect this, we matched the Database entries for famed Seattle-label Sub-Pop with their official 2019 release list.
When we pulled this reporting, the Sub Pop label had reported 42 releases for 2019. 86% of those 2019 releases were cataloged in Discogs already. The 6 not in Discogs were all digital releases. Surprisingly – the average time for the release to be submitted to Discogs is 2 days prior to the official release date! Some might cry out web subs!, but keep in mind that some Discogs users work at record shops, content houses, labels, etc. and have access to albums earlier than the official release date.
Releases by Country
We also inspected submissions by country. If you haven’t heard, we’ve officially translated Discogs into eight languages, including many of the Help and Support documents regarding submissions. This was prioritized after finding that users from countries where English is not the native language are much more likely to land in CIP. Our hope is by translating these guidelines, we can help users in countries that are underrepresented, which was a goal of last year’s S.P.IN. drive.
Below is a graph of the country of release vs the date that release was submitted. Keep in mind that country is not a required field, so earlier submissions typically have more country information.
We can see that releases from the UK were dominant in the early and Electronic-centric years (2000 to 2006), then the USA took over. The top three by a substantial margin are the USA, the UK and Germany. Japan was showing growth but has been dropping as of late.
Looking at countries 11-20, we see some bigger swings. The blue in the graph below is Russia – what could be going on? We can see that Greece dips then recovers. Also, what could be happening with Jamaican releases? Their last remaining pressing plant closed 2015 but opened its doors up again recently, albeit it looks like at a much lower volume?
The distribution between the top 43 countries, as shown below, is starkly uneven. Discogs has traditionally been an English-dominant website. Our hope is making the website more accessible to users of more languages will help diversify the information available in the Database. If you think you can help translate Discogs, join our translator community. It could be that we’re reaching a threshold in some of these countries but have ample opportunity to find undocumented releases in others.
It’s likely no coincidence that our Marketplace policy banning bootlegs, launched mid-2017, coincided with a drop in unofficial release submissions.
Are we worried? Low-quality bootleg CDs etc, possibly even created for selling on Discogs, are not something we want to encourage or be complicit in.
When we segmented by genres, we also found some changing trends. Below are release totals of those tagged with the top-4 genres in the Database. We can see the number of Rock submissions falling. Have we reached ‘peak rock’? It’s definitely something that is possible, but we need to dig in more on this to be sure.
Inspecting some of the other genres turns up some interesting finds too. Classical music has been climbing and recently surpassed Funk/Soul. The Jamaican drop mentioned above, here connected to the Reggae genre, is apparent.
We’re seeing a similar downturn in our two most-prevalent formats – Vinyl and CDs. Cassettes, however, are looking strong in 2019!
So, an interesting state of the Database in 2019. A dip in new submissions, surging release edits as contributors fill in some of the missing fields, and a look into unequal distribution by country and genre. Do you have any observations on the state of the Database in 2019? Share them with us below!