I’m in the process of testing and integrating a library into my plugin project, which has a requirement of macOS 12 Monterey or later.
I’ve tried to find reliable data on the market share of macOS versions, but for some reason everything past 10.15 Catalina seems to get labelled as Catalina.
I’m trying to ascertain what percentage of users I’d be excluding by requiring macOS 12 for my plugin.
I can see if I can write a small SQL query to get that information. But keep in mind that our target audience might be completely different than yours, so take all the data with a grain of salt.
May I ask what library that is? Targeting such a “young” OS version seems unnecessary.
macOS 10.x = 31%
macOS 11 = 8%
macOS 12 = 23%
macOS 13 = 25%
macOS 14 = 10%
So, if you cut off at macOS 12, you lose 39% of your potential customers.
Caveats: the data is for our customer base. Mostly EDM producers. Some of the queried entries might be outdated (once registered computers, no longer used, or upgraded in the meantime).
But I think expecting a 1/3 of your customers using outdated hardware/software is pretty realistic.
One of the library maintainers told me it was macOS 12, but in fact it builds fine for 10.15 and later. So that might be acceptable!
I found they were using std::filesystem::path which has an alternative: GitHub - gulrak/filesystem: An implementation of C++17 std::filesystem for C++11 /C++14/C++17/C++20 on Windows, macOS, Linux and FreeBSD.
I’ve suggested this to them, hopefully they accept and I can continue to target down to 10.9 as I currently do.
You can start with 10.11. Everything below is a pointless exercise in keeping compatibility with 0.1% of customers. Not worth the effort.
@reFX thanks for sharing that data. Is that data for people using your latest version of your plugins. Or is that just active users of any version? And over what sort of period?
Really cool of you to share this.
Would be nice to collect some more data points from others.
Also curious about the 10.x breakdown (curious if the bulk of those are >=10.13)
I actually realised I also have some data on this too 
MacOS %
10.11 0.25
10.12 0.87
10.13 4.46
10.14 7.43
10.15 10.27
11 9.4
12 35.27
13 26.11
14 3.34
That’s over the last 30 months, for all versions, all products. But if people update their OS, but otherwise the hardware stays the same, the activation is overwritten, so this is not a log of what happened historically, but rather a current state.
Our numbers for 10.x are:
10.11 0.3%
10.12 0.7%
10.13 3.9%
10.14 6.9%
10.15 14.7%
So, it is largely in line with @adamwilson numbers. This is nice to have as affirmation that our numbers are at least representative of the market and thus can be used to make development decisions.
I surely wouldn’t drop 10.14 or 10.15 support, but 10.11 or 10.12? Together they represent only 1% of our customers. It’s not fantastic, but if some JUCE feature requires us to use at least 10.13 or something, I wouldn’t even think about it for too long.
You guys seem to like this kind of statistical data.
The numbers are absolute, the bars are relative, to the top.
So for FL Studio the bar is fully filled, as it’s the most popular DAW (for NEXUS4 users), and Logic is 91% of that, so its bar is slightly shorter.
We can track Apple Silicon adoption pretty well with that. As you can see only 38.79% of our customers use native Intel, the rest are already using Apple Silicon.
Very interesting indeed, thanks for sharing!
I’d be curious to see Windows version numbers if anyone is willing to share!
Our numbers are mixed. You can see the “os” column. 54.21% is macOS and 45.79% is Windows users.
Here are updated stats from today
Ah, sorry, I meant versions of Windows (8.1, 10, 11, etc)
