Quick guide
The short answer
Mac apps do not all get updates from the same place. This report groups app reports from August 2026 by their recorded update source. It is not a survey of every Mac, a count of users, or a ranking of update tools.
The short version
We counted 56,002 saved app reports from August 2026. They cover 9,171 app identifiers. An identifier is the internal name used to group reports about the same app. These are not counts of people or Macs.
We put each app identifier in one of 10 groups, using its most recent reported update source. The groups add up to the app total. They describe this sample, not every Mac's app collection.
What this report measures
The reports cover 2026-08-01 up to, but not including, 2026-09-01, in UTC. We collected these saved reports on 2026-09-07.
An app can be reported more than once. Repeated reports do not mean more users or more unique apps. Development activity and changes to reporting can affect the totals.
This sample includes reported App Store apps. An app reported in August may also be in today's catalog. The figures are not a measure of market share or the number of apps missing from the catalog.
The headline numbers
The most recent reported update source for each app identifier, August 2026. Each identifier appears in one row only.
| Reported update source | App identifiers |
|---|---|
| Direct | 6,488 |
| App Store | 1,330 |
| Sparkle | 1,329 |
| Adobe Remote Update Manager | 11 |
| Homebrew Cask | 6 |
| Homebrew Formula | 2 |
| Microsoft EdgeUpdater | 2 |
| Docker Desktop | 1 |
| Google Software Update | 1 |
| Microsoft AutoUpdate | 1 |
| Total (each identifier counted once) | 9,171 |
451 app identifiers had more than one source label during the month. We used the latest report so we did not count them twice. Direct is a label reported by the app. It does not prove that the app has no other updater or is unsafe.
What the catalog lists
We also checked macCurrent's catalog on 2026-09-07. It listed 5,845 app records marked verified or vendor-managed, each with an enabled update source. That is not the number of apps macCurrent can install automatically.
An App Store entry in the catalog does not mean your Apple Account or Mac has an update available.
| Catalog source | App records |
|---|---|
| app store | 5,657 |
| homebrew cask | 123 |
| vendor managed | 25 |
| github releases | 22 |
| manual version page | 21 |
| sparkle appcast | 12 |
| vendor download redirect | 7 |
| tauri updater | 3 |
| brave versions | 1 |
| capture one releases | 1 |
| cua driver releases | 1 |
| duckstation releases | 1 |
| electron releases json | 1 |
| folge changelog rss | 1 |
| installomator label | 1 |
| lm studio download page | 1 |
| maxon cinema4d downloads | 1 |
| rpcs3 update api | 1 |
| warp stable versions | 1 |
| zevrix version control | 1 |
35 apps have more than one source in this table. The same app can appear in more than one row, so do not add the rows to get a unique total.
The separate Homebrew list contains 7,913 casks. Of those, 6,737 are not disabled or deprecated. Being on that list does not mean macCurrent has reviewed an installation method for every cask.
Why the sources differ
Apple manages App Store updates. Homebrew manages its own packages. Other developers choose a built-in updater, a separate tool, or a website download.
Each one checks a different set of apps. See the update-source guide for the steps you can use on your Mac.
What these numbers do not tell us
An app's reported source does not prove that an update installed successfully. A Direct label does not mean the app has no built-in updater.
These figures do not measure market share or show which updater is best. A fair comparison needs the same test apps and checks of what each tool actually did.
What this means for your Mac
If you check only the App Store, you can miss updates from other places. If you check only Homebrew, you can miss apps it does not manage.
The useful habit is simple: know where each app gets updates and check that source. Start with how to update your Mac apps.
How we counted
We group reports by their app identifier, ignoring capitalization. Reports without an identifier are left out. Versions, builds, signing developers, and installations sharing an identifier count together.
For the source table, we use the latest report for each identifier. If report times match, the higher report ID wins. The catalog table uses different rules and can count an app in more than one row.
A reported source is not a test of whether an update installed. The downloads contain totals only, with no personal or individual app records.
About the earlier July table
The July table listed 2,036 app identifiers, but its source rows added up to 2,045. The published totals did not explain the overlap. This report replaces it with a dated sample and an explicit counting rule. It does not claim growth since July.
Download the numbers and counting rules, JSON or the tables, CSV. Both use the same saved data as this page.
How macCurrent helps
macCurrent finds updates across several sources and installs supported updates. It shows which updates you can install and which need the App Store, the developer's updater, or a manual step.
Supported direct replacements can have a backup and saved update history. Other apps still need their own updater. For the current release, follow the installation steps to try it.
Frequently asked questions
Why do Mac apps use different update tools?
They come from different places. The App Store, Homebrew, and individual developers each manage their own releases.
Do these numbers show which updater is best?
No. They group reported update sources. They do not compare successful installs or rank products.
Where does the data in this report come from?
We grouped 56,002 saved app reports from August 2026 into 9,171 app identifiers. Each identifier uses its latest reported source. These totals describe the sample, not people or market share.
How often is this report updated?
This edition covers August 2026, with a separate catalog check from 2026-09-07. A later edition needs dated data and the same counting rules before we can compare results.
Can I help by reporting an app?
Coverage reporting can help us find apps that need review. What is sent depends on the reporting settings. The public report contains totals, not personal app records.