How many Firefox users have add-ons installed? 85%!

The revamped Add-ons Manager in Firefox 4 provides a rich experience for add-on discovery and management, and to power that service Mozilla receives aggregate data on add-on usage. We previously estimated that at least a third of Firefox users had chosen to install an add-on, but knew the number was higher than that.

We’re very excited to report that 85% of Firefox 4 users have chosen to install an add-on — that’s more than 60 million users every day! This number doesn’t include Personas, and even excludes popular add-ons bundled with other software that the user hasn’t actively agreed to install.

Graph of add-on usageFirefox 4 desktop add-on usage from March 22 through June 19, 2011, excluding Personas and several commonly-bundled add-ons

When we first saw this number, we expected it to drop as more and more users upgraded to Firefox 4; remarkably, it has stayed between 89% and 85% since launch in March. It’s also interesting to note that on the weekends when Firefox (and Internet) usage slumps, the percentage of users with add-ons increases, indicating those who browse on weekends are more likely to have customized Firefox.

We’ve also learned that on average, users have 5 add-ons installed, with the following distribution:

Graph of add-on distributionFirefox 4 desktop add-on distribution from June 19, 2011, excluding Personas and several commonly-bundled add-ons

With more than 2.5 billion downloads and 580 million add-ons in use every day in Firefox 4 alone, there’s no better time to browse our gallery and make Firefox your own.

17 responses

  1. Nils Maier wrote on :

    Have those users actually always “chosen” to install at least one add-on, or do the numbers also include the “force-installed” add-ons, such as Java Console, Skype Toolbar, .Net stuff, Test Pilot/Feedback and others?

    1. Nils Maier wrote on :

      Just noticed the figure captions saying “excludeing […] commonly-bundled add-ons”. Does this refer to those I called “force-installed”?

      1. Danny Moules wrote on :

        “excludes popular add-ons bundled with other software that the user hasn’t actively agreed to install.”

    2. pd wrote on :

      Yes, I’ve chosen since day one to install add-ons because without them, Firefox is a pretty ordinary slow, memory-hogging browser. Agree your point about surprise installs is a valid one though. But let’s not blame-the-big-evil-boys with their annoying toolbars and dodgy plugins that install without the user knowing. Sure that’s bad practice but people install add-ons because Mozilla deliberately simplifies and dumbs down Firefox on the basis that if you want more, add-ons are there. Clearly people want more from Firefox and it’s these people who get screwed every time there’s a so-called major release and their add-ons die.

      If I see Firefox promoted as a great browser – and also you can extend it more – once more, I’m going to go bananas. This spin should be accompanied by a timeline graph with functionality on axis and time on the other. Functionality for the average user with a handful of extensions will peak for a fortnight in between these new *major* releases where add-ons devs wake up maybe within a month of the *major* versions release, bump their maxVersions, and then low and behold, Mozilla releases another *major* version. The graph line plummets until it slowly grows as add-on devs wake up again.
      The functionality graph will look more like a heartbeat than a ever-climbing road representing new functionality always being added to the core underlying stuff.

  2. mucinch wrote on :

    Wow! This confirms popular notion that most Firefox users are crazy about their addons. I’ve already seen many cool addons starting to launch first on Chrome and many existing Firefox addon authors not updating their versions diligently. There is also the new rapid release process further creating problems for addon authors. What is Mozilla going to do to keep its developer ecosystem healthy and vibrant?

  3. pd wrote on :

    You didn’t know that your minimalist make-it-like-IE philosophy would make it highly likely people would want more? Duh!

    Now explain the quick-release logic when instead of getting something useful in a MAJOR version, I am stuck with a browser that has MUCH LESS functionality from a new MAJOR version, until extension hacks finally get around to verifying that most likely their add-on still works and kick their maxVersions?

    Mozilla should have systems to analyze the functionality that each published-via-AMO add-on relies on. If a change is made between versions that will effect an add-on, keep the maxVersion block in place. If not, just allow addon X with a maxVersion 4.0.* to work for Firefox 5 if it is not using any functionality that is changed in Firefox 5!

    Surely this is not that hard and it’s a decision that should have been pro-actively taken before the rapid-release schedule was decided. It’s really bad form for Mozilla to keep making the lives of users who are simply looking for a better browser, harder. You’re barely keeping up the pace with tools like the add-on compatibility reporter which is for all the geeky types to check their compat. Ordinary everyday users should not have to find out about *another* add-on in order to test their current add-ons for version compat.

    Get your act together Mozilla. I love ya but geez you can be as slow as buggery to do the most fundamental pro-active user-assisting ideas. You’ve got to change your mentality from an org who is there to push to big boys into action to one that is there to lead the field and challenge the rest to come along. No wonder Google didn’t extend it’s relationship with Mozilla from sponsorship to supporting the whole of Firefox as a Google-level product. Mozilla is too slow and lumbered with old code for Google. Stop trying to artificially mimic Google’s pace when you haven’t first solved your own lack of innovation speed and cultural us vs them underdog issues.

  4. Steve wrote on :

    Suggestion: If a given add-on is installed on at least 50% of Firefox installations, its functionality should be rolled into the browser itself.

    1. David wrote on :

      This is an intriguing suggestion and I totally agree with the sentiment. I would just note that the most popular add-on, ADP, regularly has between 11 and 14 million daily users according to AMO and that falls far short of your nominated threshold.

      https://addons.mozilla.org/en-US/statistics/addon/1865

    2. Johan Sundström wrote on :

      The reason this is a bad idea is that it takes away freedom of choice, and because the not having a particular feature, is a feature many people enjoy. The common base line most can agree on, is that the feature “hogs memory, CPU and user interface real-estate and adds user learning curve” (which is common to all add-ons and shipping features), counts as a negative, which is only made up for by whichever net positive that thing brings to yourself specifically.

      The way to celebrate this freedom of choice is not to take it away, but to make it easier to do, while retaining the essential freedom element, and to find the things that are good.

  5. Josh wrote on :

    Very interesting stats, thanks for sharing.

  6. Daniel Jonsson wrote on :

    Really nice numbers!

  7. Tony Mechelynck wrote on :

    I’ve always understood add-ons as the main strength of Mozilla applications, and this means:
    – Netscape 6 and its gallery of themes
    – Firefox 0.9 and its extensions; at the time it was still practical to review “All” add-ons from A to Z in alphabetical order. What a change! (and IMHO for the better) 🙂
    – Thunderbird 1.0 and /its/ extensions
    – SeaMonkey 2 and its addons; nowadays I have one constantly opened tab where I display the “most recently updated” extensions at AMO.

    When talk came around of some people’s intentions of replacing the whole extensions subsystem by only dumbed-down restartless extensions which would be able to do only a small part of what “traditional” extensions can do, and dumbed-down Personas which are far from able to change the app’s look&feel the way a “traditional” theme can (especially since they cannot be combined with any theme other than the default one), I feared for Firefox’s, Thunderbird’s and SeaMonkey’s future. Happily, AFAICT it was only a false alarm, and (again, AFAICT), powerful addons “which need restart” are here to stay.

    Your statistics make me wonder, however: is it only because of this long love story between Mozilla add-ons and me, or am I really eccentric? I’ve just counted 37 « true» lines in the “Extensions” part of the clipboard dump from about:support (i.e., 37 enabled extensions, including those bundled with SeaMonkey but not the current theme and also not the language packs).

  8. Tito wrote on :

    I’m skeptical of those numbers. The vast majority of FF users I see don’t even know what an extension is.

    Although in the last few years casual users have been switching to Chrome (or staying with IE), but still, 85% seems to high.

    I know you said you excluded some commonly/automatically installed stuff, but there’s a lot of it you have to take into account. From .Net to various toolbars to Nokia Suite etc.

  9. jrk wrote on :

    85% Firefox users have add-ons installed = 85% Firefox users have (an extension OR a theme OR a langpack) installed?

  10. Linda wrote on :

    I love the add-ons…my main issue here is..I have Firefox 5….I love to play my games..is there someone who can tell me what absolute add-ons and extensions I need to achieve this goal with no conflict? Thank you soooooo much for your help!!!!!!!!

    1. patrickjdempsey wrote on :

      Typically problems with games online for the last year has been because of hardware acceleration in Flash itself. Right-click on your game and choose Settings and uncheck hardware acceleration.

  11. bajro durakovic wrote on :

    FIREFOX Best in Class!!! NUMERO UNO!!!

    FIREFOX best in Class!!! NUMERO UNO!!!

    FIREFOX Best in Class!!! NUMERO UNO!!!