New WordPress 3.3 is just around the corner, and no matter how good new version will be, in the next couple of months less than 15% of all WordPress websites will upgrade to 3.3. And if you look to the past versions of WordPress, you will see that adoption rate looks pretty bad.
Currently all my plugins support WordPress 3.0 or newer (except one that depends on features added in WordPress 3.2). But, with the soon to be released WordPress 3.3, I would need to support 4 different major versions of WordPress. So, I was making plans to drop support for WordPress 3.0 when the WordPress 3.4 goes into development next year. But, I wanted to research this a bit to see how many websites are actually using each of WordPress versions. WordPress.org statistics page has WordPress versions distribution data, and you can see this data on the chart below.
To my surprise, WordPress 3.0 is still very dominant, while 3.1 and 3.2 together are used on about one-quarter of all WordPress powered websites, and all older WordPress version prior to WP 3.0 are used on 22.6% of all WordPress powered websites. WordPress 2.9 is almost 2 years old, and yet 11.1% still uses this version. But, for me, the most alarming thing is that WordPress 3.0 is still the most used version of WordPress.
So, what is the problem? Why global adoption rate of new versions of WordPress is so slow? I did some calculations based on the latest download counter on WordPress.org. As of today, WordPress 3.2 has 13.000.000 downloads. While that number looks impressive, in terms of adoption is really not. This number includes both 3.2 and 3.2.1 versions. Let’s assume that many users downloaded WordPress more than once, for development servers or some other reason (maybe 1.000.000 duplicated downloads, well, I have downloaded each of WP 3.2 versions at least 100 times for development). Also, let’s assume that people who decided to upgrade to WP 3.2, upgraded to 3.2.1. Some were switching to 3.2.1 directly, but that is hard to figure out. So, let’s say that based on number of downloads 6.000.000 websites actually upgraded to 3.2/3.2.1. Still looks impressive? Consider that official WordPress stats say that there are 35.000.000 websites using WordPress (this doesn’t include WordPress.com), this is only 17% of all WordPress powered websites switching to WP 3.2, and this is in line with 16.2% from the global usage chart on top of this article. As I said, 13.000.000 downloads doesn’t look so good at all. All these numbers and calculations are not very precise, but they are in the ballpark to illustrate the state of WordPress upgrades.
These numbers are for all the websites WordPress.org gathers data from. On the other hand, I have made a similar chart based on data gathered by Dev4Press website from users of our Pro plugins and themes. Thankfully, this shows the desired trend of adoption, and WP 3.0 is going down, with WP 3.1 and WP 3.2 on the rise. If this trend was to continue, WP 3.0 will almost fade out for Dev4Press users in the next 6 months, before WP 3.4 is released. Good to see that WordPress 3.3 in Beta stage is already in use.
This big difference in usage shows that Dev4Press users, in this case, are more informed about the current WordPress development than the general WordPress user base. I expect that any other WordPress based business can expect similar data and trends for their users. These are more technically oriented users or users ready to embrace the progression of WordPress development, and they are upgrading websites to take advantage of new features or to make sure that they have latest security improvements and bug fixes.
With WP 3.3 coming soon, we still have 29.000.000 websites that didn’t upgrade to WP 3.2, and almost 7.000.000 that use pre WordPress 3.0! Why is this so bad? Well, there are few reasons:
- As with any other software, it is a fact that old versions have bugs and security risks. This is true for WordPress also, and we have millions of vulnerable websites using WordPress. While it is a great thing to have big part of the Internet powered by WordPress, it is frightening to know that most of these websites are not secure. Bug fixing for new versions are great, but they don’t mean much if most website will not upgrade anyway.
- This makes plugins development very hard, considering that developers need to support several very different WordPress versions, or to decide to drop support for older versions. This can be a good thing, if the users are forced to upgrade. But, many websites are postponing upgrade because of old and outdated plugins, and this is main reason we still see 2.3 or 2.5 on the chart. New WordPress and old plugins is an old problem, and I don’t see solution any time soon.
- Money. Yes, for old websites dependent on old and no longer developed plugins, moving to a new WordPress is not cheap. In some cases, new plugins with similar functionality can be used. In most cases even if there is a new and different plugin, it will be incompatible with old plugin. So, website owner need to pay, in some case a lot of money to get old plugins fixed and upgraded, or rewritten.
- WordPress development is not really driven by the needs of wider community. Many website owners I talked with, that use old WordPress, don’t feel the need to upgrade, since they have all they need with WordPress they are using. We may like it or not, but frequent interface changes in WordPress are not something users like to see, because very often they need to learn things they may not need at all. It is easier when users upgraded to each new WordPress, because changes are comparatively smaller, making a jump from 2.3 or 2.5 to 3.3 now, is a huge change, almost like using completely different system.
- For a community so big as WordPress is, there isn’t a single website that can be considered as information center for all things WordPress and it is not a big surprise that website owners are not well-informed on current status of WordPress, plugins and themes development. And when I get a question from a user: ‘where can I find information on all things WordPress’, I honestly can’t recommend any website. I believe that lack of quality information is one of the major problems with adoption of new versions.
How to solve all these problems? How to improve the number of websites willing to upgrade and how to have a more secure websites running on WordPress. I know about all the obvious methods like better education of users on upgrade benefits, or improved security, but so far, that is not done properly at all. Individual efforts by people with work based on WordPress, have an effect on a very small number of users. Global community of WordPress website owners is not really affected at all by that. There are a couple of things I can leave as a suggestion:
- WordPress.org based information portal would be best source of information. Right now, WordPress.org website is not useful to majority of users. Codex is largely used by developers or users that want to do things themselves, Forum is again a bit too technical, Plugins Repository is mostly a mess, News and blog are very limited in scope. I know that improving that website is not easy, but in the past 2-3 years hardly anything was changed there, apart from few updates to repository and small updates to Codex.
- Slower development cycles. Right now, goal is to have two major WordPress releases each year. That is too fast, too many versions are in use already, and each year 2 more are added to the mix. One new WordPress version each year would be better, to allow users to get familiar with new versions, to have fewer changes and problems to worry about. Current core developers and contributors would have much more time to develop, much less chances to have bugs (most bugs are caused by haste) and maybe to focus their development time to WordPress.org.
- Better literature. I know ofĀ many good WordPress books, but they are mostly written by developers for developers. I am still to see a good book about using WordPress. Even when such books are published, they are already outdated due to fast development cycle of WordPress. Only book that I know to be updated from time to time (still, development book for the most part) is Digging into WordPress by Chris Coyier & Jeff Starr.
- Objective and open news websites. I am sad to say, but that there are no such websites now. There are some smaller websites that try to be informative, but they don’t have enough reach. The big, ‘the best’ WordPress websites are far from that. I know that editorial policy is internal matter for each website, but basic fairness in dealing with WordPress related services and products is not much to ask. I have personally experienced double standards from some WordPress websites and I know many fellow developers that had same bad experience. Whatever influences exclusion of information (money in most cases, or conflicting interests), is very bad for Ā the community and it creates a negative reputation for WordPress.
- Better co-operation of developers. We have a lot of plugins, even very popular plugins that don’t play well with other plugins. In many cases, there is no will to resolve the issue. And that causes problem to end users that try to make their website work after upgrade and are getting broken because of the plugins. Many developers, even developers of commercial products, simply don’t care what they plugin or theme will do to the website and other plugins.
This was a lengthy article, but I think that current state of WordPress and slow adoption is a very serious matter that is not only about the fact that users don’t upgrade on a regular basis, but that WordPress developers and core of the community is not doing the right things to inform and educate about WordPress and related products. And I tried to include all things I considered relevant for the current state.
I would like to hear from you what do you think about current state of WordPress, about updates and how often do you upgrade your websites. What are most common problems you have when it comes to upgrade?
Very informative article. Thanks!
I wonder how many WP sites that haven’t upgraded are really simply “abandoned” projects. I have one or two that I set up on a brainstorm 2-3 years ago that never went anywhere. Currently, they are just “hanging out” there. I’m guessing there are many people with sites like that, even while some of there other sites are very current – even getting beta upgrades. So same cutting-edge user may have some apparently “slow adoption” sites, but really they’re just abandoned and ignored.
I’m guessing the automatic admin interface upgrade notifications should help active sites stay much more current than before. This is very helpful to me.
Finally, I’m growing in my reception of RSS and email feeds from credible WP sources, and hopefully there will be a groundswell of this sort of thing. For example, just last week I discovered the free wpMail.me. I signed up, and when I got me first issue (#26) I was delighted to see the following news item among others: “Two years of Dev4Press”.
Thanks for the comment Jonathan. Good point about the dead or abandoned sites.It is hard to determine how many such websites are still online, but there must be a fair percentage of them.
wpMail is a good source, but it is a weekly choice of articles. Community needs all-round news website that is current, objective and capable of reaching wider user base. I know that it is not easy to create such website, but it would be good to see someone try that.
More Stats:
Thanks for this really interesting stats showing the decreasing interest in getting new versions of WordPress since 3.0 is released. No wonder we see 3.0 as dominating, and 3.1 and 3.2 far behind. As I said, 13.000.000 downloads for 3.2 is not that impressive considering number of WordPress websites and current usage statistics.
I think the ability to easily rollback to your prior version after a problematic upgrade would help. Anything that lowers the perceived risk of upgrade, and makes it simple, would help adoption.
Thanks for you comment. Yes, once you upgrade there is no going back, and than problems can start with plugins or themes. Rollback would be a problem if upgrade contains database upgrade (major versions of WP usually change database). So, right now only solution is to back up everything prior to upgrading WordPress.
This is a great analysis and as consultant I know exactly what you mean.
Slow adoption is also in part due to FEAR. People who have developed their site using WP are simply afraid to take the step to upgrade because it might break their site. They are willing to deal with possible security threat and those willing to pay someone with technical knowledge do, but many just stay on old version.
Education is a huge factor to make technical things simple for people.
Thanks for the comment Alex. And this is especially true for very old websites running 2.3/2.5/2.6 because they are so different from latest 3.x versions.
Yep, and those are the most vulnerable. Unfortunately what many of us, more technical people consider to be a trivial task – upgrade, is seen as HUGE project by blog owners.
I have seen this to be especially true with small business owners. Someone built them a site using WP and left them there.
Considering that we will get 2 WP versions each year, this problem will be much bigger in a year when 3.5 should be out, fragmentation of websites running WordPress will be alarming.
I think part of the problem is with clients not wanting their sites maintained. Letās say I design a site for a client in Jan. By October, there are already several plugin upgrades that need to be made plus any WordPress updates that are needed. So I contact the client and tell them I can update everything for them for $xxx amount but I am not sure if the update will break anything. Given that scenario, why would a client want to update anything on a web site that is working fine in their eyes?
I love the fact that WordPress and good plugin developers are constantly updating and making their products better, but in this case, it is a lot of work and cost involved for the designer to be checking the site weekly and applying in updates.
I have a client still on WP 2.9 and using WPEC 3.7 as their site would require a lot of work to update to WPEC 3.8 since it is so different than WPEC 3.7 and how the theme interacts with the plugin. From the clients point of view what incentive does he has to pay the $500+ to update everything when his site is working fine. From my point of view why should I spent 10+ hours updating everything for free.
To me, this is one biggest of the reason I miss the standard HTML based sites. Once it is completed, very little maintenance is needed.
Ed,
To solve this exact problem we are offering our clients to host their sites for a nominal fee comparable to hosting companies, just slightly higher but in exchange – we take care of all the maintenance and provide support as needed for basic things.
Small business owners love that option
Thanks for the comment. It is hard to convince the website owners that upgrade is needed for whatever reason, security mostly. But, if the upgrade is delayed, it is not only WordPress, but upgrade is needed for plugins also, and that all can cause problems and take much more time, and much more money.
I think it’s that people aren’t aware that there’s a new version out there. I can’t recall being notified of a new update via email (although I could be mistaken), nor the importance of updating being underlined with a thick marker.
So if you don’t login to your dashboard and you see no impact, there’s no reason to update.
Thanks for the comment. All WP versions report only that update is available, and there is no info on the actual update importance. As you say, many users simply overlook that new version is released.
My GD Press Tools Pro in next 4.3 version will have email notification on updates available. I doubt that WordPress anytime soon will have such option.
At least in my case, the problem is that it’s not simple to update. I have to upload the new files ā and unchanged ones, too ā to the server, instead of just being able to click on a link, an option WordPress gives me, but never works. (What’s funny about it, is that automatic plugin updates never fails.) That way, I procrastinate the update. And then, later, I delay the update again because I learn that a new version is “just around the corner”, in a vicious cycle. š
Thanks for the comment. If your core update fails, maybe the problem is with access rights for wp-admin and wp-includes folders. Maybe WP can’t access them to override and delete old files. And yes, it would be great to have less major version updates, one each year would be better than 2, less pressure.
Thank you for bringing this important topic more into focus! You can’t say it enough to clients (especially those) how important updates are! I maintain over 40 WP installations and all are on latest version of course.
To my shame I only had one little client side from 2007 that was running on WP 2.3.3 since then – but without any issues. I just switched it over to latest version, new Theme (Genesis š and all is fine now :). To my surprise there wasn’t any issue regarding security in the last 5 years. But maybe that’s this little exception to confirm the general rule…
It was a nostalgic moment to have such an old dashboard before my eyes during the time of update – and it showed how WP has evolved over the years. A lot of things were possible also in former versions but without most comfort and security and the APIs for devs that we now all have and love :).
I will spread this article to some people so they might better understand all these issues. Once again, thanx for bringing it on the table!
-Dave from Germany š
Thanks for the comment David. Old WP versions usually break on update due to old theme or the plugins. Thankfully, data is compatible and that should be fine.
You’re right, MillaN! I did the upgrade manually to avoid encoding issues – but was no problem because were only 10 pages :).
In general switching a long range of versions is not recommended – always check if all is already encoded with UTF-8 and the old install is not overloaded with old plugin’s/ theme’s options and database tables. When sites are not so big a manuel transition is mostly the best option to have a clean and performant install. Just suggestion for users out there planning on such things.
And, in the early days of my WP experience – since summer of 2006 – I’ve made a lot of mistakes. But I’ve heard here and there that a lot of WP webmasters had these experiences back then… So it’s just learning by doing (and crashing) and to evolve more and more. In these days, work without test installs, playgrounds and local stuff is the absolut regular day-to-day work basis. I just recommend such setups to everyone also the “smallest blogger”.
-Dave š
Good thing about WordPress is that there is a huge community and a lot of people had similar problems especially during transitions between versions, and you can always find some useful advises on how to handle most of the problems. This does help to speed up the transition, even if you do it manually. And doing test transitions on local servers or some sandbox server is always a good idea to try and minimize the downtime of the live website.
I couldn’t disagree more with your comment about 2 major releases a year being to much. If anything I wish there were monthly major releases, but I realize that the core dev team is already working at full steam.
I think the reason that 3x had so many downloads is because of the introduction of post-types, menus, etc and the huge press that hit the internet about this new feature set. Basically unless you live in a cave there was no way of missing the news about the huge updates in 3x.
Upgrades after 3x have been great but really did not garner the news that the 3x did. Hence there was no ‘exciting’ reason for people to update.
Having only one major release a year would make it even worse. People simply would not be thinking about updating because they are not used to it. Where if there were major updates every month or so, then you end up in the habit of updating your site.
*I may be rather unusual in that I update all my sites to the nightly build as I keep up with changes on a daily basis
Thanks for the comment Shawn. 3.0 was a big deal update and was advertised as such. And next two failed because they lacked the impact. But, if these 2 were 1 version, adoption rate would be much higher. If things continue like this we will have a great fragmentation of versions by the end of 2012 with five 3.x versions and still many pre 3.0 versions.
“How to solve all these problems? How to improve the number of websites willing to upgrade and how to have a more secure websites running on WordPress. ”
Stop putting backwards compatibility in plugins and themes. š If people are forced to upgrade to use new features in themes and plugins, they WILL take that jump.
Thanks for joining the discussion.
Well, that is easy to say :), but much harder to do. I can see free plugins doing that, but with commercial plugins it is not that simple. At least WP 3.0 should be supported nowadays, judging from usage statistics. But, with 50% of all websites using it, it is not easy to simply drop support for it. On the other hand, commercial developers have better way of tracking WP usage, so some compromise can be found. Still, we have too much different WP versions in use, adding 2 more each year. That is the problem.
I’m with Andrea on this one. Backward compatibility is pointless and a waste of your time.
If your users really want to use your plugin/theme, then find out that they must upgrade, they will. This is a major incentive to force users to upgrade, and it eases your development burden in both maintenance and support.
Users should upgrade. We all know this. And they will if given the right incentives to do so. Having plugins and themes that they want to use but need to upgrade for is a great incentive. Heck, I released a plugin today that is already 3.3 only. The next versions of all my major plugins will require 3.3, period. I don’t do backward compatibility anymore.
I’m with Otto on this one.
My next plugin release actually requires 3.3 as well. I’m hoping just the fact that the user will gain the wp_editor on the front end with my plugin will be incentive enough, but if not, I’ll add code into the plugin which forces use of 3.3 just to be safe.
I have a new plugin coming (beta right now) that will require 3.3 also, but that is due to some of the features. That is a good point for requiring update to new WordPress, an evidence that new WP will bring something new, something user will need. We developers easily forget that 99% of all WordPress users are not developers, and that going with new versions of WordPress is not as natural or easy for them. There were many good points raised about the price of such upgrade and other factors, so this is not simple issue between upgrading or not using a plugin anymore.
WordPress is developing quickly – long may it continue to do so, it’s great to see and be part of such an active open source project.
But the potential for breaking themes and plugins is definitely there and does present a problem; it would be an interesting move to see the introduction of a Long Term Support version that would be updated only in relation to certain bug fixes and security patches.
oh my god wordpress 3.3 is incredibly slow , does anyone experience that , better switch to old ones
Wow-great article and very informative! Iām not a developer but a user of WP. I currently have a highly customized WP site with 30 plugins installed. I recently had my developer do a complete scrub of the site to clean up a lot of the conflicts stemming from poorly designed plugins. I can attest to the importance of having the right plugins ā it definitely makes difference.