Sluggish Drupal 8 Adoption Lags Even D6

Filed under:

We're just past the second anniversary of 8.0.0. To see how D8 is doing compared to prior releases, we put together the chart above, based on Drupal's usage stats page.

For versions 5.x, 6.x, and 7.x, each new release brought dramatically accelerated growth.

Comparatively, D8 has dropped off a cliff. Adoption is far below that of D7, and even behind D6.

Have you moved on from Drupal? Let us know in the comments! What are you using instead?

Edit: The chart data has been updated as of December 2018.

Date posted: December 11, 2017


Drupal 8 is dead. Long live Drupal 7

The time it takes to migrate existing sites properly to Drupal 8 is sometimes a yearlong project, or longer in the case of large D7 installs. Once these projects start finishing I think adoption numbers will change. It really hasn't been that long since some of these sites have started becoming possible due to the slower release of contrib stuff. I have already completed one yearlong project and am working on another. I'd say the upside is way, way more than keeping things on the aging D7 platform due to a lot of factors.

Some people haven't even started that yet, and apparently a lot of people don't plan to. Which should be fine because D7 is still out there and supported.

Btw composer is a poor excuse for people not picking up D8. If you're in this space you should be learning new things. Drupal stopped trying to be Wordpress at least 5 years ago.

I absolutely loved Drupal 6 and 7 and the power it gave site builders to do almost anything they could dream of doing. I could easily put together a small Drupal 7 site for a client pretty quickly and at reasonable cost.

I keep waiting for Drupal 8 to have modules like Rules and Feeds. While they aren't in beta yet I don't feel that I can build client based sites using them. I also agree that using Composer is a bit of a pain.

I do a lot of work for small not for profit organisations. They wouldn't even be able to conceive of a website costing over five figures For simple sites that don't need complex data models I just just Squarespace or use WordPress these days. I never thought I would see the day when I choose WordPress over Drupal!

At the end of the day most of my clients don't really care what CMS they use. They just want something flexible, robust and easy to use and update.

Yes, I have moved on for small projects, and the answer is really simple: https://www.drupal.org/docs/8/update/update-core-option-1

Option 2 (the one labeled as mid-level) has a bold warning that says:
WARNING: if you have installed any contributed modules with 3rd party dependencies using Composer, you need to use the other update options as these instructions will overwrite your vendor/ directory.

Which basically means you have to go to options 3 or 4, that involve drush and composer respectively.

In WordPress, my clients are a click away of having both core and their theme/plugins up to date. If there's a problem (hasn't happened for more than a year to any of them), they can go back to a backup that they also generated with one click. Also no need for update.php.

For big projects with customers that can afford to pay the maintenance bill I ask for, I go with D8.

i'm i site builder and i love drupal so much, its becoming my passion. but drupal 8 make it hard for me to migrate. many of the modules i needed are not ready yet or just recently available.

rules - this is the big one. site builders dont know how to code, so depending on Rules module is sooo critical

drupal commerce - just recently stable but without the integration of Rules. this is bad. now i dont know how to fully customize commerce without Rules.

migrate path - as we all know it

i know you all want to go for 'ambitious digital experience', i'm for it, but leaving site builders behind is a really bad idea.

the move to Symphony, Twig and Composer is the right direction. I have found D8 development fast and intuitive. TWIG enables people that didn't speak PHP before to write custom layouts and modules. Expecting even site builders to learn a basic command line infrastructure like Composer doesn't seem that big of an ask - - and there is a lot of whining going on. The nerve of the Drupal community to evolve and expect people to learn new skills?!

the adoption rate speaks for itself. If what you say were true, then it would have been better than previous versions, not horribly behind.

My theory is that the sales numbers from the fewer sites in D8 today just might be somewhat closer to what the D7 sales numbers were at its 2yr point. [total, wild, uneducated, unresearched guess]
To say it another way....
Drupal 8 sites are more "Ambitious" than previous versions and that can often mean more expen$ive. In my personal experience, the D8 sites I have worked on are magnitudes of order more expen$ive.
So... It is my suspicion that even if the larger Drupal shops are building less sites [maybe some are?] the increased price-tags could quite possible make up for it.
... Individual Results May Vary ...

They all most have to be much bigger, and more expensive, so they can have the margins that are required to build with such a complex tool. With Drupal 6 & 7 I could bust out a small site for a couple grand over a weekend, and be profitable. That isn't the case with Drupal 8.

Drupal 6 & 7 were relatively easy to install and deploy without having any development knowledge ... for me learning Drupal 8 was a hard but rewarding slog ... there are still lots of issues the current update from 8.3.7 to 8.4 was incrediblely frustrating esp the issues with the media update and drush 9 ... much of my experience with drupal 8 lead me to believe that a lot of the releases are just not ready for production ... put I persist and being truthful I have to say that Drupal 8 is magnificent out of the box and if your installing from 8.4 up it is a amazing CMS its out of the box translation and media system is unsurpassed in any other CMS ... the plans for front end editing and layout discovery are very exciting.

I've brought this up several times over the past four years, but it would have been nice if there were two versions of Drupal 8: an Enterprise and a Lite version. The Enterprise version would have obviously been aimed at the big dogs Acquia was trying to attract and the Lite version could have maintained those who were successful in D7 but not ready to move to a full-blown and not-quite-ready-for-primetime D8. Yes, an Enterprise and a Lite version of D8 might be difficult to maintain, but how much difference could that have made in the adoption rate? Further, by offering a Lite version, perhaps Drupal would not have lost part of its user base who felt marginalized.

Drupal has always been dinged for being too hard to learn compared to other CMS out there. We argued that with that came power, flexibility and quality and we were somewhat right. With D8 we have doubled down on that... or octupled down on it more like. It is now hard to learn for experienced Drupal people. In trainings I saw Symfony people picking up Drupal 8 faster than Drupal people, though they complain that we don't have all the power of Symfony and error messages are a confusing mess -- more so for us old Drupal hands than them though. Even so, as before, if you endure the learning curve then you will love much of it. All the D8 experts I know would never go back to D7.

But it's still a big obstacle and there are others. You pretty much have to buy an expensive proprietary editor with an excessive price and annual fee just to cope with the basics. Modules are lagging and some major modules have disappeared or are dormant, and just a few of those have been replaced by a new module. Minor core upgrades are hairy and always have bugs, usually in contrib modules, and both these modules and many patches take weeks or months to catch up. The codebase is either fragile or contrib makes it seem that way and odd crashes/errors are common. Most new modules I install add new incoherent errors to the logs demonstrating that even the module developers are struggling.

Our site has been live 15 months and in development from 21 months ago. We needed about 17 patches to get our bleeding edge site to work decently, but all this time later most of them are still needed, including all three core patches.

Movement is slow all over. This is also reflected in how attendance at DrupalCon is down and the European DrupalCon has been killed, which in turn hit the DA hard.

User group meetings have shrunk and the community part of the community seems to be considerably smaller too. D6 and D7 left freelance designers, hobbyists and even many freelancers behind. D8 seems to be leaving small organizations and even small dev teams behind. Many of them were making big contributions. With version 8 Drupal has become a tool for teams, vendors and bigger businesses who don't care about community, and all of that other community richness and diversity that made Drupal strong in the past is now diminished or gone.

Conversations I've had with DA people or leading Drupal people are met with denial, indignation, even surprise and feelings of betrayal, but I'm asking because I care.The next year will be key, I think, but the signs are iffy.

Not a big surprise that Drupal 8 adoption lags behind other versions. Compared to Wordpress, it's not hard to spot the missing features users want that are now missing from Drupal - but used to be part of d7.

Feels like d8 is an attempt to solve problems that only exist at the enterprise scale, while making things hard for the small non-profit, e-commerce, and social sites that made it popular in the first place. The problem with that, of course, is other, more specialized platforms like Adobe Experience Manager and Oracle are a couple generations ahead on many of the things Drupal 8 aspires to do. Each of these platforms has very customer-focused product teams improving defects that used to cause organizations to consider their options. While d8 might be competing in that market, it would need to move several steps ahead on web services integration, multisite (read: organic groups), rules, and integration with metrics platforms to thrive. Just the fact you get better metrics from an out-of-the-box AEM install is a huge deal for media companies, the cost of getting the same details from multiple analytics platforms is often equal to the cost just to implement d8 over the course of a year.

Abandoning d8 and merging with backdrop to create a d9 release would not be a bad idea to consider. The enterprise market can be taken on a lot more effectively with a healthy, robust group of contributors, which doesn't seem to exist now. A merge won't happen, of course, there are a few too many egos involved and prominent developers with blinders on. But it's still worth discussing.

...I would not say "Abandon D8 and merge with Backdrop CMS to create D9". That's a bit to hard. Drupal should continue along the path it has created. I believe that there is a market out there for D8, as there is for D7/Backdrop. Something that I would love to get to the discussion table is something along these lines:

If you do not like D7 any more, then fine. Just give it (the part of the community that is not ready to move to D8 yet, and that is seriously thinking to move away to Wordpress and other solutions) to us (Backdrop CMS). Lets create something similar to Fedora and Red Hat, where Backdrop is the "community" solution, and Drupal 8 and beyond is the "enterprise" solution. I think that this is a very decent compromise.

A D7 to Backdrop "upgrade" is already in place, with the simplest D7 contrib modules already working out of the box with Backdrop, 350-some of the most popular already ported to Backdrop contrib (https://backdropcms.org/modules) or merged into its core. So D7 fans can move to Backdrop, where they will find themselves in a very familiar/comfortable place, giving them more time to prepare for the "big" move to D8 (or D9). A D7 to D8 "migration" is already in the works, and because Backdrop is so similar to D7, it can be adapted to also support Backdrop to D8 (or D9). Win-win situation.

Although as I said I do not agree with the "Abandon D8 and merge with Backdrop CMS to create D9" bit ...I fully agree with the "there are a few too many egos involved" bit. But I am hopeful :)

What a comment thread! I find myself disagreeing with many of the posts here. We've launched over a dozen Drupal 8 sites in the past 18 months or so, and we find it really refreshing to work with. We still have another ~20 D7 sites we maintain, and we find ourselves cringing a bit when going back to work on them...

Usability is noticeably better on D8, and our clients appreciate it. Upgrading and migrating content to D8 is far easier than migrating to D7 ever was. Configuration Management has a learning curve and some big gotchas, but is a huge improvement over our previous features-driven deployment. Commerce (now that it's finally out) is much quicker and easier to get going. There are definitely some rough spots that involve going to code, and many places where code is the solution instead of a module -- but coding with Drupal Console, plugins, and a cleanly defined API is a dream, compared to D7 and earlier.

I characterize all of this as growing pains. Drupal 8 has reached a level of maturity and robustness that is far better than earlier versions -- however changes in the broader marketplace and repeated traumas of multiple major version updates that turn into ever bigger projects have driven away much of the earlier Drupal market. Now Drupal is better than ever, but has a tarnished reputation.

As we see more and more successful, sophisticated projects launch, I'm confident Drupal 8 will grow -- perhaps more slowly than previously -- but its future is very bright.

As the number of sites our organization delivered grew, we found that CMS solutions didn’t scale with our operations resources. We were spending an absurb amount of labor on system administration, content manager help desk support, and more expensive hosting (compared to pure static sites.) It started with a hunch, but once we started transitioning D7 sites to static (96%) and SSGs (4%), we recovered a very large amount of reoccurring labor costs. The value we get from D7 was demonstrable in literally 1.6% of the sites we were delivering—in those few cases we kept it in place and began developing a homegrown replacement that has more in common with headless CMS systems. Everything else is now smaller scale, static sites (simple PHP + JSON, or less often node SSGs). The new approach is as effective in the market, significantly more performant in terms of page speed, at lower hosting costs, and completely eliminated the labor resource burden of training non-dev content managers and updating CMS software. Or InfoSec folks love it. Or DevOps / SysAdmin love it. Our form content managers love it—they get to spend their time on more valuable work. And we didn’t have to scale up our dev staff—we instead found ways to work smarter, automate, and optimize change management processes.

Not surprised with these posts.

Drupal was and is still a tank. As people are moving to serverless microservices and more agile frameworks like Rails/Django/Laravel, Drupal is left behind. More sites are moving back to static HTML + JS for simplicity and speed. I'm a professional developer and don't see many good reasons for switching to D8. I've even gone through the D8 boot camp in DCon LA and, while Drupal has been improved in many ways, very much of it still feels like same old.

As a developer, I want to code, not click around in the UI. I was excited to see if Drupal could deliver a better experience for site builders, but while many improvements were made, I found myself grappling with outdated patterns between a mix of old and new ways to implement things. I confess I haven't gotten as far as I should have, but the introductory experience to Drupal 8 did not yield confidence.

I'm all for frameworks and such, but when developers are faced with an option to re-learn something from scratch, they evaluate that against ALL other options. This was their opportunity to jump ship to sexier, lighter frameworks, and many did. Fan or not, there is also Wordpress gobbling up large shares of the open-source CMS market.

The idea of a CMF is no longer held in high regard and is very project dependent. The bar for a Drupal 8 implementation has increased, so smaller sites opted to by-pass it all together. Aquia moved into the 'enterprise' market and got exactly what they wanted.

Honestly, if someone asks for a marketing site, a REST API, or even a CMS, Drupal isn't the first thing that comes to mind. It has its use cases but is mostly now reserved for clients with large pockets and an enterprise requirement. I've been using Drupal since 6, so I'm no stranger to the platform. I'm super happy it has grown, I just think it hasn't grown in the right direction.

Too little too late. Don't think this toothpaste can be squeezed back into the tube. Folks have already found their alternatives. Why risk going back, when/if they ever address these issues, when the hubris that created them in the first place is still present. Even if D8 is eventually made usable/accessible for normal people, the risk that this type of enterprise/developer centric decision making will ruin another release normal users will always be there.

For me, I'm absolutely loving laravel + vue.js development.

Fool me once, shame on you. Fool me twice, shame on me.

Drupal 8+ is not the future of Drupal. The future of Drupal is Drupal 7 (Backdrop) or Wordpress!

If you're thinking of learning Drupal 8. I would consider learning Wordpress instead.

Nothing has really changed for Drupal 8. I found this thread after just breaking my D8 development site with composer update drupal/core something something code which I always copy from drupal.org as I don't remember the exact command .

After first glaze to contrib modules in drupal.org the only reason for me to even try to learn D8 is the upcoming end of life for D7 that got me on my toes. I thought I could write the 'glue' that I need as custom modules but as banging my head to the wall while learning the mess I begin to think that if I have to code more modules what purpose is drupal for?

So I move to Laravel, study that well and just write what I need? That's what others who didn't move to wordpress did right?
Love for drupal has definitely changed to some other feelings.

I'm a sitebuilder and well capable of reading code and writing small modules. D8 under the hood just feels like eating surprise soup blindfolded, you can only guess what you're getting. It used to be Drupal, now it's some weird symbiosis with drupal, composer, symfony, git, drupal-console and I haven't even dived in that deep yet. Community is dead. Year old documentation does not apply and/or is confusing.

Something that was as easy as hitting a child in D7 is incredibly complicated in D8. I'm starting to look away.

I have only just made the move from Drupal 7 to Drupal 8 and it has been scary as hell. Most of the modules are not ready for prime time. Conditional field is a complete fail just doesn't work with any fields :). Form displays is also very buggy and the list just goes on and on. I now spend more time going through modules trying to find ones that actually work which has increased development time to unacceptable levels. I can see an atomic bomb of potential with Drupal 8 very powerful but it is not ready for prime time. What really worries me is that Drupal 8 has the potential to be an incredible CMS but people are already talking about Drupal 9 when Drupal 8 is plagued by so many issues at the moment.
Composer is fantastic once you get through the very vague and confusing documentation and get it up and running. I also have no Idea why we have an uninstall for modules and an available updates as everything now needs to be done in composer. I hope things improve quickly for Drupal 8 otherwise I can see myself moving to another platform which I really don't want to do.

Add new comment

Restricted HTML

  • Allowed HTML tags: <a href hreflang> <em> <strong> <cite> <blockquote cite> <code> <ul type> <ol start type> <li> <dl> <dt> <dd> <h2 id> <h3 id> <h4 id> <h5 id> <h6 id>
  • You can enable syntax highlighting of source code with the following tags: <code>, <blockcode>, <cpp>, <java>, <php>. The supported tag styles are: <foo>, [foo].
  • Web page addresses and email addresses turn into links automatically.
  • Lines and paragraphs break automatically.

Metal Toad is an Advanced AWS Consulting Partner. Learn more about our AWS Managed Services

Schedule a Free Consultation

Speak with our team to understand how Metal Toad can help you drive innovation, growth, and success.