A week or so ago, Joost de Valk (Emilia Capital, formerly Yoast SEO) wrote “WordPress, and what should be on its roadmap“, and before I continue, you should probably read that first. He covers a lot of well-understood ground: WordPress has major UX problems, WordPress has technical debt and outdated code, WordPress needs to focus on enabling modern integrations more easily in core, and so on.
I say these are “well-understood” ideas, but what I should probably say is “well-understood by anyone outside of whoever is taking leadership responsibility for Gutenberg” (otherwise articles like this wouldn’t need to be written). It’s a very good article, even as it’s one of many articles (and videos and WordCamp presentations and podcasts and livestreams) outlining these similar concerns that WordPress has lost its direction the last few years.
I’m feeling a bit more optimistic after State of the Word and some recent work inside of Gutenberg. For every UX head-scratcher (and believe me there are always few in each release), there’s also a lot of focus on some issues I think are important for my narrow view: the agency developer deploying multiple sites per year. Version control is getting easier, responsive design is no longer a hard “no”, underlying APIs are getting more advanced, the separation of “content” and “design” modes in the editor will be fantastic, and partially-synced patterns are getting closer to giving us a “component” experience in core.
But some of those features are a far cry from being production-ready, and there are tools out there right now that are actively solving these problems inside WordPress: Elementor, Kadence, GenerateBlocks. Bricks, and so on.
As Dries Buytaert (Drupal) mentions in this public conversation with WordPress founders Matt Mullenweg and Mike Little, one of the main value propositions in open source is different teams providing different solutions to the same problem:
But the beauty about Open Source is you have all of these people that can contribute their own version and even fork existing versions to make it better.
And eventually just like evolution in their, you know, in their Darwinian evolutions, sort of the best solutions bubble to the top and bad solutions tend to disappear.
https://youtu.be/QYhIItlPPOs?si=1y4qOMVOZwXSaVpz&t=1207
Even I forget that sometimes. Why do we need two dozen form plugins, why can’t we have just one good one? Well, this is why. Because software continually changes and evolves and what is good in one particular time or one particular context will always change. If you don’t like where Wix goes, tough. With WordPress you always have a choice.
(If there’s one variable beyond just “good code good” that open source leaders tend to ignore, it’s the importance of marketing.)
What is content management
What is a WordPress website really? Yes, there’s this massive backend monster of a content management system, but let’s just talk about the front end of a typical website for a minute. The front end of your WordPress site is actually pretty simple, pretty classic. It’s mostly just HTML and CSS, generated from PHP but probably cached as static content somewhere for most visitors. This hasn’t really changed in twenty years. It’s the simplest, most basic of websites. There’s no frontend framework burning through JavaScript, no continual fetching of more and more and more resources.
At the end of the day, WordPress is still the quickest and easiest way for a non-technical person to regularly publish words and images as simple HTML pages on the web. Browse a WordPress website from ten years ago (and I’m sure most of the ones you see are that old), and it just works.
Sure other CMS tools like Sanity and Contentful may be more powerful, but the websites you build for them are not nearly as simple to create or maintain. Matt Medeiros spun up this new landing page for MasterWP in less than a few hours using the WordPress site editor. And when he’s ready, he’ll be able to pick a much better color palette for it without the help of a developer or any complicated deployment process. He can continue editing his site content without an annual license fee. That’s the democratizing power of WordPress.
But as Dries mentioned – open source means we get different approaches to the same problems on the same platform. As happy as many of us are with Gutenberg (a smaller number to be sure), there’s a much larger group that’s been using a tool like Elementor just as happily. This is a testament to the power of WordPress, as I don’t see Elementor, or Awesome Motive, or even WP Engine truly forking WordPress to create their own CMS any time soon. They’re doing great standing on the shoulders of giants.
All of this to say that I don’t disagree with anything in Joost’s article. Maybe I’d prioritize a few things differently, have other focus areas he didn’t mention, but overall it feels right. The reaction to his article, though, I’m not sure I agree with.
Is the block editor a failure?
I want to focus on one image from his post that’s been widely shared, because when I look at it, I do see something different than most people are seeing.
This is an image showing the total market share of Elementor when compared to the block editor and specifically the site editor (full site editing).
So what does this image show? Well first of all, it shows that full site editing adoption is extremely low, less than 1%. If anyone is surprised by that, they haven’t been paying attention. You could see that as a criticism of the feature itself (and I wouldn’t argue with you), but you could also see it as a potential new market opportunity, as some people have. Depends on how much faith you have in the core team.
The second insight is that even though almost 9% of all tracked websites use Elementor, around 5% of sites use Gutenberg to some extent, though it’s unclear if this accounts for crossover (Elementor for theme/pages, blocks for posts). It looks like a huge difference in this chart, but if you zoom out and look at the original source, we can see that the block editor has already outpaced every other page builder. Zoom in on this year and only the block editor, site editor, and Elementor are showing consistent visible growth. Although YouTube would tell you otherwise, novelty builders like Bricks don’t even crack the top ten yet, though I’m sure they’re making a great niche product and running a profitable business.
This growth is happening relatively slowly, and while I personally would love to see the block editor’s pace of development match these other page builders, you do get what you pay for.
Where the block editor shines.
Even as we’re seeing the page builder market get crowded with more competition, there’s another alarming trend that’s not on this chart. The middle-ground “$20k marketing website” industry has been falling off a cliff the last two years, just ask any web agency in between their numerous rounds of layoffs. Whether it’s due to new technologies, the complete market saturation of web development services, or the decreasing costs of technology/labor, the web development industry is shifting.
Websites are becoming increasingly cheap or increasingly expensive- and where you as a developer will land on that scale depends on a number of things, not least your skills as an individual. Web development has always been a shifting landscape, with continued education simply the price of admission.
Meanwhile the block editor is becoming the builder of choice at the higher-end of the WordPress market, powering enterprise and attracting back all the news/media that abandoned WordPress for proprietary CMSs. It’s a much smaller market, for sure, and it definitely isn’t good news for all the smaller agencies and product companies, but let’s be clear: the WordPress ecosystem’s golden age is passed. The introduction of the mobile internet was a boon for web developers (and app developers and search/ad companies so on) but that boon is over and the market itself is more mature, with flatter lines.
Who is this new market?
There are a few other things that graph doesn’t show us – it doesn’t show us market share relative to new websites. We all know that’s important: we hope to see more WordPress websites launching each year. It’s really hard to derive “trends” from data where you might be comparing a site built last week to a site built last decade.
It also doesn’t show us the economic value of those individual sites, by which I mean what types of hosting they purchase, what types of software licenses they’re paying, how much active development they fund, the initial cost for their website, and so on.
Another thing we don’t see specifically measured is existing WordPress sites without a page builder. It’s pretty hard to determine from this data, since it uses “all websites” as it’s base, but I am definitely curious where all the rest of the WordPress websites sit. Many of the agency developers I know never used page builders. Like, at all. And many weren’t using the block editor until pretty recently.
This is from an email I got just this week from a developer who had been using the Understrap framework to build their WordPress sites, which typically meant pages were built with good old PHP and Advanced Custom Fields:
Just came across your site for the first time, though I’ve known (of) you for many years via UnderStrap, my go-to for parent/child theme dev. Just finished a site build and have a little down time and started looking into where to go next WP theme dev. This site was possibly my last “old style” WP build as I’m hoping to use this time to move to a more current block-based theme (probably using ACF Blocks as I have decades of PHP in me).
The open source starter theme frameworks like Underscores, UnderStrap, Roots/Sage, and the newly popular TailPress- not to mention the countless starter themes inside of larger agencies like 10up and WDS- were almost always built without any page builder in mind. And many of these themes are embracing native blocks, as Ben Word from Roots eloquently wrote last year, or ACF Blocks.
I don’t know how to estimate the size of this cohort when compared to the page builders. I have to imagine it’s not as large, but again, this is a completely different set of users. These are typically more advanced or hands-on developers using their own bespoke tools and frameworks to launch sites that require custom development, custom integrations, version control, redundancy, and so on. They don’t have clients who might lose business because they auto-updated and deployed a bug in their page builder without testing.
My point here is that there are different types of “builders” in WordPress – there are the extenders leveraging page builders and there is the much smaller group of coders who are increasingly adopting the block editor. That’s a good thing. They serve different aspects of the industry, work at different price points, and provide different value propositions to their clients, such as speed of development vs resilience of deployment.
A page builder is designed for an easy site-builder experience, Gutenberg is designed for a robust content-editing experience. Different markets, different solutions. I think we’re going to see that WordPress – once the darling of the DIY market – is moving up to focus on increasingly higher-value websites. And that bug I linked to above is just one more example of why having a modern, consistent approach to the frontend markup of all WordPress sites actually is an important goal.
Why is core still so important?
I’m not saying the block editor- and especially the site editor- don’t have a lot of work to do to gain more adoption. The site editor especially needs some major work. Gutenberg has some advantages (being shipped in core, being free) and is facing some major disadvantages (massive negative press from the initial launch, no paid marketing efforts like Elementor), but ultimately what really wins is the best software and the best marketing.
There’s a lot of fuss from the page builder community about how they want Gutenberg to be better, but most of them aren’t using it, right? There’s a common refrain of “WordPress needs to focus on the core CMS features” by which they usually mean existing content modeling solutions like ACF, but I’m just curious why they don’t push that same criticism onto the site builder they’re actually paying for? If it’s still such a critical gap in WordPress, where is the rush to build and monetize it? Why are we asking so much of the one part of our stack we’re not even paying for?
Part of the answer might be something Felix Artnz mentioned in our recent interview. I asked him about adding new features to WordPress as canonical plugins, and he responded:
Realistically, I’ve noticed that it’s hard to get adoption for this kind of plugin because it’s still not part of Core. I don’t know what exactly…but I think it’s important. Developers have to be confident that this is a thing that is going to be around. […] I think only once it hit Core, it really reaches a critical mass.
https://webmasters.fm/google-performance-ai-services-for-wordpress-ft-felix-arntz/
It’s the same argument for keeping Gutenberg in core. In the long run, being in core will mean a level of adoption that each individual builder can not typically enforce. It’s counter-intuitive – shouldn’t the tool I pay for be more reliable than the open source one? But there you have it.
What’s next for WordPress
I want to clarify that Joost’s article is not wrong. I agree with his specific points, and he knows the history and the market better then most people. But the general internet chorus seems to be falling into the same trap that WordPress itself often falls into: using “market share” as a proxy for success or health.
If you’re a developer, it’s not enough to just focus on finding the next popular tool to speed up your workflow, it’s also about finding the opportunities to level up what you can personally offer, because selling cheaper, faster websites is only going to be a race to the bottom.
So if I had my wish for 2025, my WordPress roadmap post, here’s what I’d like from the core WordPress team: First, say what you’re trying to accomplish, then do it well. That’s it. The rest will sort itself out.
Leave a Reply