The Fall of WordPress
How Elementor finally killed WordPress for me, and many other web developers.
So, I’ve finally had enough of WordPress
I started working with WordPress probably 14 years ago. I worked for a company building websites for small to medium sized businesses, using the Genesis Theme Framework. I eventually left that company and started working for a marketing firm around the time Elementor came on the scene.
I used to consider Gravity Forms to be the best plugin ever made for WordPress. That is, until Elementor. It made building websites so fast and easy, and more importantly, easy to hand off to clients. That kind of work built enough trust that clients usually paid to have me take care of the site for them. Sure, there were other visual page builders out there, but most of them had horrible interfaces. Although anything was better than the Gutenberg block editor that comes with WordPress.
Incremental Updates
Over time, Elementor received many updates. Some were useful, while others were a little bit annoying. Like pushing flexbox on everyone when the rows and columns system they had worked just fine. Still, the simplicity of dragging and dropping elements onto the page and adjusting the styling made life simple. Sure, occasionally you needed some CSS tricks to get an element to behave the way you wanted. But the most important fact was it was faster than hand-coding the website.
Disaster
Then along came Elementor V4, which introduced Atomic Elements. These new elements created a whole new level of complexity.
I can see the appeal for a designer, who wants an extra level of control over how elements look. However, for a developer like myself, it slows down the development process dramatically. With Elementor V3 and under, I could produce an entire website in a couple of days. With Elementor V4, it now takes that long just to do the home page, and generate the reusable elements for the website.
This is madness. I can hand-code a website faster than this.
Performance
I’ve also noticed a significant performance hit. Pages load much slower. Speed scores are much worse. And gods help you if you have a particularly long page. The whole editor can freeze up.
Not to mention if you blank out a field, instead of reverting it, you can’t save the page and you get a rather unhelpful error message. You have to retrace your steps to find the problem.
The Debate
Combined with some new issues my latest company was facing with Cloudflare and with increasing security concerns over plugins in the AI era, a colleague suggested we flatten the WordPress pages into HTML. Which of course doesn’t really make sense because of plugins like Gravity Forms, and honestly defeats the purpose of using a CMS in the first place. Would these websites even still work?
Epiphany
Wait a second…
What exactly is our product?
We build websites for small to medium sized businesses. The content is mostly static anyway. We need pages, blog articles, and forms. None of this needs to be dynamic. It just needs to be consistent and easy to update.
We could use a static site generator (SSG)!
…but which one?
Astro
After a little research, Astro seemed to be the best fit. We’re using cPanel hosting, so trying to shoehorn something like Next.js into it would be a disaster. We need to maintain our hosting for analytics purposes anyway. And since there’s little dynamic content anyway, something lightweight like Astro fits the bill.
So, I created a reusable boilerplate around Astro, that would make it easy to translate either existing websites or Figma designs into a working website using agentic development tools (I have recently upgraded to Cursor from VS Code + Copilot).
I’d originally considered using Formspree for the contact forms. However, I found their pricing structure to be awful. I mean, why do they expect $200/mo just so you can use email templates? That’s absurd. So, I abandoned that approach in favor of a PHP-based API for emailing the form submissions and autoresponders.
The Experience
This new approach is an absolute dream for developers. You get all of the power of working with a framework, and what it spits out is tiny compared to a WordPress installation. Development is handled through Cursor, which even allows you to select parts of the website preview you want to make changes to. Import your Figma design or HTML exports from an existing website, let the agent handle the details. Guide the agent through optimizations and SEO/AEO. Deployment is handled through GitHub to your hosting.
The end result is a fully custom, highly optimized, very fast website that can be made to pass any test you throw at it.
Conclusion
While I consider myself more of a “Prompt Engineer” than a “Vibe Coder” (the difference being I know the languages I use), anyone should be able to build websites using this method. If you lack design skills, you can use Google Stitch or hire a web designer to do it in Figma for you. Then use your favorite agentic development tool (I really like Cursor) to build it. No more WordPress, Wix, Squarespace, etc. Real, high-performance websites for all.
Keep reading
eBay POCSAG HAMNET DAPNET Pager Battery
It took some tooth pulling, but I finally identified the battery used in the POCSAG pagers you can get on eBay from China. They don't come with the battery because of shipping hassles. It's an 802040 LiPo cell, 3.7v 600-700mAh. There's plenty of them on Amazon. Update: Wait til the pager comes in. T
FireFox Green Bars On Video
If you are getting green bars on video using FireFox this is the solution. I've found too many old sites with instruction on turning off hardware acceleration in Flash, but no one uses Flash any more. Duh. The reason you're getting green bars on your videos in FireFox is your video card is old. You'
PDS Tornado Watch: The Particularly Dangerous Situation That Wasn't
On April 26, 2016 the Storm Prediction Center issued a PDS Tornado Watch that included a large portion of the North Texas area. This is a special kind of watch issued only in a Particularly Dangerous Situation. That is to say when computer models are predicting the possibility of extraordinarily sev