Core Web Vitals: WordPress and Astro Versus Everyone Else

We often treat performance as a binary. Either a site is fast or it is slow. But when you look at the data from the HTTP Archive, you realize that the relationship between how a site is built and how it actually feels to a user is far more complex. The latest Core Web Vitals Technology Report reveals a surprising truth, which is that the numbers we see in a lab environment do not always predict the actual experience of a human being browsing a page.

The Gap Between Lab Scores and User Experience

The core of the recent report is a comparison of seven different content management platforms. The most interesting takeaway is that page weight and the scores we get from PageSpeed Insights Lighthouse audits are not always reliable predictors of Core Web Vitals performance. This suggests that we might be obsessing over the wrong metrics if our only goal is a better user experience.

I find this distinction critical because it changes how we approach optimization. If a low Lighthouse score does not automatically mean a poor user experience, then the way we prioritize our technical debt needs to shift. We have to stop treating the audit as the final word and start looking at how the platform handles the delivery of content in the real world.

Why These Metrics Actually Matter for Business

Google created Core Web Vitals to quantify things that were previously subjective, such as how stable a page feels while it loads and how responsive it is when a user clicks a button. While these metrics are only a minor factor in how Google ranks a site, the business implications are much larger.

When a page is smooth and fast, users are more likely to convert and ad performance generally improves. On the other hand, low scores usually mean the user is encountering friction. This friction leads to frustration, which leads to higher abandonment rates. From a practical standpoint, the tradeoff here is between adding "flashy" features and maintaining a frictionless path to purchase or sign up. The decision for a site owner is whether the perceived value of a complex feature outweighs the potential loss in conversion caused by a slower experience.

The Misconception of Page Weight

There is a long held belief in the industry that page weight is the primary driver of Core Web Vitals scores. The logic is simple, which is that smaller pages load faster. However, the data suggests this relationship is not as straightforward as we think. While weight matters, it is not always the dominant factor.

This is a vital realization for anyone managing a site. If you spend weeks shaving kilobytes off your images but ignore how your JavaScript executes, you might see your page weight drop without seeing any improvement in your actual CWV scores. The tradeoff is between "weight reduction" and "execution optimization." You have to decide if you are optimizing for a number in a report or for the actual behavior of the browser.

Understanding the Data Sources

To understand these findings, we have to look at where the data comes from. The report blends two different sources, the Chrome UX Report (CrUX) and the HTTP Archive project. This combination is important because it provides a global view across various devices and connection speeds.

CrUX provides field data, which is anonymized real world performance from actual Chrome users. The HTTP Archive provides lab data, which is collected by crawling and testing sites in a controlled environment. By comparing the two, we can see the difference between how a site performs in a vacuum and how it performs in the wild. For a developer, the decision here is to prioritize field data over lab data, as the field data reflects the actual human experience.

Duda and the Efficiency Lead

Duda came out on top in this comparison, with about 85 percent of its sites achieving a good Core Web Vitals score. This success seems linked to its efficiency, as it also had one of the lowest median page weights at roughly 1.78 MB.

In this case, the correlation between light weight and high performance is clear. Duda has managed to pair a lean payload with a highly optimized delivery system. The lesson here is that when a platform controls both the build and the hosting environment, it can eliminate a lot of the bloat that usually plagues DIY setups.

Wix and the Middle Ground

Wix followed closely in second place, with roughly 80 percent of sites scoring well on Core Web Vitals. Its median page weight was about 2.55 MB, which is heavier than Duda but still lighter than many of the platforms that struggled.

This reinforces the general trend that lighter pages tend to cluster at the top of the rankings. However, it also shows that you can have a slightly heavier page and still maintain a great user experience, provided the platform handles the loading process efficiently.

The Shopify Anomaly

Shopify ranks third, with about 79 percent of sites receiving a good CWV score. This is the most surprising part of the report. E-commerce sites are usually bogged down by tracking scripts, third party apps, and complex product galleries. In fact, Shopify sites had some of the worst page weight and Lighthouse audit scores in the entire group.

The fact that Shopify performs so well despite these hurdles proves that page weight is not the only path to success. Shopify likely achieves this through stable rendering, layouts that prevent layout shift, and a very aggressive CDN and hosting infrastructure. The tradeoff here is "payload versus delivery." Shopify accepts a heavier payload but optimizes the delivery so effectively that the user does not feel the weight.

For a business owner, this means you do not necessarily need to strip every single feature from your store to get a good score. Instead, you should focus on the stability of the rendering and the quality of the hosting environment.

The Astro Approach to Performance

Astro sits in fourth place, with 67 percent of its sites scoring well. Interestingly, Astro had the lightest median page weight of all the platforms tested. This supports the intuition that being lightweight helps, but it also raises a question about the types of sites being built with Astro.

Astro is often used for simpler, content driven sites like blogs, which naturally lack the heavy functionality of a full e-commerce store or a complex CMS. The performance here is a reflection of both the tool and the use case. The decision for a user is whether they need the heavy lifting of a traditional CMS or if they can move to a static first approach like Astro to gain an immediate performance advantage.

Drupal and the Value of Stability

Drupal ranked fifth, with about 64 percent of sites achieving good CWV scores. Its median page weight was around 2.28 MB, placing it right in the middle of the pack. One notable detail is the stability of Drupal, as its scores remained exactly at 64 percent from January to April 2026.

While stability is a positive trait, it also suggests a lack of aggressive improvement. In a competitive landscape, staying still is often the same as falling behind. The tradeoff for Drupal users is reliability versus innovation.

The Execution Gap in Joomla

Joomla ranked sixth, with roughly 58 percent of sites scoring well. What is strange here is that Joomla's median page weight of 2.53 MB is actually better than some platforms that ranked higher. Despite the lower weight, Joomla had the lowest Lighthouse scores of all the platforms.

This indicates that the problem is not the size of the files, but how they are executed. Issues like render blocking resources, poor image handling, and low quality extensions are likely dragging down the real world experience. This is a perfect example of why weight is a misleading metric. You can have a small page that feels slow because the browser is struggling to process the code.

The Persistent Struggle of WordPress

WordPress finished last, with only about 49 percent of sites receiving a good Core Web Vitals score. It also struggled with page weight, averaging around 2.63 MB, and ranked second to last in Lighthouse audits.

The gap between WordPress and platforms like Astro or Duda is massive. When you compare a 2.63 MB median weight to Astro's 1.65 MB, you are looking at two different categories of performance. The problem with WordPress is often its flexibility. Because it allows users to stack endless plugins and themes, the resulting bloat is often insurmountable for the average user.

The decision for a WordPress user is whether the ecosystem of plugins is worth the performance tax. To move the needle on CWV, WordPress users cannot just rely on a caching plugin, they have to fundamentally rethink how many third party resources they are loading.

Comparing Median Weight to Real Performance

When we look at the median page weight versus the actual CWV performance, the lack of a perfect correlation is the most important lesson. We see platforms with low weight that struggle and platforms with high weight that excel.

This tells us that the "how" is more important than the "how much." How the browser parses the HTML, how the CSS is delivered, and how the server responds are the real drivers of the user experience. If you only focus on reducing the size of your images, you are missing the bigger picture of browser execution.

The Final Verdict on Platform Choice

Choosing a platform is always a tradeoff between ease of use, flexibility, and performance. If raw speed and Core Web Vitals are the priority, the data points toward Astro or Duda. If you need a strong e-commerce engine and are willing to accept a heavier page in exchange for a highly optimized delivery network, Shopify is a strong choice.

Introduction

The key issue here is HTTP Archive's latest Core Web Vitals Technology Report ranks seven content management platforms and offers the surprising insight that page weight and PageSpeed Insights Lighthouse scores do not always predict Core Web Vitals performance. My read is to treat it as a decision point: what signal needs to become clearer, what part of the system is currently weak, and what evidence would show that the work is improving visibility rather than only adding activity.

That is the difference between reacting to a trend and building a useful search system. Connect this point back to the page template, internal linking, entity signals, content depth, crawl accessibility, and the way the brand is represented across the wider web before deciding what to change first.

Practical next steps

The useful part is not only the idea itself, but the operating habit behind it. Use it as a checklist for decisions: what deserves attention now, what should be monitored, what needs a stronger evidence base, and what can wait until the system has more scale.