PageSpeed MattersPageSpeed Matters
    See Pricing

    Data Study · Website Migration

    Page Builder Rebuilds: 25.3% of WordPress Sites With Good LCP Lost It

    1,879 top 1M WordPress sites added a visual page builder. Of the 1,715 with real-user LCP data before and after, the share with a good LCP fell from 55.0% to 48.8% within three months, while WordPress sites that never used a builder held at 69.0% to 69.3%.

    • 1,879 WordPress sites
    • HTTP Archive, Sept 2026

    Matt Suffoletto

    Founder, PageSpeed MattersPublished 7 min read

    25.3%

    Key Findings

    1. Among 1,715 top 1M WordPress sites that added a page builder and had CrUX LCP data on both sides, the share with a good LCP fell from 55.0% to 48.8%; 47,753 WordPress sites that never ran a builder held at 69.0% to 69.3% over a matched window.

    2. Of switchers that had a good LCP before the builder appeared, 25.3% lost it within three months, against 5.8% of WordPress sites that never used a builder.

    3. The median switcher's homepage grew by 171 KB, against 19 KB for WordPress sites without a builder; switchers' median JavaScript weight grew 26 KB and image weight 15 KB.

    4. The median switcher's LCP got 88 ms slower over the switch, while the median non-switcher got 9 ms faster.

    5. Top 100K switchers fell from 72.2% to 65.9% good LCP (176 sites), while never-builder WordPress sites in the same tier held at 78.0% to 77.8% (6,832 sites).

    6. Sites that switched between Dec 2024 and Jul 2025 saw the biggest drop, from 53.6% to 45.5% good LCP (895 sites), against 54.0% to 51.4% for switches from Aug 2025 to Jan 2026 (494 sites).

    7. Elementor was the builder behind 1,577 of the 1,879 switches (84%), followed by WPBakery (192), Divi (78) and Beaver Builder (32).

    Change in good LCP share after adding a page builder, in percentage points
    All switchers-6.2 pts
    Top 100K switchers-6.3 pts
    100K to 500K switchers-6.7 pts
    500K to 1M switchers-5.6 pts
    Never used a builder0.3 pts

    Source: PageSpeed Matters. HTTP Archive and Chrome UX Report history, Sept 2024 to Sept 2026.

    Negative values mean fewer sites with a good LCP after the switch.

    In this study

    Share of WordPress homepages with good LCP before and after adding a page builder

    GroupSites with LCP both sidesGood LCP beforeGood LCP afterMedian homepage growth
    All switchers1,71555.0%48.8%171 KB
    Never used a builder47,75369.0%69.3%19 KB
    Switchers, top 100K17672.2%65.9%250 KB
    Switchers, 100K to 500K87955.5%48.8%158 KB
    Switchers, 500K to 1M66049.8%44.2%172 KB
    Switched Dec 2024 to Jul 202589553.6%45.5%165 KB
    Switched Aug 2025 to Jan 202649454.0%51.4%220 KB
    Switched Feb 2026 to Jul 202632660.4%54.0%121 KB

    Source: PageSpeed Matters, Sept 2026. Real-user LCP from CrUX; sites without enough traffic for CrUX are not included.

    Summary

    Rebuilding a WordPress site with a visual page builder is one of the most common redesign paths. Builders make layouts easy to edit, but they also add markup, CSS and JavaScript. We wanted to see what happens to real-user loading speed when a site makes that switch.

    1,879 top 1M WordPress sites added a visual page builder. Of the 1,715 with real-user LCP data before and after, the share with a good Largest Contentful Paint fell from 55.0% to 48.8% within three months. WordPress sites that never used a builder held at 69.0% to 69.3% over a matched window.

    What we measured

    We used HTTP Archive's monthly crawls from September 2024 to September 2026 to track the four most common WordPress page builders. A switch is the first month a builder was detected on a site's homepage, after three straight WordPress months without one, with the builder still seen in at least three later months. 1,879 sites met that test.

    For LCP we used Chrome UX Report field data, which only covers sites with enough real Chrome traffic. 1,715 switchers had LCP data both before and after. We compared a three-month average before the switch with the switch month and the two months after. A site has a good LCP when its 75th percentile LCP meets Google's "good" threshold. The comparison group is 47,753 WordPress sites never detected with a builder, each measured over a matched three-month window.

    Builder detection relies on homepage markup, so a builder used only on inner pages is not counted. Elementor was behind 1,577 of the 1,879 switches (84%), followed by WPBakery (192), Divi (78) and Beaver Builder (32). We report all four together and do not compare builders with each other on speed.

    Good LCP fell after a builder arrived

    Among the 1,715 switchers with LCP data on both sides, the share with a good LCP fell from 55.0% to 48.8%, a drop of 6.2 percentage points. The 47,753 WordPress sites that never ran a builder moved the other way, from 69.0% to 69.3%.

    The comparison matters because LCP shifts over time for reasons unrelated to a rebuild, such as Chrome updates, hosting changes and seasonal traffic. The control group barely moved, so the drop among switchers is not a general trend across WordPress. It goes with the rebuild. A rebuild often brings a new theme, new images and new plugins at the same time, so the builder is likely one part of the change rather than the only one.

    A quarter of fast sites lost their good LCP

    Looking only at sites that started with a good LCP makes the effect clearer. Of switchers that had a good LCP before the builder appeared, 25.3% lost it within three months. Among WordPress sites that never used a builder, 5.8% lost a good LCP over the same kind of window.

    Some sites drop out of the good range in any three-month window, which is what the control group shows. Switchers dropped out far more often. For a site that already passes, that is a real risk to weigh before a rebuild.

    25.3%

    Share of WordPress sites with a good LCP that lost it within three months of adding a page builder.

    Homepages got heavier and slower

    The median switcher's homepage grew by 171 KB over the switch, against 19 KB for WordPress sites without a builder. Switchers' median JavaScript weight grew 26 KB and their median image weight 15 KB, both far smaller than the total. That suggests much of the extra weight sits in other resources such as HTML, CSS and fonts, which builders tend to add.

    LCP moved in the same direction. The median switcher's LCP got 88 ms slower, while the median non-switcher got 9 ms faster. The change for a typical site is modest, but it is likely enough to push sites close to the threshold out of the good range.

    The drop held across tiers and switch dates

    The pattern held at every popularity tier. Top 100K switchers fell from 72.2% to 65.9% good LCP (176 sites), while never-builder WordPress sites in the same tier held at 78.0% to 77.8% (6,832 sites). Switchers ranked 100K to 500K fell from 55.5% to 48.8%, and those ranked 500K to 1M from 49.8% to 44.2%.

    By switch date, sites that switched between Dec 2024 and Jul 2025 saw the biggest drop, from 53.6% to 45.5% good LCP (895 sites). Switches from Aug 2025 to Jan 2026 fell less, from 54.0% to 51.4% (494 sites), and switches from Feb 2026 to Jul 2026 went from 60.4% to 54.0% (326 sites). Every group lost ground. The differences between periods are likely a mix of sample size and what else changed on those sites at the time.

    What this means for your site

    • Measure before you rebuild. Record your current LCP, page weight and CrUX status for your key templates so you can tell what the rebuild changed.
    • Set a weight budget. Builder rebuilds in our data added a median 171 KB to the homepage. Decide in advance how much heavier the new pages are allowed to be, and test against that budget on staging.
    • Trim what the builder loads. Turn off widgets and modules you do not use, avoid nesting containers several levels deep and check that only the CSS and JavaScript each page needs is loaded.
    • Treat the LCP element with care. Keep the hero image properly sized, compressed and loaded early, rather than as a background image set by the builder.
    • Watch field data after launch. CrUX reflects the last 28 days of real visits, so check it in the weeks after launch and fix regressions before they settle in.

    Sources

    • HTTP Archive monthly crawls, September 2024 to September 2026
    • Chrome UX Report (CrUX) field data, September 2024 to September 2026
    • web.dev, Largest Contentful Paint: https://web.dev/articles/lcp

    Embed this research

    Paste this on your site to embed the chart. It links back to the source automatically.

    <iframe id="psm-page-builder-rebuilds" src="https://www.pagespeedmatters.com/embed/data-studies/page-builder-rebuilds" width="100%" height="600" style="width:100%;border:1px solid #E2E3E1;border-radius:12px" loading="lazy" title="Page Builder Rebuilds: 25.3% of WordPress Sites With Good LCP Lost It - PageSpeed Matters"></iframe>
<script>window.addEventListener("message",function(e){if(e&&e.data&&e.data.psmEmbed==="page-builder-rebuilds"&&e.data.height){var f=document.getElementById("psm-page-builder-rebuilds");if(f){f.style.height=e.data.height+"px";}}});</script>
<p style="font:14px/1.5 system-ui,sans-serif">Source: <a href="https://www.pagespeedmatters.com/resources/data-studies/page-builder-rebuilds">Page Builder Rebuilds: 25.3% of WordPress Sites With Good LCP Lost It - PageSpeed Matters</a></p>

    Frequently asked questions

    Do page builders slow down WordPress sites?

    Sites that added one tended to get slower. Among 1,715 WordPress sites that added a page builder, the share with a good LCP fell from 55.0% to 48.8%, while sites that never used a builder held at 69.0% to 69.3%. This is a correlation, and rebuilds often change more than the builder.

    How much weight does a page builder add to a homepage?

    The median switcher's homepage grew by 171 KB, against 19 KB for WordPress sites without a builder. Median JavaScript weight grew 26 KB and image weight 15 KB.

    Which page builder do most WordPress rebuilds use?

    Elementor, by a wide margin. It was behind 1,577 of the 1,879 switches we found (84%), followed by WPBakery (192), Divi (78) and Beaver Builder (32).

    How do I check if a rebuild hurt my LCP?

    Compare your Chrome UX Report LCP for the months before the rebuild with the months after, for example in PageSpeed Insights or Search Console's Core Web Vitals report. In our data, 25.3% of sites with a good LCP lost it within three months of adding a builder.

    A rebuild planned with a weight budget and checked against field data can keep a fast WordPress site fast. Need help with this? See our WordPress migration services.

    Share

    How we measured

    Methodology & Source

    Data source
    HTTP Archive monthly crawls with Chrome UX Report field data
    Collection date
    September 2026
    Sample
    1,879 WordPress sites
    Release
    HTTP Archive, Sept 2026

    Monthly HTTP Archive crawls from Sept 2024 to Sept 2026 for the four most common WordPress page builders. A switch is the first month a builder appeared after three WordPress months without one, with the builder seen in three or more later months. Good LCP compares a three-month CrUX average before the switch with the switch month and two after; the control is 47,753 WordPress sites that never ran a builder, over a matched window. Builder detection relies on homepage markup, and LCP covers only sites in CrUX. Builders are not compared with each other on speed.

    Using this data?

    Cite this study

    Link back to the study when you cite it.

    APA

    Matt Suffoletto. (2026). Page Builder Rebuilds: 25.3% of WordPress Sites With Good LCP Lost It. PageSpeed Matters. https://www.pagespeedmatters.com/resources/data-studies/page-builder-rebuilds

    Plain link

    Page Builder Rebuilds: 25.3% of WordPress Sites With Good LCP Lost It - PageSpeed Matters - https://www.pagespeedmatters.com/resources/data-studies/page-builder-rebuilds

    Ready to Get Started?

    Get your site optimized in as little as 3 days. From $500.

    See Plans & Pricing