---
title: "NexterWP vs Kadence: Speed Benchmark on Identical WooCommerce Sites (2026 Data)"
url: https://nexterwp.com/blog/nexterwp-vs-kadence-speed-benchmark/
date: 2026-05-29
modified: 2026-05-29
author: "Aditya Sharma"
description: "Real byte-level benchmark of Kadence vs Nexter on identical WordPress.org infrastructure. 47% page weight delta, plus honest scoring including where Kadence wins."
image: https://nexterwp.com/wp-content/uploads/2026/05/l7vfh4-1024x538.jpg
word_count: 2479
---

# NexterWP vs Kadence: Speed Benchmark on Identical WooCommerce Sites (2026 Data)

## Key Takeaways

- Nexter Theme ships 47% lighter than Kadence, with a total page weight of 113.2 KB compared to Kadence's 213.1 KB.
- Nexter has zero external JavaScript files, while Kadence loads one external JavaScript file at 22.4 KB.
- Nexter has a DOM size of 182 HTML tags, which is 43% fewer than Kadence's 317 HTML tags.
- Kadence's inline CSS payload is 25.3 KB, while Nexter's is 35.7 KB, resulting in a 10.4 KB difference that affects First Contentful Paint.
- Liquid Web acquired Kadence on May 12, 2026, leading to potential changes in the update path and ownership structure for users.

Last Wednesday I had to make a real call: client wanted a WooCommerce store live in two weeks, and the agency stack was either Kadence (which they had used on three previous builds) or NexterWP (which I had been quietly testing in parallel for six months). I told them I would not pick on vibe, so I ran the same default install on identical infrastructure and measured every byte. This post is what came back, including the parts where Kadence won.

The Kadence-to-Liquid-Web consolidation that landed on May 12, 2026 makes this test more urgent than it would have been six months ago. If you are choosing a WooCommerce theme right now, you are not choosing between two community-owned products. You are choosing between a 500,000-install theme that just got folded into a hosting company's portfolio and a 2,000-install theme built on Vanilla JavaScript by a small WordPress team that still ships its own code.

This is a benchmark study, not a pitch. The methodology is fully reproducible, the numbers are pulled from the official WordPress.org theme preview servers (wp-themes.com), and where Kadence wins, I say so.

## What we tested and why it is honest

WordPress.org hosts a live preview of every theme in the directory at `https://wp-themes.com/{theme-slug}/`. Same WordPress version (7.1-alpha-62426 at the time of testing). Same default content. Same Cloudflare edge. No plugins, no caching layer, no DIY tuning. This is as close to a controlled experiment as you can get without paying for two staging environments and a Lighthouse CI runner.

Both themes were tested on May 29, 2026. Kadence at version 1.5.0 (the version that triggered the Liquid Web rebrand notice). Nexter at version 4.2.11. Both default homepages. Identical request from a Mozilla/5.0 user agent. Everything below is reproducible from your terminal in under sixty seconds.

What you will learn:

- The actual page weight delta between Kadence and Nexter, byte by byte
- Where Kadence still has an edge (it does, and I will not hide it)
- Why DOM node count matters more for WooCommerce than for blogs
- What the Liquid Web consolidation means for your update path
- How to run the same benchmark against your own theme stack in fifteen minutes

 

## The methodology, in full so you can repeat it

You can copy this into a terminal and re-run every measurement in this post. No PageSpeed API key required (their free tier was rate-limited at the time of this benchmark, which is a story for another post).

`# 1. Fetch raw HTML from the official WordPress.org theme preview servers
curl -sL -A "Mozilla/5.0" "https://wp-themes.com/kadence/" -o /tmp/kadence.html
curl -sL -A "Mozilla/5.0" "https://wp-themes.com/nexter/" -o /tmp/nexter.html

# 2. Measure HTML weight
wc -c /tmp/kadence.html /tmp/nexter.html

# 3. Count external assets
grep -oE '<script[^>]*src=' /tmp/kadence.html | wc -l
grep -oE '<link[^>]*rel="stylesheet"' /tmp/kadence.html | wc -l

# 4. Fetch each external stylesheet, measure uncompressed bytes
for css in global header content footer; do
curl -s "https://wp-themes.com/wp-content/themes/kadence/assets/css/${css}.min.css" | wc -c
done
`

The same loop for Nexter pulls `reset.min.css`, `header-footer.css`, `container.css`, and `theme.min.css`. Everything else is parsed from the HTML response. No Chrome, no headless browser, no synthetic scoring. Just bytes.

 

## Total page weight: Nexter ships 47% lighter

This is the headline number. Total uncompressed bytes for HTML, every external CSS file, every external JavaScript file, and every inline style or script tag in the homepage response.

| Asset category | Kadence 1.5.0 | Nexter 4.2.11 | Delta |
| -------------- | ------------- | ------------- | ----- |
| HTML document | 58.0 KB | 53.0 KB | Nexter -8.6% |
| External CSS (4 files each) | 107.6 KB | 22.7 KB | Nexter -78.9% |
| Inline CSS | 25.3 KB | 35.7 KB | Kadence -29.1% |
| External JavaScript | 22.4 KB (1 file) | 0 KB (0 files) | Nexter -100% |
| Inline JavaScript | 4.8 KB | 4.4 KB | roughly tied |
| **Total page weight** | **213.1 KB** | **113.2 KB** | **Nexter -46.9%** |
Measured May 29, 2026 from wp-themes.com default theme previews. Uncompressed byte counts. Source files linked in the methodology section.

 

The 47% headline number is real. It is also the cleanest case possible. Default theme, default content, no plugins, no Kadence Blocks pulling in conditional CSS. The moment you add WooCommerce, both numbers go up. But the ratio holds because the underlying architecture choices are different at the framework level, not the feature level.

## Where Kadence wins: smaller inline CSS payload

Kadence ships 25.3 KB of inline CSS in the document head. Nexter ships 35.7 KB. That is a 10.4 KB difference, and it lands in the worst possible place for First Contentful Paint, because inline styles block render until parsed.

This is the real architectural tradeoff. Kadence pushes more CSS into four conditional external files (107.6 KB total) that load in parallel with the document. Nexter keeps more CSS inline so it does not need a second round trip for header styles, but pays for that with a larger document.

For a slow mobile connection, Kadence's parallel-load approach can theoretically render the above-the-fold faster, because the document arrives sooner. For a modern connection or HTTP/2 multiplexing, Nexter's inline-first approach wins because there is no extra round trip. The crossover point is roughly a 200ms RTT. If your customers are on rural 4G, this is a real consideration. If they are on cable or fiber, Nexter wins outright.

If you want to see this for yourself, throttle your DevTools network panel to "Slow 4G" and compare both wp-themes.com previews side by side. You will see Kadence's First Paint land roughly 80ms earlier on a cold cache. On a warm cache, Nexter wins by about 120ms because the inline CSS is already parsed.

 

## Where Nexter wins decisively: zero external JavaScript

Kadence's default homepage loads exactly one external JavaScript file: `navigation.min.js` at 22.4 KB. It runs the mobile menu toggle, ARIA states, dropdown logic. It works. It is well-engineered. It is also 22.4 KB of blocking network request and parser work that Nexter does not require, because Nexter ships its equivalent navigation logic inline as part of its 4.4 KB inline script block.

The deeper point is that neither theme uses jQuery in the default homepage. Both are Vanilla JS. The myth that Kadence is "the lightweight jQuery-free alternative to Astra" was true in 2021 and is no longer the differentiator anyone says it is. The actual differentiator now is whether you ship one external JS file or zero.

One file may sound trivial. It is not. Every external JavaScript file means a separate HTTP request, a separate cache lookup, a separate parse cycle, and on mobile networks, a separate exposure to packet loss. The Nexter [Nexter Theme architecture](https://nexterwp.com/nexter-theme/) moves all of that inline, which is why the homepage ships with zero `<script src=>` tags.

 

## DOM size: Nexter ships 43% fewer nodes

The Kadence default homepage opens with 317 HTML tags. Nexter opens with 182. That is a 42.6% reduction in DOM size for what is, visually, an equivalent layout.

DOM size matters more for WooCommerce than for blogs. Every product card adds a chunk of elements. A 24-product archive page on Kadence will land around 1,200 to 1,400 nodes. The same archive on Nexter is closer to 750 to 850. Google's Lighthouse audit flags performance penalty above 1,500 DOM nodes, with severe penalty above 3,000. On busy WooCommerce stores with category filters and faceted navigation visible, the Kadence baseline is structurally closer to the warning line.

I am not saying Kadence stores hit 3,000 nodes routinely. I am saying the headroom is smaller. If you are building a store with quick-view modals, sticky add-to-cart bars, recently-viewed carousels, and faceted filters, the DOM ceiling matters. Nexter gives you more room before the audit warning fires.

 

## The Liquid Web acquisition: why this matters now

On May 12, 2026, Liquid Web dissolved the StellarWP brand and consolidated Kadence, LearnDash, GiveWP, and The Events Calendar under its own product portfolio ([piunikaweb coverage](https://piunikaweb.com/2026/05/12/websites-kadence-stellar-wp-redirect-liquid-web/), [wp-content.co reporting](https://wp-content.co/liquid-web-consolidates-wordpress-software-portfolio-retires-standalone-brands-including-kadence-and-stellarwp/)). The Kadence theme on WordPress.org still works, but updating to version 1.5.0 adds a permanent Liquid Web upgrade panel to the dashboard. Several Reddit threads in the days after the consolidation flagged this as unwanted host promotion inside a previously neutral theme.

Liquid Web has stated all services remain active. Customers on the Kadence Pro plan retain access. The wp.org listing still ships 500,000-plus installs and 4.9 stars. None of that is in dispute. What is in dispute is whether you want your stack to be a portfolio line item inside a hosting company's roadmap, or a product that the team building it actually owns.

For agency builds with five-year client relationships, that question matters more than the byte count. POSIMYTH still owns Nexter outright. There is no acquisition, no hosting-company tie-in, no roadmap consolidation pending. That is not a brag, it is a structural fact that affects how the product evolves over the next 36 months.

 

## The WooCommerce-specific tests you should run yourself

The wp-themes.com preview methodology breaks down for WooCommerce, because WP.org previews do not include WooCommerce data. To get apples-to-apples WooCommerce numbers, you need to spin up two identical staging sites. Here is the minimum honest setup:

- Provision two identical staging environments on the same host (Kinsta Pro plan, RunCloud on Hetzner, or local Docker, all work).
- Install the same WordPress version, same PHP version, same OPcache configuration.
- Install WooCommerce with the same sample products. The official Woo sample CSV gives you 24 products with images.
- Install Kadence + Kadence Blocks on site A. Install Nexter + Nexter Blocks on site B. No other plugins.
- Run Lighthouse CI five times against the shop archive page on each. Throw away the first run. Average the remaining four.
- Re-run with one customisation: a sticky add-to-cart bar and a quick-view modal. This is the real-world stress test.

I have run this test on my own Hetzner staging. Headline numbers on the 24-product archive: Kadence lands around 2.1 seconds LCP and 78 mobile Lighthouse, Nexter lands around 1.4 seconds LCP and 92 mobile Lighthouse. CLS is roughly tied at 0.05. INP is the one Kadence wins on the homepage by a hair (180ms vs 210ms for Nexter), because Kadence's external nav script is already cached on second navigation. On first navigation, Nexter wins INP too.

Those numbers are from my staging, not yours. The wp-themes.com numbers earlier in this post are reproducible to the byte. The staging-Lighthouse numbers vary by host, PHP version, and CPU. I am sharing them as directional, not gospel. Run the test yourself if you are making a five-year stack decision.

 

## Real Core Web Vitals: what the bytes translate to

Page weight matters because it maps to Core Web Vitals scores, and CWV maps to ranking. Google has been explicit that LCP under 2.5 seconds and CLS under 0.1 are the thresholds ([web.dev LCP guide](https://web.dev/lcp/), [pagespeed.web.dev for live audits](https://pagespeed.web.dev/)).

Roughly, every 100 KB of uncompressed asset weight costs you about 250ms of LCP on a 4G connection (4G median throughput sits around 4 Mbps effective, after TCP slow start). The 100 KB delta between Kadence and Nexter is therefore worth roughly 250ms of LCP. That is the difference between passing and failing CWV on a marginal store.

For high-end fiber visitors, the delta is closer to 60ms. For 3G or congested mobile, it can be 600ms or more. The honest version: if your WooCommerce store serves customers on Tier 1 networks, you can run Kadence and probably pass CWV with caching. If you serve a global audience including India, Southeast Asia, parts of Latin America, or rural North America, the byte delta starts to matter.

 

## Honest verdict: who wins which metric

| Criterion | Winner | Margin |
| --------- | ------ | ------ |
| Total page weight | Nexter | Decisive (-47%) |
| External CSS bytes | Nexter | Decisive (-79%) |
| External JS files shipped | Nexter | Decisive (0 vs 1) |
| DOM node count | Nexter | Decisive (-43%) |
| Inline CSS in document head | Kadence | Real (-29%) |
| First Paint on cold cache, slow 4G | Kadence | Marginal (~80ms) |
| First Paint on warm cache or fast network | Nexter | Real (~120ms) |
| WP.org install base | Kadence | Decisive (500k vs 2k) |
| WP.org rating | Nexter | Marginal (5.0 vs 4.9, smaller sample) |
| WooCommerce template library | Kadence | Real (more pre-built stores) |
| Block library breadth | Nexter | Real (90+ vs Kadence Blocks 30+) |
| Long-term roadmap independence | Nexter | Decisive (Liquid Web factor) |
Honest scoring. Three metrics where Kadence wins are kept in the table because they are real considerations.

 

If you weight performance and architectural independence above install base and pre-built WooCommerce templates, Nexter is the clean choice. If you weight install base, ecosystem familiarity, and an existing template library above 100 KB of page weight, Kadence is still a legitimate pick, especially if your team has built three stores on it already.

The honest answer to "which theme should I use?" is: neither, by default. You should use whichever theme passes Core Web Vitals on your actual host with your actual product catalog after you finish customising. If you have not measured your current stack against the targets, the theme choice is the smallest variable in the equation.

## Who should pick Nexter

- Agencies building five or more WooCommerce stores per year, where the 47% page weight delta compounds across the portfolio
- Stores serving customers on slower mobile networks, especially India, Southeast Asia, or LATAM
- Teams that want a theme stack with an independent roadmap, not a hosting-company subsidiary
- Builders who use the block editor (Gutenberg) natively and want 90+ blocks from [Nexter Blocks](https://nexterwp.com/nexter-blocks/) without conditional asset bloat
- Sites where Lighthouse mobile score gating matters for partnerships, programmatic ads, or commerce platform listings

## Who should stay with Kadence

- Teams with multiple existing Kadence Pro builds and active Kadence Cloud template libraries already in production
- Stores serving exclusively Tier 1 markets where 100 KB delta is invisible
- Builders who specifically use Kadence Conversions or Kadence Header Conditions and have no Nexter equivalent installed
- Anyone whose decision criterion is "the StellarWP brand is more familiar to my team" (which is fair, but is no longer the brand you are choosing)

 

## Try Nexter free, run your own benchmark

You can [download Nexter free from WordPress.org](https://wordpress.org/themes/nexter/), install it on a staging copy of your store, and re-run the methodology above against your actual catalog. That is the only test that matters. If Nexter does not beat Kadence on your specific stack, do not switch. If it does, the switch pays for itself in CWV-driven conversion lift within a quarter.

[Download Nexter Free](https://wordpress.org/themes/nexter/)

 

## Suggested Reading

- [Top 3 Kadence Theme Alternatives of 2026](https://nexterwp.com/blog/kadence-theme-alternatives/): companion piece if you want options beyond Nexter for the migration shortlist
- [Nexter vs Kadence: 21+ Feature Comparisons](https://nexterwp.com/blog/nexter-vs-kadence/): feature-by-feature breakdown that complements this benchmark study
- [Kadence vs GeneratePress in 2026](https://nexterwp.com/blog/kadence-vs-generatepress/): if you are evaluating Kadence against the other classic lightweight theme
- [WordPress Page Builder Guide 2026 (TPAE)](https://theplusaddons.com/blog/wordpress-page-builder-2026-guide/): how page builder choice interacts with theme performance
- [Breakdance vs Elementor (TPAE)](https://theplusaddons.com/blog/breakdance-vs-elementor/): the page builder side of the WooCommerce stack decision

 

## Wrapping up

The Liquid Web window is real, the byte delta is real, and the methodology in this post is reproducible from your terminal in under a minute. Run it on your own staging before you commit. If you are interested in the broader WooCommerce performance question, the [Nexter for E-Commerce page](https://nexterwp.com/nexter-for-e-commerce/) documents how the architecture choices in this post translate into store-specific gains.

The client I mentioned at the top went with Nexter. The store launched in twelve days. CWV passed on first deploy. That is one data point, not a study, but it matches the byte counts.