There are thousands of WordPress themes. Many theme developers claim that their theme is super-fast and very lightweight. Speed Up Press tries to verify those claims and tests thoroughly the performance of interesting and popular WordPress themes in a standardized and transparent speed test environment.
To reduce the effect of variability, which is inherent in such performance tests, three testing platforms are used:
- Lighthouse installed on a local machine near the server hosting the tested website.
Lighthouse is run through the command line. For consistency, all tests in a calendar year use the same Lighthouse version.
This third-party platform also uses the Lighthouse software, but we have no influence on the Lighthouse version.
Additionally, each speed test on one platform is a series of 15 runs, out of which median and average values are calculated. Furthermore, a single run of webpagetest.org consists of 5 tests and the result of a single run is the median value of these 5 tests. Each row in the table contains a link to more detailed test results with additional metrics, screenshots and PDF reports.
There are several metrics used in web performance tests. In the past, measuring the fully loaded time and the onload event were a popular way of testing how fast a website is. Since the introduction of Web Vitals as a search engine ranking factor by Google, the focus of performance optimization has switched to the following metrics: Largest Contentful Paint (LCP), Cumulative Layout Shift (CLS) and Total Blocking Time.
In the below table, tested themes are ranked according to their Largest Contentful Paint timings. An average value of the results from the three platforms is calculated, and the table is sorted by that column. This approach reduces significantly the variability and should show us the fastest WordPress themes, both free and premium.
|Rank||Theme||Edition||Version||LCP (Avg.)||LCP (LH)||LCP (GTM)||LCP (WPT)||CLS||Blocking Time||Requests||Byte Weight||Test details|
|1. Go 1.6.0||Go||-||1.6.0||0.302 s||0.373 s||0.238 s||0.223 s||0.002||0||11||81,270||Link to the Test|
|2. GeneratePress (free) 3.1.0||GeneratePress||free||3.1.0||0.304 s||0.385 s||0.256 s||0.191 s||0.000||0||8||33,035||Link to the Test|
|3. OceanWP (free) 3.2.1||OceanWP||free||3.2.1||0.371 s||0.416 s||0.313 s||0.338 s||0.000||0||22||230,412||Link to the Test|
|4. Twenty Nineteen 2.2||Twenty Nineteen||-||2.2||0.388 s||0.430 s||0.361 s||0.330 s||0.000||0||10||64,033||Link to the Test|
|5. Blocksy (free) 1.8.16||Blocksy||free||1.8.16||0.389 s||0.430 s||0.408 s||0.287 s||0.000||0||7||49,264||Link to the Test|
|6. Kadence (free) 1.1.11||Kadence||free||1.1.11||0.392 s||0.431 s||0.360 s||0.334 s||0.000||0||11||50,764||Link to the Test|
|7. Lightning (free) 14.19.1||Lightning||free||14.19.1||0.420 s||0.350 s||0.465 s||0.515 s||0.008||0||20||303,848||Link to the Test|
|8. Astra (free) 3.7.6||Astra||free||3.7.6||0.448 s||0.526 s||0.398 s||0.341 s||0.000||0||8||45,249||Link to the Test|
|9. MH Magazine (Lite) 2.9.2||MH Magazine||Lite||2.9.2||0.523 s||0.608 s||0.454 s||0.422 s||0.022||0||12||130,856||Link to the Test|
|10. Avada 7.6.1||Avada||-||7.6.1||0.570 s||0.519 s||0.580 s||0.663 s||0.075||0||14||472,121||Link to the Test|
|11. Phlox (free) 2.9.3||Phlox||free||2.9.3||0.601 s||0.594 s||0.659 s||0.567 s||0.000||37.916||28||441,175||Link to the Test|
New themes will be tested and added on a regular basis.
Acronyms and Terminology
LCP – Largest Contentful Paint
LH – Google Lighthouse
WPT – WebPageTest.org
GTM – GT Metrix
CLS – Cumulative Layout Shift
Methodology and Workflow
As such comprehensive tests involve a lot of data, the workflow is semi-automated. WebPageTest and Lighthouse provide very detailed JSON files with test results. Some of the metrics may be rounded in the results shown on the WebPageTest.org website or in Google’s PageSpeed Insights, but in the JSON files you will find exact values.
The JSON files provided by these tools are uploaded to WordPress and a custom-made plugin parses the relevant values and saves them into WordPress custom fields, which are managed by Meta Box. The tables are then generated automatically. GT Metrics provides HAR files instead of the JSON format, but they can be converted to JSON. The data provided by GT Metrix in the HAR files is not as detailed as in the other tools. The metrics missing in the HAR files are entered manually as custom fields.
For the purpose of this theme performance comparison, no caching plugins or other caching technologies (e.g. Varnish, Redis) are used. By using a caching plugin, you can significantly reduce the time to first byte and the overall page load times. However, using caching technologies would increase variability and might not show us the true performance of the tested theme.
The above theme ranking shows page loading times for a very simple page with practically no content and a very simple navigation menu. System stack fonts are used if the given theme offers such an option through its graphical user interface. If deactivating web fonts is only possible through adding custom code, then the theme’s default fonts are used.
To calculate the Average LCP Score the following weighting is used: Lighthouse – 50%, GT Metrix – 25% and WebPageTest – 25%. Both GT Metrix and WebPageTest are set to measure the performance on desktops, whereas the Lighthouse result simulates the performance on mobile devices.
As Lighthouse tests are run through the command line on Windows, the testing environment is under full control and kept constant. Both GT Metrix and the operators of WebPageTest may upgrade the Lighthouse or browser versions, which may or may not affect the test results over time. These are the reasons why the Lighthouse results are given a little more importance when calculating the average LCP score for the purpose of this WordPress theme performance comparison.
The purpose of this theme ranking is to benchmark the performance of popular WordPress themes in a standardized environment. The actual page load times of your website will obviously be different, as they depend on many factors, most importantly on your content, server performance and caching technologies.
Nevertheless, this comparison should be useful to see how one theme relates to another in terms of performance.
Do you want to have quick speed improvements, without rebuilding your site and spending a fortune?
1. Invest a few dollars in a super-fast, yet affordable and scalable Vultr High Frequency server through Cloudways. Especially, if you are coming from shared hosting, the performance improvements should be significant. Choose a server location close to your audience.
2. Buy a license and install WP-Rocket – a very robust and powerful caching plugin for WordPress with a simple set-up. Enable link preloading – it will significantly improve perceived page load times for visitors navigating your site.