How we review

Every review on lovewp is based on tests we actually run. This page explains exactly how, so you can judge the results for yourself.

Themes and plugins: the test bench

We test every theme and plugin on the same clean WordPress site, built fresh for each test with WordPress Playground (the current version of WordPress, PHP 8.3). The site uses Twenty Twenty-Five, WordPress’s default theme, and one standard test page: a small bakery homepage with text, an image, a list, two columns, a quote, a table and a button.

We measure that page without the product, then install the product from WordPress.org, activate it on its default settings and measure the same page again. Each measurement is taken three times with Google Lighthouse (mobile profile) and we report the median.

What we report:

  • Page weight: the total size of every file the page needs, before compression. It doesn’t depend on the server, so it compares directly between products.
  • Where the weight comes from: CSS, JavaScript, fonts, images or HTML.
  • Extra requests and any files loaded from other websites.
  • PHP errors and warnings logged while the product runs.
  • Admin menu items added, so you know how much it adds to your dashboard.
  • Screenshots of the product’s own admin screens, taken during the test.

The limits, in plain terms:

  • Load times from the test bench come from a local test server, so we only use them as a rough before-and-after guide, never as real-world speeds.
  • Our test page doesn’t use the product’s own blocks or shortcodes, so the result is what the product adds to every other page of your site.
  • Premium-only features can’t be tested without a licence. When we get one from the company, we say so on the review.
  • Some products can’t show their main benefit on our test server (for example caching that only works on certain web servers). The review says so when that happens.

Hosting: the hosting bench

For hosting, we measure a live WordPress site on the host from our test computer in the UK. Each test makes 20 normal requests and 20 with caching bypassed, every one on a fresh connection, then measures the page three times with Lighthouse. We run the test twice, at different times of day, and report the average.

  • Time to first byte: how long the server takes to start sending the page. We report the median and the slowest 10% of requests.
  • Caching bypassed: what you get when the host has to build the page fresh, which shows the server’s raw speed.
  • Failed requests, page caching and CDN, as we found them.

We pay for the hosting we test. If a host ever gives us a free account, the review will say so. Each hosting review also says which site we measured.

The verdicts

Every review ends with one of three verdicts:

  • Install it: worth paying for, or free and excellent. We’d put it on our own site.
  • Free is enough: the free version does the job. Upgrade only if you hit a wall.
  • Skip unless…: good for a narrow case. We tell you exactly which one.

The love score (out of five hearts) is how much we enjoyed using the product on a real site. It is only given after hands-on use, never from the test bench alone.

What we never do

  • Change a verdict because of a commission. Verdicts are decided before any affiliate link is added.
  • Accept payment for a review or a place in a list.
  • Present a company’s own benchmarks or claims as our results. When we quote them, we link to the source.
  • Hide the downsides. Every review lists what to watch out for.

Keeping reviews current

Every review shows the WordPress version and the date it was tested. We re-test after major releases and at least once a year, and update the date only when we’ve actually re-checked something.

Corrections

If you think we got something wrong, please tell us through the contact page. We check every report, fix what’s wrong and note the correction and its date on the review.

Some links on lovewp are affiliate links. Read our affiliate disclosure for how that works.