Learn how to speed up a slow WordPress site with 7 fixes ranked by effort and impact, from image compression to caching, plus how to measure each one.

Photo by Rumman Amin on Unsplash

If you want to know how to speed up a slow WordPress site, start with three moves: turn on page caching, compress your images, and update PHP and WordPress. Those changes fix most slowdowns for most sites, and they can be done in an afternoon. Below, we rank seven fixes by effort and impact, and we show which metric to check after each one so you know what actually worked.

Most slow sites lose time in one of three places. The server takes too long to build the page, the browser downloads too much before the main content appears, or the page runs heavy code. Each fix below targets one of those problems, so you can match the fix to the symptom instead of guessing.

Seven fixes at a glance

Rank	Fix	Effort	Typical impact	Metric to check
1	Turn on page caching	Low	High	TTFB
2	Compress and resize images	Low to medium	High	LCP
3	Update PHP and WordPress	Low to medium	Medium to high	TTFB
4	Audit and remove plugins	Medium	Medium to high	TTFB and INP
5	Fix lazy loading for the hero image	Low	Medium	LCP
6	Add a CDN	Medium	Medium, depends on your visitors	LCP
7	Move to better hosting	High	High if the server is the bottleneck	TTFB

TTFB means time to first byte, the delay before the server starts sending the page. LCP means Largest Contentful Paint, the time until the main content appears. The impact column is our estimate for a typical WordPress site. Your theme, plugins, and traffic will change the results.

Set a baseline before you change anything

Without a baseline, you cannot tell whether a change helped. Run these steps first:

1. Pick three URLs: your homepage, a typical blog post, and your most important landing or product page.
2. Test each one in Google PageSpeed Insights on both mobile and desktop.
3. Record LCP, INP, CLS, and TTFB, along with the date and time.

Use these thresholds to read the results. For field data, measured at the 75th percentile of real visits, a good score means LCP of 2.5 seconds or less, INP of 200 milliseconds or less, and CLS of 0.1 or less. Poor means LCP above 4 seconds, INP above 500 milliseconds, or CLS above 0.25, according to Anderson Collaborative's Core Web Vitals guide.

INP, or Interaction to Next Paint, replaced First Input Delay as a Core Web Vital in March 2024. It measures how quickly the page responds when someone clicks or taps. LCP breaks into four parts: time to first byte, resource load delay, resource load duration, and element render delay. Knowing which part is slow tells you which fix to try first. Server work shows up in TTFB, while a heavy hero image shows up in resource load duration.

1. Turn on page caching (low effort, high impact)

Without caching, WordPress builds every page from PHP and database queries each time someone visits. A caching plugin saves a finished copy of the page and serves that copy to visitors. Popular options include WP Rocket and W3 Total Cache. Turn on page caching first, then check that the plugin clears the cache when you publish or update content, so visitors never see outdated pages.

Caching mostly improves TTFB. Test it while logged out, because most caching plugins skip the cache for logged-in users, including you when you manage the site.

What to measure: TTFB on your homepage before and after. If TTFB drops sharply while the rest of the page stays the same, caching did its job.

2. Compress and resize your images (low to medium effort, high impact)

Images are often the heaviest part of a page, and the largest one above the fold is frequently your LCP element. Start with size. A 3000-pixel photo displayed at 800 pixels sends far more data than the screen needs. Resize each image to its display width, compress it, and use a modern format such as WebP. WebP files are often smaller than JPEG files at similar quality, though the savings vary from image to image.

Most image optimization plugins can compress your media library in bulk. Keep copies of the originals so you can re-export them later if you change your layout.

What to measure: LCP on the pages with the biggest images. PageSpeed Insights also lists image opportunities in its diagnostics, so you can see exactly which files are the worst offenders.

3. Update PHP and WordPress (low to medium effort, medium to high impact)

Newer PHP versions generally run WordPress faster, and most hosts let you switch the PHP version from the control panel. Choose a version that your theme and plugins support, then update WordPress core, your theme, and every plugin. Back up your site first and test on a staging copy when you can, because a jump in PHP version can break older plugins.

What to measure: TTFB. Changes in server processing appear there first.

4. Audit and remove plugins (medium effort, medium to high impact)

Every plugin adds code to your pages, and some run database queries on every load. Deactivate and delete anything you do not use. Look for overlaps, such as two SEO plugins or two caching plugins competing with each other. For each plugin you keep, confirm that it still does something you actually need.

Change one plugin at a time. Disable one, retest, and then decide whether to keep it. This turns a messy cleanup into a clear list of what helped and what did not.

What to measure: TTFB for server-side plugin work, and INP for scripts that make the page slow to respond when someone clicks.

5. Fix lazy loading for the hero image (low effort, medium impact)

WordPress 5.5 and later lazy loads images natively, which means images further down the page load only as readers scroll toward them. That saves bandwidth. The trap is the image at the top of the page. If your hero image is lazy loaded, the browser waits before fetching it, and that delays LCP.

Keep lazy loading for images below the fold, and make sure the first large image loads right away. Some themes and plugins apply lazy loading to every image, so check the page source for your hero image.

What to measure: LCP. If PageSpeed Insights flags a resource load delay on your main image, lazy loading is a likely cause.

6. Add a CDN (medium effort, impact depends on your visitors)

A content delivery network, or CDN, stores copies of your static files, such as images, CSS, and JavaScript, on servers around the world. Visitors download those files from a nearby location instead of from your main server. The gain is largest when many visitors live far from your host. If most of your audience sits close to your server, expect a smaller improvement.

A CDN mainly speeds up resource load duration. It helps less with HTML that is generated fresh for each visitor, unless the CDN also caches full pages.

What to measure: LCP, compared before and after the change on the same pages.

7. Move to better hosting (high effort, high impact when the server is the bottleneck)

Hosting comes last because it takes the most work. A migration means copying files and databases, updating DNS, and checking everything afterward. But if TTFB stays high after caching, updates, and plugin cleanup, the server itself is probably the bottleneck. Crowded shared servers often show this pattern.

Look for hosts that offer server-level caching, a current PHP version, and enough resources for your traffic. Ask the host directly how it handles caching and whether it provides staging. Then compare the new host's TTFB with your baseline using the same pages. If you are still deciding which platform to build on, read our Webflow vs WordPress comparison before you commit to a hosting plan.

What to measure: TTFB before and after the move. If TTFB does not improve, the problem is more likely in your theme, plugins, or database than in the host.

Also worth checking: the Heartbeat API and database cleanup

The WordPress Heartbeat API keeps the editor and dashboard in sync by sending regular AJAX requests to admin-ajax.php. These requests are normal, but on a server with limited resources they add load while you edit. Some optimization plugins let you slow Heartbeat down or limit it to certain screens. Database cleanup, such as removing spam comments and expired transients, also helps, though the gains are usually smaller than the seven fixes above.

How to measure changes without fooling yourself

Measurement is where most speed projects go wrong. Keep these rules in mind:

• Change one thing at a time. Make a single change, retest the same URLs, and write down the result.
• Lab results come first. PageSpeed Insights lab data is collected in a controlled test on a mid-tier mobile device over a simulated mobile network, so you can see a change within minutes.
• Field data takes time. PageSpeed Insights field data uses a rolling 28-day collection period, so a fix can take weeks to appear. Google's PageSpeed Insights documentation explains how the two data types differ.
• Lab improved but field is still poor? Do not assume the fix failed. Field data reflects real visitors on many devices and networks, and it covers a 28-day window. Give it time before you judge.
• Low-traffic sites need a different approach. If your site does not get enough visits for Google to collect field data, rely on lab tests. Run each test several times and compare typical results, not one lucky run.

Ahrefs also recommends testing changes on a staging copy first and comparing before-and-after results, as covered in its WordPress speed guide.

Frequently asked questions

Can I speed up WordPress without a plugin?

Yes, partly. You can update PHP and WordPress, remove unused plugins and themes, resize images before you upload them, and choose a host with caching built in. Page caching is easiest with a plugin, although some hosts provide it at the server level.

Is it safe to delete the cache?

Yes. WordPress and your caching plugin rebuild the cache on the next visit. The first few visitors after clearing may see a slightly slower page, which is normal. Clear the cache after major design or content changes.

Do I need a developer to speed up my WordPress site?

Probably not for caching, image compression, updates, or plugin cleanup, since those are settings and uploads. A developer can help with hosting migrations, custom theme code, or database work, especially if your site handles orders or memberships. If you are still choosing a platform, our Webflow vs WordPress comparison covers the trade-offs.

Why is my mobile score lower than my desktop score?

Mobile tests use a slower simulated device and a slower network, so the numbers are harder to hit. Compare mobile results with earlier mobile results, not with desktop.

How long until field data shows my changes?

Field data uses a rolling 28-day window, so expect several weeks. Lab results update almost immediately after each test.

The takeaway

Start with page caching, image compression, and updates, since they deliver the biggest gains for the least work. Then clean up plugins and fix lazy loading for your hero image. Add a CDN or change hosts only if your numbers point to that bottleneck. Measure each change on the same URLs, one at a time, and give field data a few weeks. That approach tells you exactly what worked.