The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Check a WordPress site in layers rather than trusting one speed score. Start with Tools > Site Health in the dashboard, test representative public pages in PageSpeed Insights, use WebPageTest when you need a waterfall-level diagnosis, and interpret LCP, CLS, INP, FCP and TTFB together. Make one controlled change at a time, rerun several times under the same conditions, and classify the bottleneck as front end, server or third party before changing plugins, hosting or code.
Antihost offers managed WordPress hosting with free migration, one-click staging and a 14-day free trial.
What a WordPress performance check should tell you
“Is my site fast?” is too broad to produce a useful fix. A proper check answers four separate questions:
As an Amazon Associate I earn from qualifying purchases.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minute- Loading: How quickly does the browser receive and display the main content?
- Visual stability: Does the layout stay in place, or do images, ads, fonts and widgets move it while loading?
- Responsiveness: After a visitor taps, clicks or types, how quickly does the page respond?
- Server delivery: How long does the server take to begin returning the document and its data?
WordPress performance is a chain. A slow database query can delay the HTML; a large hero image can delay the largest visible element; a consent script can block interaction; and a third-party video or advertising request can keep the page busy after the server has responded. The measurement method must therefore match the question you are trying to answer.
Prepare a representative test set
Testing only the homepage can hide a problem in a template used by most visitors. Before opening a testing service, create a short list of final, public URLs that represent the site’s important experiences.
#1 Best Overall
- Used Book in Good Condition
Pages to include
- Homepage
- A typical blog post or article
- A landing page with the site’s main conversion goal
- A category, tag or other archive page
- A product detail page and checkout page when the site sells online
This is a practical sampling method, not a mandatory WordPress list. Choose pages that use different templates, image treatments, forms, embeds and plugins. Record each exact final URL, including whether it redirects, so every later run tests the same destination.
Record the test conditions
Write down the device mode (mobile or desktop), browser, test location, connection profile and date. Keep those variables unchanged when comparing a before-and-after result. If a page is personalized, bypass the logged-in dashboard view and test the public visitor experience separately. Do not compare a cached page in one run with an uncached page in another without noting the difference.
Start with WordPress Site Health
Site Health is the WordPress-native inspection and should be your first check because it exposes configuration problems that a public speed test cannot explain.
- Sign in to WordPress and open Tools > Site Health.
- On the Status tab, review the groups for critical issues, recommended improvements and passed tests.
- Open the Info tab and expand the sections for WordPress, themes, plugins, media, server, database, constants and filesystem.
- Use the export option to save the information before making changes. It gives you a baseline for a developer or hosting provider.
What Site Health can reveal
Site Health may flag an outdated PHP version, failed updates, loopback requests that do not complete, blocked HTTP requests or other operational conditions. Those findings do not automatically prove that a page-speed score is low, but they can explain failed background tasks, missing cache generation, broken integrations or unreliable administration.
The Server section documents the PHP version, limits and web-server environment that are relevant to response time. The Database section helps establish what WordPress is using, while the Plugins and Themes sections show how much code is active. Treat this information as a diagnostic record, not as a score: a passed test means that particular check did not find a problem, not that every page is fast.
Rank #2
- Keep track of everything from attendance to test scores
- Spiral bound
- Measures 8-1/2" x 11"
Use PageSpeed Insights for a broad performance view
PageSpeed Insights combines two different kinds of evidence. Its laboratory test runs a controlled page load under a specified mobile or desktop simulation. Its Chrome User Experience Report (CrUX) section uses real-user field data for pages that have enough eligible visitors. These sections answer different questions and should not be blended into one number.
How to run a useful test
- Paste one representative public URL into PageSpeed Insights.
- Run the mobile view and record the reported metrics and opportunities.
- Run the desktop view separately; do not assume that a desktop result predicts mobile behavior.
- Read the lab section first when you need to reproduce a problem. Then inspect the field section to see how real visitors experience the page when data is available.
- Repeat the process for the other templates in your test set instead of assuming every URL shares the homepage’s bottleneck.
Lab data is useful for controlled troubleshooting: it gives you a consistent device profile and a reproducible list of blocking or oversized resources. Field data reflects visitors using different phones, networks and browsers over time, so it is better for deciding whether a problem affects production users. A page can have a respectable lab result and weak field data, or the reverse, without either report being wrong.
Use WebPageTest when you need a diagnosis
WebPageTest adds controls that are useful when a headline score is not enough. You can select a test location, browser and connection speed, then inspect the detailed request sequence.
What to examine
- Waterfall: Shows when DNS, connection setup, server response, downloads and browser-blocking resources occur.
- Optimization opportunities: Identifies request and delivery patterns that deserve investigation.
- Image analysis: Helps find oversized or inefficiently delivered images.
- Environmental report: Provides an estimate of the page-load environmental impact.
Choose a location close to the audience you care about and keep the browser and connection profile fixed for comparisons. The waterfall can separate a slow initial server response from a fast server that is followed by heavy JavaScript, late fonts or third-party embeds. That distinction determines whether you investigate PHP and hosting or the browser-side asset queue.
Read the metrics without confusing their jobs
| Metric | What it measures | How to use it |
|---|---|---|
| Largest Contentful Paint (LCP) | When the largest visible content element becomes available. | Use it to investigate the main headline, hero image, video poster or other dominant element and the resources that delay it. |
| Cumulative Layout Shift (CLS) | Unexpected movement of visible page elements while the page loads. | Look for images or ads without reserved dimensions, late fonts, injected banners and widgets that change their size. |
| Interaction to Next Paint (INP) | How promptly the page responds to user interactions. | Inspect long JavaScript tasks, event handlers, third-party scripts and overloaded main-thread work. |
| First Contentful Paint (FCP) | When the browser first displays any content. | Use it as an early rendering signal; a late FCP often points to connection, server, CSS or render-blocking work before useful pixels appear. |
| Time to First Byte (TTFB) | The time until the first byte of the response arrives from the server. | Investigate hosting, network distance, PHP execution, database work and cache misses when this is slow. |
LCP, CLS and INP describe the visitor’s loading, stability and interaction experience. FCP adds an earlier rendering marker, while TTFB provides server-response context. None of these metrics alone identifies the fix, and the tools do not publish a universal pass-rate statistic that can replace examining the page’s request sequence and environment.
Rank #3
- Used Book in Good Condition
Classify the bottleneck before changing anything
Once you have Site Health, PageSpeed Insights and (when needed) a WebPageTest waterfall, classify the dominant delay. This prevents the common mistake of installing several optimization plugins when the actual issue is a slow database query or a third-party widget.
| Observed pattern | Likely area | First investigation |
|---|---|---|
| High TTFB before the document starts | Server or back end | Check PHP configuration, cache-hit behavior, database work, hosting resources and the Site Health Server section. |
| Fast TTFB but late LCP | Front end | Inspect the hero image, CSS, fonts, preload choices and scripts that block rendering. |
| Large CLS with otherwise acceptable loading | Layout and asset sizing | Reserve image, ad and embed dimensions; investigate late font swaps and injected interface elements. |
| High INP after the page appears | JavaScript or third party | Find long main-thread tasks, heavy event handlers, carousels, popups and external scripts. |
| One template is slow while others are normal | Template, plugin or query specific to that page | Compare its request waterfall and active components with a fast template. |
| Lab results vary sharply between runs | Test conditions or variable third parties | Fix the location, device, browser and connection profile, then repeat multiple requests. |
Separate front-end, server and third-party causes
Front-end work
Large images, web fonts, carousels, popups, bloated themes and plugin assets add download, parsing and main-thread work. A page may receive its HTML quickly yet feel slow because the browser must process too much JavaScript or wait for a large visual asset before the dominant content is complete. Use the waterfall and asset details to identify which files are early blockers and which are merely late, nonessential requests.
Server-side work
Hosting capacity, network infrastructure, server hardware, database operations and PHP configuration influence TTFB. A cache miss, expensive query or slow PHP hook can delay every front-end resource because the document itself has not arrived. Site Health documents the environment, but host-controlled settings may require the provider’s assistance. Give the provider the exact URL, timestamp, TTFB observations and exported Site Health information rather than a vague statement that “WordPress is slow.”
Third-party work
Analytics, advertising, video players, social embeds, chat widgets, consent managers and externally hosted fonts can introduce DNS lookups, connection delays and JavaScript tasks outside your WordPress installation. Temporarily testing a page with one third-party service disabled can show whether it changes the waterfall or INP. Make that an isolated experiment, not a permanent production change, and preserve compliance or accessibility requirements.
Profile deeper when page tests cannot locate the cause
A page-speed report tells you that a page is slow; profiling can show which query, hook or server operation caused it. WordPress Core Performance guidance identifies several tools and data sources for this deeper stage.
- Query Monitor: Helps inspect database queries, hooks, PHP errors and component timing while diagnosing a page in a controlled environment.
- XHProf: Provides PHP profiling detail when you need to trace function-level server execution.
- Server-Timing: Exposes back-end timing information to compatible browser tools so server operations can be related to the page load.
- Web Vitals: Helps observe the loading, stability and responsiveness metrics in the browser context.
- Field data: Shows production behavior that a synthetic test cannot reproduce, especially for visitors on slower devices or networks.
Use profiling on a staging copy or during a controlled maintenance window when possible. Profiling itself can add overhead and should not be left broadly enabled for ordinary visitors.
Run a repeatable benchmark
Performance is variable. Even two requests for the same URL can produce slightly different results because of network conditions, machine load, cache state and third-party timing. A single impressive or disappointing run is not enough to declare a fix or regression.
- Keep the WordPress version, database contents, theme, active plugins and page content fixed while establishing a baseline.
- Choose one device, browser, location and connection profile for the comparison.
- Run several requests for each representative URL and record the metrics, test date and cache state.
- Change only one variable, such as an image format, a script setting or a server configuration.
- Repeat the same requests under the same setup.
- Compare the distribution of results and the waterfall, not just a single composite score.
- Restore the variable if the change harms another metric, a key interaction or a required feature.
Keep a small change log containing the URL, exact setting, test conditions, observed metric movement and any functional side effect. It prevents two optimizations from being credited for one improvement and makes regressions easier to reverse.
Free tools Windows power users keep installed
One-click scans. No signup required.
Apply fixes in the order the evidence supports
When images dominate
Start with the images identified in the LCP and waterfall evidence. Deliver dimensions appropriate to the displayed space, avoid forcing a mobile visitor to download a desktop-sized asset, and ensure the dominant image is not competing with unnecessary below-the-fold downloads. Recheck CLS after changing dimensions because an image that loads faster can still shift the layout if its space is not reserved.
When scripts and styles dominate
Identify which plugin or theme component enqueues each large file. Remove assets from templates that do not use them where it is safe to do so, defer noncritical work only when the feature remains functional, and test menus, forms, search, media players and checkout flows after every change. Minification or compression can reduce transfer size, but it will not solve a server-side delay or an event handler that blocks the main thread.
When fonts or layout shifts dominate
Check whether font loading changes text dimensions after the first paint. Reserve space for images, advertisements and embeds, and avoid inserting banners above already-rendered content. Use the CLS trace to identify the element that moved rather than applying a blanket layout setting.
When TTFB dominates
Confirm whether the slow response is a cache miss, a particular template, a database operation or the underlying server. Site Health can document PHP and server details; Query Monitor or PHP profiling can identify expensive WordPress work. If the bottleneck is outside your control, provide the host with repeatable measurements and ask for server-log or resource-usage evidence.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
When third parties dominate
Measure the page with each external service isolated. Delay, remove or replace a service only after checking its business, privacy, accessibility and legal purpose. A faster synthetic score is not an adequate reason to break consent handling, analytics requirements or a purchase flow.
Check Core Web Vitals in the right context
To check Core Web Vitals, use the LCP, CLS and INP values shown by PageSpeed Insights for the tested URL, then look for matching field data when the page has eligible CrUX coverage. Use the lab view to reproduce a loading or interaction problem and the field view to determine whether actual visitors encounter it. FCP and TTFB are supporting diagnostics: they help explain why LCP or the first visible content is late, but they are not substitutes for the three experience metrics.
Review more than one URL and template. A site-wide conclusion based on a single page can miss a product script, archive query or landing-page animation that affects a different audience.
When to involve a developer or hosting provider
Escalate when the evidence points beyond a safe dashboard change:
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →- Site Health reports persistent loopback, HTTP-request or update failures.
- TTFB remains high across several public templates with the same test conditions.
- A database query, PHP hook or plugin component is responsible for most server time.
- A required third-party script causes INP or blocks the page and cannot be configured from its own settings.
- A performance change breaks checkout, forms, navigation, media playback, accessibility or consent behavior.
Send the exact URL, test location, device profile, several run results, the relevant waterfall view and the exported Site Health information. This is actionable evidence; a single score without conditions is not.
A concise maintenance routine
After the initial investigation, keep performance checking part of normal site maintenance. Re-test representative templates after WordPress, theme, plugin, PHP, hosting or major content changes. Save a baseline before the change, repeat the same scenarios afterward, and investigate a regression before adding another optimization layer. This approach catches performance regressions while the responsible change is still identifiable.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
