How to Test Landing Page Speed (and What to Do With the Results)
Most founders who test their page speed test the wrong page. They run their homepage through PageSpeed Insights, see a score, and assume the number applies to their ad landing pages. It does not. Each URL on your site has its own speed profile, and the page your paid traffic lands on may perform completely differently from the page you tested. This article shows you how to test the right page with three tools, and what to do with what you find.
The Most Common Mistake: Testing the Wrong Page
Your homepage is often the fastest page on your site. It is the one you care about most visually, the one most likely to have been optimised at launch, and the one that may have no third-party scripts beyond basic analytics. Your ad landing pages are a different story. They may have A/B testing scripts, retargeting pixels, chat widgets, embedded video, and form integrations that your homepage does not carry.
When you test your homepage and see a score of 90, nothing about that score tells you what a visitor experiences when they land on yourdomain.com/lp/offer-name. That page may score 45 on the same tool. The gap between those two numbers is the gap between the page you tested and the page your ad dollars are driving traffic to.
The rule: always test the exact URL your ad campaigns send traffic to. If you have multiple landing page variants, test each one separately. Speed profiles are per-URL, not per-domain.
Tool 1: PageSpeed Insights - Start With LCP on Mobile
PageSpeed Insights is the right tool to start with because it combines lab data (a simulated load) with field data (real Chrome user measurements from your actual visitors). Go to pagespeed.web.dev, paste your landing page URL, and run the analysis.
When the results load, look at the mobile tab first. Mobile scores matter more for paid traffic because the majority of ad clicks happen on mobile devices, and mobile connections are slower than desktop connections. The score shown at the top is a composite - do not use it as your primary metric.
Instead, scroll to the Core Web Vitals section and look at Largest Contentful Paint (LCP) first. A passing score is under 2.5 seconds. If the LCP value is shown in orange or red, that is your first priority. The other metrics matter but LCP has the most direct relationship with whether paid visitors stay or leave before your page is usable.
Tool 2: WebPageTest - Use the Waterfall to Find Blockers
PageSpeed Insights tells you what is failing. WebPageTest tells you why. Go to webpagetest.org, enter your landing page URL, select a mobile device and a mid-tier connection (3G Fast or 4G are representative of real paid traffic conditions), and run the test.
The waterfall chart is the most useful output. Each row is one request your page makes - your HTML, CSS, JavaScript files, images, fonts, and third-party scripts. The horizontal bars show when each request starts and how long it takes. Look for three patterns: a very wide bar near the top (large file downloading early), a gap where nothing is loading (server waiting), or a cluster of external-domain requests that appear before your main content.
Each of those patterns points to a specific fix. A wide image bar points to compression. An early gap points to TTFB and hosting. A cluster of external requests before your content points to render-blocking scripts. The waterfall does not tell you the fix directly, but it shows you exactly where the time is going.
Tool 3: Nebula Audit - Structural CWV Analysis
PageSpeed Insights and WebPageTest measure load performance at the technical level. Nebula's audit approaches the same question from the structural side: does your landing page have the conditions that typically cause Core Web Vitals to fail? This includes checking for unoptimised images, synchronous third-party scripts, missing font display settings, and hosting configurations that drive high TTFB.
The structural approach is useful because it catches problems before they appear in field data. Field data in PageSpeed Insights requires real user visits over a 28-day window - if your landing page is new or running on limited traffic, there may not be enough data yet to get reliable measurements. A structural audit identifies the same patterns without requiring traffic history.
The Load Speed signals in the Nebula audit map directly to the three killers covered in the LCP article: image compression, render-blocking scripts, and TTFB. Each signal has a pass or fail result and a specific recommended action.
What to Do With the Results
After running all three tools, you will have a list of issues ranked by priority. Work through them in this order: LCP first, then First Input Delay or Interaction to Next Paint, then Cumulative Layout Shift. Within LCP, fix image compression before script loading before TTFB - image compression is the cheapest fix with the largest typical impact.
After making each fix, re-run PageSpeed Insights on the specific URL you changed. Do not assume a fix worked without measuring it. The lab data in PageSpeed Insights reflects the current state of the page and updates immediately - you do not need to wait for field data to confirm a technical change.
A landing page with a passing LCP on mobile is not the ceiling - it is the floor. Once LCP passes, look at the remaining Opportunities in PageSpeed Insights and work through them by estimated savings. Each improvement compounds with the others, and the difference between a score of 60 and a score of 90 is often a handful of discrete changes that each take less than an hour.
Get a Structural Speed Audit in Two Minutes
Nebula's free audit runs a structural analysis of your landing page's Core Web Vital conditions without requiring 28 days of traffic data. You get a prioritised list of load speed failures and specific fixes for each one.
Related Articles
Frequently Asked Questions
How do I test my landing page speed?
Use Google PageSpeed Insights (pagespeed.web.dev) with the actual URL of the landing page receiving paid traffic, not the homepage. Look at the LCP score first on mobile. A score under 2.5 seconds is passing. The tool also identifies the specific resources causing the slowdown.
Should I test my landing page on mobile or desktop?
Mobile first. Over 60% of paid ad traffic lands on mobile devices, and mobile scores are consistently lower than desktop. A page that passes on desktop can fail on mobile by a significant margin. Google's ranking signals also use the mobile version of the page as the primary index.
What is the difference between PageSpeed Insights and the Nebula audit for speed?
PageSpeed Insights measures real load performance by actually loading the page. The Nebula audit checks structural signals in the page HTML, such as render-blocking scripts, image compression, and font loading strategy. Both are useful: PageSpeed gives the measured outcome, Nebula identifies the structural causes.