What Is a Good Landing Page Load Time? Core Web Vitals Targets

A good landing page load time is under 2.5 seconds to Largest Contentful Paint on a mid-range mobile phone, which is the threshold Google classifies as good in Core Web Vitals. Under 1.5 seconds is where fast pages sit. Past 4 seconds you are losing visitors who never saw your copy at all. Measure on throttled mobile rather than your office connection, because that is where most landing page traffic actually arrives and where the gap between assumption and reality is largest. Last updated July 2026.

Speed gets discussed as an SEO problem, which undersells it. On a landing page, a slow load is a conversion problem first: the visitor clicked an ad you paid for, waited, and left before the page rendered. You paid for that click either way.

What is a good landing page load time?

Use Core Web Vitals thresholds rather than a single stopwatch number, because "load time" means several different things and the one that matters is when the page becomes useful to a human.

LCP is the one to lead with on a landing page: it marks when the biggest visible element, usually your hero image or headline, has actually painted. That is the moment the visitor can start reading. Google's threshold is that 75 percent of your real-user visits should hit the good bucket, not your median.

How does page speed affect conversion rate?

The mechanism is simple even though the size of the effect varies wildly by site. Some share of visitors abandon while waiting, and that share grows with every additional second. Those people never see your headline, your offer, or your form, so no amount of copy work reaches them. Speed does not persuade anyone; it just determines how many people get the chance to be persuaded.

Be skeptical of the widely circulated "one second costs 7 percent of conversions" figure. It comes from a specific ecommerce study and gets repeated as a universal law. The honest version: the effect is real, it is larger on mobile than desktop, larger on cold paid traffic than on branded traffic, and larger the slower you already are. Going from 6 seconds to 3 matters far more than going from 2.0 to 1.7.

Why is my landing page slow?

On a landing page specifically, the causes are predictable and rarely exotic.

Unoptimized hero images. A 2 MB PNG behind the headline is the single most common cause of a bad LCP. Serve WebP or AVIF, size the image to the largest slot it actually occupies, and preload it rather than lazy-loading the one image that defines your LCP.

Third-party tags. Landing pages accumulate them faster than any other page: analytics, two ad pixels, a heatmap tool, a chat widget, a consent banner, a form embed. Each is a separate connection and some block rendering. Audit what is on the page against what anyone actually reads.

Render-blocking CSS and fonts. A full site stylesheet loaded for a page that uses a fraction of it, plus custom fonts that hide text until they download. Use font-display swap, subset the fonts, and inline the critical CSS.

Client-side rendering for static content. If your landing page is a JavaScript app that fetches its own copy after load, the visitor waits for the framework, then the request, then the paint. Static or server-rendered HTML wins every time on a page whose content does not change per visitor.

Slow server or no CDN. A high TTFB means the page had not even begun before any of the above mattered. Check TTFB first, because optimizing images will not help if the server took 1.6 seconds to answer.

How do I test my landing page speed?

Use two kinds of measurement and do not confuse them. Lab data, from PageSpeed Insights or Lighthouse, runs a simulated load and is good for diagnosis: it tells you what is slow and why, reproducibly. Field data, from the Chrome User Experience Report or your own real-user monitoring, tells you what your actual visitors experienced across devices and networks. Lab data finds the cause; field data tells you whether it matters.

Always throttle to a mid-tier mobile device and a slow 4G connection when testing. Your unthrottled desktop result is close to meaningless for a page whose traffic comes from phones on cell networks.

Speed is also not a one-time fix. Pages regress when someone adds a tag, swaps an image, or ships a plugin update, and nobody notices until a campaign underperforms. Pairing a monthly audit with continuous monitoring that checks the page every 30 seconds catches both the outage and the slow creep, which is the version of this problem that actually costs money because it never announces itself.

What is the fastest way to speed up a landing page?

In rough order of effort to payoff:

Compress and correctly format the hero image, and preload it. This alone fixes most bad LCP scores and takes minutes.

Remove tags nobody uses. Every landing page has at least one pixel from a channel that has been off for six months.

Defer non-critical JavaScript. Chat widgets, review embeds, and video players almost never need to load before the visitor can read the page.

Set explicit width and height on images and reserve space for banners and embeds. This fixes CLS, which is a real conversion problem when a button moves the instant someone taps it.

Put a CDN in front of the page and cache the HTML if it is not personalized. This is the fix for TTFB and it requires no code changes.

Is speed more important than copy?

No, and framing it as a contest leads teams astray. Speed sets the ceiling on how many people can read your page. Copy determines how many of those people act. A page that loads in 900 milliseconds and says nothing convincing converts badly. A brilliant page that takes 7 seconds converts badly too, for a completely different reason. Fix whichever one is currently the binding constraint.

The practical test: if your bounce rate is high but time on page for non-bouncers is healthy, you likely have a speed or message-match problem at the entry. If visitors are arriving, reading, and leaving, the page is fast enough and the argument is what needs work. Our guide to reading landing page bounce rate covers how to tell those apart, and why a landing page is not converting walks the rest of the diagnosis.

Auditing speed alongside everything else

Speed rarely travels alone. The page that loads slowly is usually also the page with a vague headline and three competing CTAs, because both come from the same absence of ownership. Converto audits a live URL across copy, layout, CTAs, page speed, and AI-readability in one pass, so you see the load-time problem ranked next to the message problem instead of solving them in separate tickets. It reads the page you already shipped, with nothing to install, and generates the copy and CTA variants alongside the speed findings. Suggestions are data-informed, not guarantees.

For the dedicated speed view, see page speed audit, and for the full pass on a live page, landing page audit.