How Page Speed Affects Conversion Rates (The Data)

Slow pages do not just rank lower — they convert worse. Here is what the research says about the relationship between load time and conversions, and what to focus on.

The relationship between speed and conversions

The data on page speed and conversions is unusually consistent. Across studies from Google, Deloitte, Portent, and individual companies, the finding is the same: faster pages convert better.

The relationship is not linear. Improvements matter most at the slow end of the scale. Going from 8 seconds to 4 seconds has a larger impact than going from 2 seconds to 1 second. But improvements at every level correlate with measurable conversion gains.

What the research shows

Google / Deloitte research (retail and travel):

A 0.1 second improvement in mobile site speed improved conversion rates by 8.4% for retail sites and 10.1% for travel sites.

The improvement was most pronounced on product pages and checkout flows.

Google's own data (mobile):

53% of mobile visits are abandoned if a page takes longer than 3 seconds to load.

The probability of bounce increases 32% as page load goes from 1 to 3 seconds.

At 5 seconds, the bounce probability increase is 90%.

Portent study (B2B and B2C):

The highest-converting sites in their dataset loaded in under 1 second.

For every additional second of load time, conversion rates dropped by an average of 4.42% in the first five seconds.

Walmart:

Every 1-second improvement in page load time resulted in a 2% increase in conversions.

Why speed affects conversions

The mechanism is straightforward:

Patience thresholds. Users have a fixed tolerance for waiting. On mobile, that threshold is lower than on desktop and lower than it was five years ago. Users who leave before your page loads cannot convert.

Perceived trust. A slow, janky page signals low quality to visitors. People are less likely to hand over payment details to a site that does not feel polished.

Form abandonment. Checkout and sign-up flows with slow field responses or page transitions have higher abandonment rates, even when users initially engage.

SEO → traffic → conversions. Speed affects rankings (via Core Web Vitals), which affects organic traffic volume, which affects total conversions even if the on-page conversion rate is unchanged.

Where to focus for conversion impact

Not all pages contribute equally. Focus speed improvements in this order:

1. Checkout and payment pages

These are where money changes hands. Any friction — slow field validation, delayed button responses, layout shifts near the payment button — directly reduces revenue.

Measure INP on checkout pages specifically. A high INP score means users experience lag between clicking a button and seeing a response.

2. Landing pages from paid ads

Paid traffic is expensive. If your Google Ads or Meta Ads are sending users to a page that takes 5 seconds to load on mobile, you are paying for visits that leave before seeing your offer.

3. Product pages (ecommerce)

The product page is the primary conversion point in ecommerce. Image load speed and LCP on the product hero image directly affect add-to-cart rates.

4. Homepage

Your homepage is often the first impression for direct and branded search traffic. A fast homepage sets the tone for the rest of the visit.

How much is a slow site costing you?

A rough calculation: if your site receives 10,000 monthly visitors with a 2% conversion rate at £50 average order value, that is £10,000/month.

If a speed improvement increases conversions by 10% (within the range reported in the Deloitte study), that is an additional £1,000/month from the same traffic.

The return on investment for performance work is often faster than for most other conversion rate optimisation activities because it benefits every page simultaneously.

Start with a baseline measurement

Run a free speed check with Loadzen on your highest-traffic landing pages and checkout flow. See your current PageSpeed scores and which specific issues are most likely affecting your conversion rate.