How Page Speed Affects Conversion Rates: What the Research Shows

The Speed-Conversion Relationship Is Well Documented

The connection between page speed and business outcomes isn't speculation — it's one of the most studied relationships in web performance. Multiple independent studies from Google, Deloitte, Akamai, Portent, and others have quantified the impact. The consistent finding: slower pages lose visitors and revenue, and the relationship is nonlinear — the first few seconds matter far more than people expect.

This article compiles the major research findings in one place, so you can make data-informed decisions about whether performance investment makes sense for your specific situation. We'll cover the studies, what they found, what they didn't account for, and what it means in practice.

Google's Research: Bounce Probability by Load Time

Google's mobile page speed research, based on analysis of millions of mobile page loads, established the foundational data points that the industry still references. The key findings on how bounce probability increases as load time grows from 1 second:

• 1s to 3s: bounce probability increases by 32%
• 1s to 5s: bounce probability increases by 90%
• 1s to 6s: bounce probability increases by 106%
• 1s to 10s: bounce probability increases by 123%

The critical insight: the curve is steepest between 1 and 3 seconds. Going from a 1-second load to a 3-second load loses nearly a third of your visitors. Going from 3 to 5 seconds loses more, but the marginal impact per second decreases. This means optimizing a slow site from 5s to 3s has a bigger business impact than optimizing from 3s to 1s.

Google also reported that 53% of mobile site visits were abandoned if a page took longer than 3 seconds to load. Given that mobile devices account for roughly 60% of web traffic globally, this means a 3+ second load time is losing you more than half of your mobile visitors before they even see your content.

Google's Ad Revenue Data

In a separate study focused on ad-supported sites, Google found that pages loading within 5 seconds had significantly better engagement compared to slow-loading pages:

• 25% higher ad viewability — users scrolled far enough to see ads
• 70% longer average session duration — users stayed and browsed more
• 35% lower bounce rate — more users engaged with the content

For ad-supported sites and content publishers, these numbers translate directly to revenue. Longer sessions mean more page views, more page views mean more ad impressions, and higher viewability means higher CPMs. A publisher going from a 15-second load time to a 5-second load time can see meaningful revenue increases just from the improved engagement.

Deloitte's Speed Study: The 0.1-Second Effect

A Deloitte study conducted in partnership with Google examined the impact of even tiny speed improvements on business metrics for retail and travel sites. They found that a 0.1-second improvement in load time led to measurable improvements: retail consumers spent roughly 10% more, and lead generation sites saw increased page engagement.

What makes this study particularly compelling is the scale of improvement — 0.1 seconds. That's not the difference between a slow site and a fast site. It's a marginal improvement on an already-functioning site. The implication: every bit of performance optimization has a measurable return, even for sites that are already reasonably fast.

The study also found that the improvements were more pronounced on mobile than desktop, which makes sense given that mobile users typically have slower connections and less patience for delays.

Portent's Conversion Rate Analysis

Portent conducted an analysis of conversion rate data across multiple e-commerce sites and found a clear relationship between load time and conversions:

• Pages loading in 1 second had conversion rates roughly 3x higher than pages loading in 5 seconds
• The steepest decline occurred between 1 and 3 seconds
• The highest e-commerce conversion rates were found on pages with load times between 0-2 seconds
• Each additional second of load time reduced conversions substantially

The practical takeaway: if your e-commerce pages load in 4-5 seconds and you can bring that down to 1-2 seconds, the data suggests you could see a significant increase in conversion rate without changing anything else about your product, pricing, or marketing.

Akamai's E-Commerce Data

Akamai (one of the largest CDN providers) has published multiple studies on speed and e-commerce. Their analysis found that a 100-millisecond delay in load time could decrease conversion rates by a measurable percentage for online retailers. They also found that a 2-second delay in page load time increased bounce rates significantly.

Their recommendations, based on analysis of millions of e-commerce transactions: target a page load time of under 2 seconds for optimal conversion. They found that the relationship between speed and conversions was consistent across industries, device types, and geographies — faster sites consistently outperformed slower ones.

The SEO Dimension: Core Web Vitals as a Ranking Signal

Since mid-2021, Google has used Core Web Vitals (LCP, CLS, and INP) as ranking signals. While Google has said that content relevance is still the primary factor, page experience — including speed — acts as a tiebreaker between otherwise equal pages. For competitive search queries where many pages have similar content quality, performance becomes a meaningful ranking advantage.

This creates a compounding effect: a slow site faces a double penalty. Fewer visitors convert into customers (the direct speed-to-conversion impact), and fewer people find the site in the first place because it ranks lower in search results. For businesses that depend on organic search traffic, the combination can be severe.

Google Search Console provides a Core Web Vitals report showing how your real users experience your site. If your pages fail the Good threshold for any Core Web Vital, you'll see them flagged. Fixing these is one of the highest-ROI SEO activities because it improves both rankings and the conversion rate of the traffic you do get.

Mobile Performance Matters More Than Desktop

As of 2025, mobile devices account for approximately 60% of all web traffic globally. This share is higher in many industries — e-commerce, local services, and media skew even more heavily toward mobile. And mobile users are more sensitive to speed for three compounding reasons:

• Slower connections: many mobile users are on 3G or congested 4G, with higher latency and lower bandwidth than desktop users on WiFi or Ethernet
• Lower-powered devices: mid-range phones have 2-4x less CPU power than even a basic laptop, making JavaScript execution significantly slower
• Less patience: mobile sessions are often task-oriented (looking up a store, checking a price, finding information) with a lower tolerance for waiting

Google's mobile-first indexing means the mobile version of your site is what Google evaluates for rankings. If your desktop site scores 90 on PageSpeed but your mobile site scores 45, the 45 is what matters for SEO. The performance numbers from the studies above are measured primarily on mobile, where the impact is most severe.

What About B2B and Non-E-Commerce Sites?

Most of the major studies focus on e-commerce because conversions are easy to measure (add to cart, purchase). But the speed-engagement relationship applies to all site types:

• Lead generation: form submissions decrease with slower load times. The Deloitte study specifically included lead gen sites.
• SaaS: sign-up flows and onboarding pages need to be fast. Users evaluating SaaS products often compare multiple options — the one that loads fastest gets more attention.
• Content/media: engagement metrics (time on site, pages per session, scroll depth) all correlate with speed. Slower pages get less content consumption.
• B2B: while B2B buyers are more patient than B2C consumers, they're also comparing you to competitors. A slow site signals outdated technology and erodes trust.

Limitations of the Research

It's worth noting what these studies don't tell you. Correlation isn't always causation — faster sites may also have better design, clearer messaging, and stronger products. The studies control for this to varying degrees, but no study perfectly isolates speed as the only variable.

Additionally, the impact varies by industry, audience, and intent. A user searching for emergency information will wait longer than a user comparison-shopping. A loyal customer returning to a known site is more tolerant than a first-time visitor from an ad.

That said, the consistency of findings across independent studies, different methodologies, and different industries makes the overall conclusion robust: speed meaningfully affects user behavior and business outcomes. The exact magnitude varies, but the direction is clear and consistent.

What This Means In Practice

The research consistently points in one direction: speed improvements have measurable and often substantial business impact. But there's a practical ceiling to how much you can optimize within a given platform's constraints. If your site is on a no-code platform that ships 500KB of JavaScript and 400KB of CSS that you can't control, no amount of image optimization will get you to a 2-second load time.

The question isn't whether speed matters — the research is clear that it does. The question is whether your current platform allows you to reach the speed thresholds that matter for your business. If it does, optimize what you can and you're done. If it doesn't, the performance improvement from migration may pay for itself through improved conversion rates alone.