The cost of a slow website rarely shows up as a line item on a P&L statement, which is exactly why it's so easy to ignore. It shows up instead as visitors who leave before the page finishes loading, carts that get abandoned, and search rankings that quietly slip. This post walks through what the data actually says about load time and revenue, so you can put a real number on the problem instead of a vague feeling that "the site feels slow."
How much does a slow website actually cost?
The short answer: more than most business owners assume, and the effect compounds. Even fractions of a second show up in the numbers: Deloitte's "Milliseconds Make Millions" study, which tracked 30 million mobile sessions across 37 brands, found a 0.1-second speed improvement lifted retail conversions by 8.4%. The same mechanism works in reverse — every bit of speed a site loses quietly erodes conversions.
In Portent's 2022 analysis of more than 100 million page views across 20 sites, B2B lead-generation pages that loaded in one second converted three times as well as five-second pages and five times as well as ten-second pages. On B2C online stores, one-second pages converted 2.5 times as well as five-second pages — about 3.05% at one second, falling to 0.67% by four seconds.
That single paragraph is worth sitting with, because it reframes site speed from a technical nice-to-have into a direct revenue lever. If your product pages load in four or five seconds instead of one, you're not just delivering a worse experience — you're very likely converting at less than half the rate you could be.
Why does load time affect revenue so directly?
Three mechanisms are at work, and they compound rather than operate independently.
- Abandonment before the page even renders. In a 2009 Forrester survey commissioned by Akamai, 47% of online shoppers expected a page to load in two seconds or less, and 40% said they'd wait no more than three seconds before leaving. Those visitors never even see your offer — and expectations have only risen since.
- Reduced trust in what does load. A page that stutters or shifts while loading reads as unpolished, and unpolished reads as untrustworthy — especially for first-time visitors evaluating whether to hand over payment details.
- A drag on every subsequent interaction. Slow doesn't stay contained to the homepage. If your server and asset pipeline are slow on page one, they're slow on the checkout page too, and checkout is where the cost of friction is highest.
What does 100 milliseconds actually mean for revenue?
It sounds too small to matter, but it has been measured. Deloitte found that making mobile pages 0.1 seconds faster raised retail conversions by 8.4% and average order value by 9.2%. For a store doing $50,000 a month, a lift of that size is worth several thousand dollars a month — not a rounding error. (You'll also see the claim that every 100ms cost Amazon 1% of sales; it comes from internal tests a former Amazon engineer described in 2006 and was never published as a formal study, so treat it as an anecdote rather than a benchmark.)
How does bounce rate change with load time?
Bounce rate is the other half of the story, because a visitor who bounces never gets the chance to convert at all. Google's 2018 mobile benchmarks found that as page load time goes from one second to ten, the probability of a mobile visitor bouncing rises by 123%.
Here's what Portent's 2022 data looks like side by side. Note the two columns measure different things: B2B "goal" conversions (a form fill or download) run far higher than store checkouts.
| Load time | B2B lead-gen conversion rate | B2C e-commerce conversion rate |
|---|---|---|
| 1 second | ~40% | ~3.05% |
| 3 seconds | ~29% | — |
| 4 seconds | — | ~0.67% |
| 5 seconds | About ⅓ of the 1-second rate | About 1 ÷ 2.5 of the 1-second rate |
| 10 seconds | About ⅕ of the 1-second rate | — |
Source: Portent, 2022. These are averages across the sites Portent studied, not a prediction for any specific site — your actual numbers depend on traffic source, device mix, and what you're selling. But the direction and rough magnitude are well established across multiple independent studies, which is why site speed keeps showing up as a priority in conversion-rate-optimization work rather than being treated as a purely technical concern.
How do you estimate the revenue loss for your own site?
Putting a real number on your own situation takes three inputs: your current load time, your typical monthly traffic to the pages that matter (product pages, landing pages, checkout), and your current conversion rate or average order value. From there, you can apply the conversion-by-load-time pattern above as a directional estimate rather than a guarantee — the point isn't precision to the dollar, it's understanding the order of magnitude you're dealing with.
For example, a store with 20,000 monthly visitors to product pages and a 2.5% baseline conversion rate is working with roughly 500 conversions a month at its current speed. If that site is loading in four seconds instead of one or two, the conversion research above suggests it could plausibly be converting at less than half of what a faster version of the same page would achieve — which, at even a modest average order value, adds up to a meaningful monthly gap. This is exactly the kind of calculation an automated audit can do for you automatically, using your site's actual measured load time rather than an assumed one.
Is this a one-time fix or an ongoing concern?
Ongoing, in most cases. Sites tend to get slower over time as more plugins, scripts, images, and third-party trackers get added without anyone removing what's no longer needed — a phenomenon sometimes called "performance creep." A site that was fast at launch can quietly slide back into the same revenue-losing territory eighteen months later. Treating speed as a metric to check quarterly, the same way you'd check analytics or search rankings, is a more durable approach than treating it as a single project you finish once.
What should you actually do about it?
Start by measuring, not guessing. Most sites that "feel slow" have a specific, fixable cause — often unoptimized images, render-blocking scripts, or hosting that can't keep up with traffic. Once you know your actual Largest Contentful Paint and load time, you can translate that directly into an estimated revenue-loss range using the conversion research above, which turns an abstract performance score into a number your team (or your client) will actually act on.
If this is on your list, see our guides on the most common reasons a website is slow and what Core Web Vitals measure.
