- Google and SOASTA (2017) found the chance of a visitor leaving rises 32% as load time goes from 1 to 3 seconds, and 123% from 1 to 10 seconds.
- Deloitte and Google (2020) measured a 0.1 second speed gain lifting retail conversions by 8.4%.
- You can turn a speed gain into money: monthly revenue times the conversion lift for the seconds saved.
- Mobile visitors leave sooner, and Google ranks pages partly on Largest Contentful Paint (good is under 2.5 seconds).
- Server response time sets the floor: front-end tuning cannot make up for a slow server.
How does page speed affect conversions?
Slower pages lose visitors before they buy, book or get in touch, and the effect has been measured many times. Google and SOASTA found that going from a 1 second to a 3 second load raises the chance of a visitor leaving by 32%. Deloitte and Google measured a 0.1 second gain lifting retail conversions by 8.4%.
Every extra second is another chance to give up. A visitor who leaves before your content appears never sees the product, the offer or the checkout. A slow page also looks less trustworthy, and people act on that. The loss repeats at each step of your sales path, so it adds up.
Deloitte and Google, Milliseconds Make Millions, 2020: a 0.1 second mobile speed gain raised retail conversions by 8.4%.
What does a one-second delay cost in revenue?
The earliest hard figures came from the biggest sites. Amazon reported that every 100 milliseconds of added delay cost sales. Google found that adding 500 milliseconds to search results cut traffic by 20%. Aberdeen Group put the cost of one second of delay at about 7% of conversions. Different studies, same direction.
| Source (dated) | Finding |
|---|---|
| Amazon (widely reported, 2006 to 2012) | Every 100ms of added delay reduced sales. |
| Google (Marissa Mayer, 2006) | Adding 500ms to search results cut traffic by 20%. |
| Aberdeen Group (2008) | A 1 second delay went with a 7% drop in conversions. |
| Deloitte and Google (2020) | A 0.1s mobile speed gain lifted retail conversions 8.4% and average order value 9.2%. |
Published sources linking speed to sales. Treat them as a direction; your own numbers will differ.
Read these as a direction, not a guarantee
These figures come from particular sites at particular times. What they agree on is the direction: faster earns more. How much it matters for you depends on your traffic, your margins and how slow you are now, so measure your own.
How much does bounce rate rise as load time grows?
Steeply, and faster as pages get slower. Google and SOASTA studied millions of mobile pages in 2017. As load time grew from 1 to 3 seconds, the chance a visitor left without doing anything (a bounce) rose 32%. At 5 seconds it was up 90%, and at 10 seconds up 123%. The first few seconds matter most.
| Load time (from 1s to) | Rise in bounce probability |
|---|---|
| 3 seconds | +32% |
| 5 seconds | +90% |
| 6 seconds | +106% |
| 10 seconds | +123% |
Google and SOASTA, Think with Google, 2017 (mobile load time against bounce probability).
If your pages take 2 to 4 seconds when the site is busy, a large share of your visitors sit on the steep part of this curve. You paid to bring them to the site, and they leave before it shows anything.
How do I turn a speed number into a revenue number?
You need four numbers: monthly revenue, current load time, the seconds you expect to save, and a conversion effect per second. The studies above suggest roughly 2% to 7% of conversions per second saved. Multiply that lift by your revenue for the monthly gain, then by 12 for the year.
An example. A shop takes £20,000 a month and loads in 4 seconds. Better hosting brings it to 2 seconds, saving 2 seconds. At a cautious 3% per second, that is a 6% lift: 0.06 x £20,000 = £1,200 a month, or £14,400 a year. Halve the per-second figure to be safer, and the yearly number is still large next to a hosting bill.
| Monthly revenue | Lift at 3% a second (2s saved = 6%) | Extra per month | Extra per year |
|---|---|---|---|
| £5,000 | 6% | £300 | £3,600 |
| £10,000 | 6% | £600 | £7,200 |
| £20,000 | 6% | £1,200 | £14,400 |
| £50,000 | 6% | £3,000 | £36,000 |
An illustration using a cautious 3% lift per second across 2 seconds. Use your own conversion data before acting on it.
Do not guess your load time. Measure it every minute, so you see what visitors get, including the slow moments a one-off test misses. Our uptime monitoring checks your site every minute and records how fast it answered, so you can compare before and after a change.
Why does mobile speed cost more conversions than desktop?
Because mobile visitors load pages more slowly and leave sooner. About six in ten web visits happen on phones, where networks are patchier and processors weaker. A page that loads in 2 seconds on a desktop can take 5 to 8 seconds on a mid-range phone with a weak 4G signal.
Think with Google (2017): 53% of mobile visits are abandoned when a page takes longer than 3 seconds to load.
There is a second cost. Google looks at the mobile version of your site first when ranking it, so a slow mobile site loses search visitors as well as sales. For WooCommerce and other online shops, the checkout is often the heaviest and slowest page, so the loss lands where the money is.
How do LCP and Core Web Vitals affect sales and rankings?
Core Web Vitals are three speed and stability scores Google measures on real visits, and they count towards rankings. The one that matters most for sales is Largest Contentful Paint (LCP): the moment your main content appears. Good is under 2.5 seconds; over 4 seconds is poor.
| Core Web Vital | Measures | Good | Poor |
|---|---|---|---|
| LCP (Largest Contentful Paint) | Time until the main content appears | under 2.5s | over 4.0s |
| INP (Interaction to Next Paint) | How quickly the page reacts to taps and clicks | under 200ms | over 500ms |
| CLS (Cumulative Layout Shift) | How much the page jumps while loading | under 0.1 | over 0.25 |
Google Core Web Vitals thresholds (web.dev).
LCP depends heavily on how fast your server sends the first byte of the page, so hosting is the base the score is built on. Our Core Web Vitals guide for WordPress explains each score and how to fix it.
Is better hosting the fastest way to speed up a site?
Often, yes, because server response time sets the floor under everything else. Compressing images, tuning caching and trimming plugins all help. But if the server is slow to answer, every visitor waits before any of that work starts. That first wait, the time to first byte, is the part front-end work cannot remove.
Why shared hosting has a slow floor
On cheap shared hosting, your site shares its PHP capacity with strangers. When their traffic spikes, your pages wait, so your speed rises and falls with other people's busy moments.
On G7Cloud WordPress hosting, each site runs in its own container with its own database, so its speed depends on its own traffic. We can move your site for you. We build a preview of your moved site first, you compare it with the live one, and nothing switches over until you approve. See how assisted migration works, or read managed versus shared hosting.
Frequently asked questions
Does website speed really affect conversion rate?
Yes, and it has been measured many times. Google and SOASTA (2017) found the chance of a visitor leaving rises 32% as load time moves from 1 to 3 seconds. Deloitte and Google (2020) measured a 0.1 second gain lifting retail conversions 8.4%.
How fast should my website load for good conversions?
Aim for your main content to appear in under 2.5 seconds, Google's threshold for a good Largest Contentful Paint. Over 4 seconds is rated poor. This matters most on mobile, where most visits happen and phones and networks are slower.
How do I calculate the revenue impact of a slow website?
Take your monthly revenue and the seconds you can save, then apply a conversion lift of roughly 2% to 7% per second from the published studies. Saving 2 seconds at a cautious 3% per second is a 6% lift, or £1,200 a month on £20,000 of revenue. Refine it with your own conversion data.
Is upgrading hosting the fastest way to speed up my site?
Often, yes. Server response time is the floor under every front-end fix, so if your host is slow, caching and image work cannot fully make up for it. Hosting where your site has its own container and database removes the slowdowns that come from sharing a server.
Does site speed affect Google rankings as well as conversions?
Yes. Core Web Vitals, including Largest Contentful Paint, count towards Google rankings, and Google looks at your mobile site first. A slow site loses sales from the visitors who arrive, and fewer visitors arrive.
About G7Cloud Engineering
Articles written by the engineers who build and run G7Cloud: UK managed hosting and the AI Website Builder. We write about what we operate every day: containers, backups, databases, and the small-business websites that run on them.
More about G7Cloud →