Why a slow site loses customers (and what it costs monthly)

A beautiful site that takes 8 seconds to load is not a beautiful site. It’s a house with the front door jammed.

The good news is that speed is not a matter of taste. It’s a set of public numbers, measured by Google, that you can check yourself in two minutes. The bad news is that on most company sites we audit, nobody had ever checked them.

What Google actually measures

The main metric is LCP (Largest Contentful Paint): how long it takes for a visitor to actually see the main content of the page, not how long it takes everything to finish loading.

The thresholds are published by Google on web.dev: a good LCP is 2.5 seconds or less, and a poor one is over 4 seconds. Anything in between needs improvement.

One detail most people miss: Google recommends looking at the 75th percentile of page loads, split between mobile and desktop. It does not matter how fast the site loads for you, on fibre, with the page already cached. What matters is how fast it loads for the third visitor out of four, on an ordinary phone. That number is almost always far worse than the one the site owner looks at.

What speed is worth, in measured money

This is where the conversation gets concrete. Vodafone ran an A/B test on a landing page: the same page in two versions, one optimized for Web Vitals and one not, half the traffic to each. The results, published on web.dev in March 2021:

• 31% improvement in LCP
• 8% more total sales
• 15% uplift in the lead-to-visit rate
• 11% uplift in the cart-to-visit rate

That is not an agency estimate and not a correlation spotted from the outside. It is a controlled test where the only variable changed was speed. A third faster meant almost a tenth more sales.

Apply the proportion to your own traffic. If your site brings in 30 quote requests a month, 8% is two more per month, 24 a year, for work you do once.

Don’t know what your LCP is? Ask for a free audit: we send you your site’s real numbers and the list of what to fix, in order of impact. Or talk it through in a 30-minute call.

What you lose, concretely

Here is what we see on almost every site we audit, as our own observation rather than a market statistic: out of 10 people who land, roughly 9 leave without a signal, meaning no call, no form, no message. Rarely because the offer is weak. Usually because the friction is too high: the page drags, the main image shows up late, the contact button hides three screens down.

At 100 visits a month, that is dozens of potential customers who end up with the competitor whose page loaded faster. They never tell you they left. They simply never show up in your numbers as quote requests.

The 3 numbers worth knowing

• LCP, when the page becomes visible. Target: under 2.5s. Over 4s puts you in the poor band, by Google’s own thresholds.
• Speed Index, how quickly the page “feels” complete.
• Mobile PageSpeed score, the overall grade on a phone. Below 75 you are leaving money on the table.

All of them are free, from Google, in PageSpeed Insights. They are not agency opinions and you do not have to take anyone’s word for them.

Why it happens, almost every time

The causes repeat, and they are the same ones Romanian hosting companies list:

1. Unoptimized images. A 4 MB photo straight off a phone, displayed at 600 pixels wide. The whole thing transfers, a quarter of it shows.
2. Unused JavaScript. The site ships an entire bundle on every visit when it needs a fraction of it. The result: the visitor’s phone spends several seconds on a half-empty page with a busy processor.
3. Too many modules. Every plugin installed “so we have that too” adds its own files. Ten plugins for one feature each is ten tolls paid on every load.
4. No caching. Without a cache, the server rebuilds the same page from scratch for every visitor, including everyone reading the same article.
5. Hosting. The cheapest shared plan, with hundreds of sites on one server, is the saving you pay back in seconds of waiting.

The order is not accidental. The first two explain most of the cases we see, and they are also the cheapest to fix.

What to do in the next 10 minutes

1. Put your address into PageSpeed Insights, on mobile, not desktop.
2. Write down your LCP and your overall score.
3. If LCP is over 4 seconds or the score is under 75, you have a customer leak you can stop.
4. Look at the “Opportunities” section: the first two or three entries tell you, in order of impact, what is slowing you down.

You do not need a developer for any of those steps. You need one at step 4, and only for what shows up there.

What we fix first

The order matters more than the list. When we take over a slow site, we go by impact, not by what is easiest to tick off:

1. The main image on the homepage, almost always the element that dictates LCP. Correct dimensions, modern format, priority loading.
2. Render-blocking code, moved or removed, so the page appears before everything finishes.
3. Caching and delivery, so the second visitor does not pay the same wait as the first.
4. Measuring again, on mobile, so there is a number before and a number after, not a feeling.

Speed is not a technical vanity metric. It is the difference between a visitor who calls you and one who closes the tab.

Frequently asked questions

How fast should a website load? Under 2.5 seconds for LCP, measured on mobile at the 75th percentile of visitors, according to the thresholds Google publishes on web.dev. Over 4 seconds the site officially falls into the poor band.

Does site speed affect Google rankings? Yes, page experience is a ranking factor, though not the main one. The bigger effect is different: a slow site loses visitors before they contact you, no matter what position it holds in the results.

How do I check my site speed for free? In Google’s PageSpeed Insights, at pagespeed.web.dev. Enter the address, pick the mobile report, and you get your LCP, the overall score, and the list of opportunities in order of impact.

Why is my WordPress site slow? Most often unoptimized images, too many plugins, no caching layer, a heavy theme, or an overcrowded shared host. The PageSpeed Insights report shows which of those weighs most in your case.

How much does it cost to make a site faster? It depends on the cause, and that only becomes clear once you look at the report. The usual fixes (images, render-blocking code, caching) are hours of work, not weeks. A rebuild is only necessary when the site sits on a foundation you cannot repair, and then the ranges are the ones in how much a website costs.

Is it worth switching hosting to gain speed? Only if the report shows a high server response time. If the problem is images and code, you move the same problems onto a more expensive server and keep the same seconds of waiting.

Want to see exactly what is slowing your site down? Ask for a free audit and we will send you the real numbers plus what to fix in order of impact, or book a 30-minute call.