Talk to Jason

Your Website Is Slow. Here Is What That Is Costing You

Most answers to this question multiply your traffic by a statistic from someone else's e-commerce study. This one uses Google's published thresholds and then tells you to measure your own site.

By Jason Flatford Updated

The short answer

  • Most of the speed statistics you see are based on big e-commerce sites, and they do not apply to small brochure websites.
  • What you can check: Google considers a Largest Contentful Paint of 2.5 seconds or less as good, and anything above 4 seconds as poor, based on real user visits.
  • Google looks at the slowest 25 percent of visits and checks mobile separately. This means your site might seem fast on your laptop but still fail Google's tests.
  • The real costs are lower search rankings, visitors leaving before your site loads, and paying more per click if you run ads.
  • Most of the time, slow sites are caused by unoptimized images, not by hosting. Start by fixing your images.

First, what this page will not tell you

If you search this topic, you will see the same statistics everywhere: a one-second delay costs seven percent of conversions, mobile bounce rate goes up thirty-two percent between one and three seconds, and similar claims.

These numbers are real, but they come from big e-commerce studies and do not apply to a small twelve-page brochure site. If your business gets forty visitors a day and uses a contact form, those percentages do not fit your situation. If someone multiplies your traffic by seven percent and gives you a revenue number, they are just guessing.

Here is what you can actually check for your own site.

Google publishes the thresholds, and uses them

Page speed affects your search ranking. Google measures this using Core Web Vitals, which are based on real visits to your site, not just lab tests. Google also publishes the exact thresholds it uses.

Google's Core Web Vitals thresholds for Largest Contentful Paint Google rates Largest Contentful Paint as good at 2.5 seconds or under, needs improvement between 2.5 and 4 seconds, and poor above 4 seconds. These are Google's published thresholds, not estimates. Good Needs work Poor 0s 2.5s 4.0s 6.0s Largest Contentful Paint: how long until the biggest thing on screen has loaded. These bands are Google's own published thresholds, measured on real visits to your site.
Google's Core Web Vitals thresholds. A page must clear all three to pass.
Metric What it measures Good Poor
LCPTime until the biggest element has loaded2.5s or underOver 4.0s
INPHow fast the page responds when tapped200ms or underOver 500ms
CLSHow much the layout jumps while loading0.1 or underOver 0.25
Google grades a page on the slowest 25 percent of real visits, so a site that feels fine on your laptop can still land in the poor band.

There are two details that often surprise people. First, Google grades your site based on the slowest 25 percent of real visits, so even if your site feels fast on your office wifi, it might still fail. Second, Google checks mobile performance separately, which matters because most local searches happen on mobile, and most sites are slowest there.

What it actually costs you

Here are the three real costs, listed in order of importance for a small business.

Ranking. Speed is just one of many factors, and it will not help a page that people do not want. But when two results are similar, Google uses speed as a tiebreaker, and local search often has many similar results.

People leaving before your site loads. This is the most important cost and the hardest to track, because if someone leaves before seeing your site, they never show up in your inquiry count or analytics. You will not see them as a lost sale. They are just missing.

Ad spend, if you run ads. Google Ads weighs your landing page experience when it ranks and prices your ad, and Google says higher quality ads often pay less per click. A slow page works against you there. If you advertise, this is the easiest cost to measure.

You do not need to use industry averages to estimate your loss. Instead, you can measure your own site in about a minute, which is a better use of your time than reading more statistics.

What makes a small business site slow

Usually, slow sites are caused by one of these five things, listed in order of how common they are.

  • Unoptimized images. For example, a 4MB photo uploaded straight from a phone and only scaled down in the browser instead of the file. This is the most common cause and usually the easiest to fix.
  • Too much JavaScript. Things like page builders, tracking scripts, chat widgets, popups, and font loaders all run before your page is usable. Every plugin or extension adds more code.
  • Pages built on demand. Platforms like WordPress and Squarespace create pages from a database every time someone visits, then use caching to speed things up. A static file, on the other hand, is already finished and loads faster.
  • Web fonts. If you use three font families with four weights each, your site has to load twelve files before any text appears.
  • Cheap shared hosting. This can cause slow sites, but it is much less common than the other four reasons, even though people often blame hosting first.

Notice that four out of five causes are things on your page, not the server. Upgrading to faster hosting only fixes the least common problem.

What to do about it

  1. Measure your site speed. Use the Website Grader, which uses Google's PageSpeed data and gives you a report in about a minute. It is free and does not require an email.
  2. Start by fixing your images. This is usually the main issue and often only takes an afternoon to solve.
  3. Remove anything you are not using, like a chat widget nobody answers, an analytics tool you never check, or a popup plugin from an old campaign.
  4. Next, check your website platform. If your site is still slow after fixing images and removing unused scripts, the platform itself may be the problem. Deciding to change platforms is a bigger step, which is covered in the platform comparison.

Follow these steps in order. Many people start with the fourth step, which is the most expensive, but the real problem is usually with the second step.

Common questions

Does speed really affect my Google ranking?

Yes, but it is just one of many factors. Google uses Core Web Vitals from real visits and considers page experience in its rankings. Speed will not help a page that people do not want, but when two results are similar, speed can be the deciding factor. This is common in local search.

What counts as fast enough?

Google considers a Largest Contentful Paint of 2.5 seconds or less as good, and anything above 4 seconds as poor. It checks the slowest 25 percent of real visits and scores mobile separately, so your site might feel fast on your laptop but still fail Google's test.

How much money is a slow site costing me?

Honestly, no one can tell you that using industry averages. The common figures, like a 7 percent drop in conversions per second, are from big e-commerce studies and do not apply to a small brochure site with forty visitors a day. The best approach is to measure your own site and fix any issues it shows.

What is usually the cause?

Unoptimized images are by far the most common cause, like uploading a large photo straight from your phone and only scaling it down in the browser. After that, too much JavaScript from page builders and widgets can slow things down. Cheap hosting can also be a cause, but it is much less common, even though people often blame it first.

Sources

I opened each of these and checked it against the figure it supports, on the date shown.

  1. Aberdeen Group, The Performance of Web Applications: One-Second Wonders for Winning or Losing Customers (press release, 2008): the one-second delay and 7 percent conversion figure, from a survey of more than 160 organizations. Checked .
  2. Google, Milliseconds Make Millions: the probability of bounce rising 32 percent as mobile page load time goes from one to three seconds (Google/SOASTA research). Checked .
  3. Google Search Central, Understanding page experience in Google Search results: Core Web Vitals are used by Google's ranking systems, and page experience can contribute to success when lots of helpful content is available. Checked .
  4. web.dev, How the Core Web Vitals metrics thresholds were defined: LCP good at 2.5 seconds or under and poor over 4 seconds, INP good at 200ms or under and poor over 500ms, CLS good at 0.1 or under and poor over 0.25. Checked .
  5. Google Search Console Help, Core Web Vitals report: grading at the 75th percentile of real visits, mobile and desktop assessed separately, and a URL group taking the status of its worst metric. Checked .
  6. Google Ads Help, About Ad Rank: landing page experience counted in Ad Rank, and higher quality ads often paying less per click. Checked .

Measure it first, then decide

Run the grader, then bring the report to a fifteen-minute call. If the fix is an afternoon of image compression, that is what you will hear, and you can do it yourself.

Run the grader Book a 15-minute call