Check your website on your phone: what we found on 637 Georgia sites
You can check what matters most about your website with the phone in your pocket. This post walks through what to look for and why, using what we found when we tested 637 Georgia business sites the same way.
Pick up your phone, turn off Wi-Fi, and open your own website. Count to five. If the page is still blank or half built, or you can’t tell what the business does and how to call it, a stranger looking you up can’t either. This fall we ran 637 Georgia business websites through a stricter version of that check, and more than half of them took longer than 5 seconds to show their main content. The check below costs nothing, and you can do it from wherever you’re sitting.
Check your own site
Do this on your own phone, the way a new visitor would see it:
- Open a private or incognito tab, so your phone doesn’t reuse parts of your site it saved on earlier visits. Turn off Wi-Fi so you’re on your cellular connection.
- Type in your web address and count to five. At five, look at the screen as a stranger would. Can you tell what the business does, where it is, and how to call?
- Find your phone number without scrolling, then tap it. It should offer to dial. If you can’t see it, or it doesn’t dial, a caller has to work to reach you.
- Paste your web address into PageSpeed Insights, Google’s free test, and open the mobile results. Look for Largest Contentful Paint, which is how long the main content takes to show. If your site gets enough visitors, the page also shows how real Chrome users’ visits went, and that number deserves more weight than any lab test, ours included.
Two more things live in your site’s code. One is local business markup, a short block of code that gives Google your business’s name, address, phone, and hours in a format it reads directly. Google’s local business guide says you can use it to tell Google details like your hours, and Google’s Rich Results Test will show whether your homepage has it. The other is a way to count visitors. If nobody at your business can log in and see how many people visited last month, ask whoever built your site whether Google Analytics or another counting tool is installed.
Why the first five seconds matter
Until the main content shows, a visitor on a phone is looking at a page that hasn’t finished arriving. I won’t quote a figure for how many people give up, because the ones that get passed around are hard to trace to a study I can check. Count to six, about the median for the law firm sites we tested, and decide how long you’d wait.
Google gives a yardstick for real visits. For this measure, 2.5 seconds or less counts as good, and more than 4 seconds counts as poor (web.dev on Largest Contentful Paint). Speed counts in search too, though it isn’t everything. Google says its ranking systems use Core Web Vitals, a short list of measures of how fast and stable a page feels, and this is one of them. It also says it will still show the most relevant page when the page experience is poor (Google on page experience), so a fast page that doesn’t answer the search won’t win on speed alone.
The phone number matters for a plainer reason. Someone ready to call wants to see your number on the first screen, the part of the page they see before scrolling, and tap it. I’d treat every extra step as a chance for them to call someone else.
What we found on 637 Georgia sites
Between September 28 and October 16, 2026, we tested 309 Georgia law firm homepages and 328 Atlanta-area home service homepages. We used the same phone, connection, and processor settings that Google’s Lighthouse uses for its mobile test: a mid-range Android phone on a slow 4G connection, with an empty cache. Lighthouse is the test behind the lab results in PageSpeed Insights, and our numbers come out close to Google’s but not identical. “Main content shows” means the moment the biggest thing on the first screen appears. We left out sites that blocked the test or failed to load, and we don’t name any business we tested.
These aren’t a random sample of Georgia businesses. They’re businesses we found while looking for clients: Georgia law firms, mostly personal injury firms, and home service companies around Atlanta that have websites. It’s also a lab test, so it doesn’t measure real visitors’ phones.
| Law firms | Home services | |
|---|---|---|
| Sites tested | 309 | 328 |
| Tested | Sept 28 to Oct 16 | Sept 28 to Oct 13 |
| Median seconds until the main content shows | 5.9 | 7.3 |
| Took over 5 seconds | 56% | 66% |
| Took over 10 seconds | 31% | 38% |
| No tap-to-call link on the first screen | 40% | 25% |
| No phone number on the first screen | 57% | 52% |
| No local business markup for Google | 52% | 45% |
| No Google Analytics or Tag Manager | 32% | 19% |
| Median page weight | 1.6 MB | 2.2 MB |
The home service sites were slower and heavier. Their median homepage took 7.3 seconds to show its main content, against 5.9 for the law firms, and weighed 2.2 MB against 1.6. The law firm sites did worse on calling and counting. Of the law firm sites, 40% had no tap-to-call link on the first screen, against 25% of the home service sites, and 32% had no Google Analytics or Tag Manager, against 19%.
Google’s yardstick is for real visits, so our lab numbers can’t tell you whether any one site passes it. For a rough comparison, though: in our test, 85% of the law firm sites and 93% of the home service sites took longer than 2.5 seconds, and 66% and 80% took longer than 4.
There was good news, too. Nearly every site fit a phone screen. Only 2% of the law firm sites scrolled sideways, and among the home service sites, that share rounded to zero. And more than half of the sites with no tap-to-call link on the first screen had one somewhere else on the homepage. Only 18% of the law firm sites and 5% of the home service sites had none at all, so for a lot of owners the fix is moving a link they already have.
Home service sites by trade
| Trade | Sites | Median seconds | Over 5 seconds | Over 10 seconds | No tap-to-call on the first screen |
|---|---|---|---|---|---|
| HVAC | 93 | 5.7 | 58% | 27% | 22% |
| Roofing | 89 | 7.8 | 70% | 39% | 34% |
| Landscaping | 53 | 7.1 | 66% | 40% | 25% |
| Tree service | 52 | 8.4 | 69% | 46% | 13% |
HVAC sites were the fastest of the four, with a median of 5.7 seconds. Tree service sites were the slowest at 8.4, yet they were the most likely to put a tap-to-call link on the first screen: only 13% didn’t. Roofing sites were the least likely, with 34% missing one. The rest of the home service sites, 38 in a mixed group of other trades and 3 remodeling companies, aren’t broken out.
By city
We only report cities with 10 or more sites tested, and even those groups are small enough that a few slow sites can swing them. For law firms, only Atlanta and Marietta qualify. Atlanta had 111 of the sites, with a median of 6.2 seconds, and Marietta’s 20 firms came in faster at 4.8, though half of them had no tap-to-call link on the first screen. Among the 12 Atlanta-area cities with 10 or more home service sites, the medians ran from 4.8 seconds in Buford and Fayetteville to 12 in Lilburn, and Kennesaw (11.6) and Acworth (10.9) were also over 10.
What I’d fix first
If your own check turned something up, I’d start with the phone number. Put a tap-to-call link with your number at the top of the first screen. On a lot of sites it’s already somewhere on the page, so the job is moving it.
Next, look at the words on that first screen. Say what you do and where you do it, in plain words, before anything else. If you run a law firm, check your state bar’s advertising rules before you change what your homepage says.
Speed usually takes more work, and the right fix depends on what’s slowing the page down. Our page speed post covers the fixes that make the biggest difference. Markup and a visitor counter come last on my list. Neither one changes what a visitor sees, and both are small jobs for whoever manages your site.
If you run a firm, our page on websites for injury law firms shows what we build around these problems, and local SEO for Georgia law firms covers what Google reads about a firm. For trades, see websites for home service companies.
What this test can’t tell you
This was a lab test on a slow setup, with nothing saved from earlier visits. Visitors on newer phones, on Wi-Fi, or coming back to a site they’ve seen before will often get faster loads. We tested homepages only, and a slow homepage can sit in front of fast inner pages, or the other way around. Any single load can also run slow or fast, so one site’s number is rough. The medians across hundreds of sites are steadier.
The phone number checks read the page’s text, so a number shown inside an image counted as missing, and the analytics check looked only for Google’s tools. The sample leans toward the kinds of businesses we work with. It’s also a snapshot from fall 2026, and some of these sites have likely changed since.
If you’d rather have us look, our free website review starts with this same phone test and puts what we find on one page, in plain words.