Jacksonville Local SEO Statistics: 158 Service-Business Websites, Tested

Key findings

Finding 01

About one in four listings has no website to send people to.

Google gives every business profile a website button. On 44 of 211 listings that button was missing. Another 7 pointed to a Facebook page or a directory instead of a site the business owns. It varied a lot by trade. Pool companies and event planners almost always had a site. Auto detailers and exterior cleaners often didn’t.

The businesses without a site also had far fewer reviews: a median of 5, against 45 for businesses with one. That doesn’t mean a website earns reviews. Newer and smaller operations tend to have neither yet.

Finding 02

Most of these sites are slow on a phone.

We loaded each site the way Google’s PageSpeed test does: a mid-range phone on a slow 4G connection. The number that matters most to a visitor is how long the main content takes to appear (Google calls it Largest Contentful Paint, or LCP). Google’s own guidance says 2.5 seconds or less is good and anything over 4 seconds is poor. The median Jacksonville site in this sample took 6.7 seconds.

43 of 127 took longer than 10 seconds. Weight is a big part of it. The median page was 2.7 MB, and 39 of 127 were over 5 MB, which is usually large photos and video that were never resized for a phone. Overall, 20 of 127 sites scored 90 or better on Lighthouse’s performance score, and the median was 70 out of 100.

These are lab numbers under deliberately tough conditions. A customer on a new phone with good signal will see faster loads. A customer in a driveway with two bars may see something close to this.

Finding 03

14 listings send people to a website that doesn’t open.

Of the 158 websites linked from these Google listings, 14 failed before any content loaded. We only counted a link as broken when it failed twice in our test browser and twice more in a plain request.

1. Domain no longer exists6 sites
2. Browser security warning (expired or invalid certificate)4 sites
3. Page not found4 sites

A further 17 sites couldn’t be measured because they blocked or stalled our test browser. Those may work fine for a real visitor, so we left them out of both the broken count and the speed numbers.

Finding 04

The SEO basics are mostly in place. A few common gaps remain.

Lighthouse’s SEO check covers the basics a search engine needs to read a page. The median score was 92 out of 100, and 101 of 127 sites scored 90 or better. The first two rows below come from our own read of each page’s source, and the other two from Lighthouse.

1. No business structured data in the page source (the labels Google and AI assistants read)60 of 139 sites
2. No tap-to-call phone link on the home page32 of 139 sites
3. Links with vague text such as “click here”26 of 127 sites
4. No meta description18 of 127 sites

Finding 05

Accessibility is where the most sites fall short.

The median accessibility score was 89 out of 100, and 59 of 127 sites reached 90. Automated checks catch only part of what makes a site usable, so the real picture is likely worse. These failed most often.

1. Text too faint against its background83 of 127 sites
2. Links a screen reader can’t name (often icon-only links)67 of 126 sites
3. Headings out of order66 of 125 sites
4. Images with no description28 of 124 sites
5. Pinch-to-zoom turned off, or no mobile viewport set14 of 125 sites

Finding 06

WordPress runs the largest share of these sites.

We could identify the platform from the page source on 139 sites. “Other or custom” means we found none of the common builders’ fingerprints. The groups are small, and a few sites may be misidentified. Read the speed column as a description of this sample. It does not rank platforms. We build custom sites ourselves, so weigh this table with that in mind.

What to do with this

If you run a service business in Jacksonville

1. Tap your own website button. Open your Google Business Profile on your phone and tap Website. If it errors or shows a security warning, fix that first. It costs nothing and it’s the first thing a customer does.
2. Test your home page on a phone. Run it through Google’s free PageSpeed Insights (pagespeed.web.dev) and look at one number: Largest Contentful Paint on mobile. Under 2.5 seconds is the goal.
3. Shrink your photos. Heavy pages were common in this sample, and oversized photos are a usual cause. Resize photos to the width they display at before uploading them, and keep video off the top of the home page.
4. Make the phone number tappable. A customer on a phone should be able to call you with one tap from the top of the page.
5. Check your text contrast. Faint text was the most common accessibility failure we found. Light gray on white and white text over a photo are the usual culprits. If you squint to read it outside, so do your customers.

Methodology

How we ran it, and what it can’t tell you

The sample comes from Google Maps results 904 Digital Media had already collected between October 6 and November 8, 2026, for searches such as pressure washing, pool service, appliance repair, mobile detailing and event planner in the Jacksonville area (18 result sets, 409 listings before removing duplicates). We kept businesses located in Duval, St. Johns and Clay counties, removed duplicates, and grouped them into five trades. That left 211 businesses, 119 of them in Jacksonville and 60 in St. Augustine. 158 of them linked to a distinct website of their own, counting a brand with several locations once.

On November 9 and 10, 2026, we ran Google Lighthouse 13.6.0 once against each site’s linked page, with Lighthouse’s default mobile settings (a simulated mid-range phone on slow 4G), from one computer in Jacksonville. Sites that errored were retried once. 127 sites returned a full report. We also fetched each home page’s HTML to check for a tap-to-call link, structured data and the platform it was built on.

• Our own clients are excluded. We build websites for a living, so we removed every 904 Digital Media client from the sample before measuring.

• No business is named. The data file uses anonymous IDs. This is a picture of a market and says nothing about the quality of any company’s work.

• This is not a random sample. It covers five trades and the businesses Google showed for those searches, which leans toward businesses that already rank. Other trades may look different.

• One lab run per site. Lighthouse scores move a few points between runs and differ by test machine. We did not use Google’s field data from real visitors, because most small sites don’t have enough traffic to appear in it.

• One page per site. We tested the page each listing links to, which is usually the home page, and nothing deeper.

• What we did not measure: rankings, traffic, leads, content quality, or whether a faster site earns more business. Nothing here shows cause and effect.

• Listing data is a snapshot. Review counts and website links are as Google showed them on the day each result set was collected.

Get the data

Free to use with credit to 904 Digital Media (CC BY 4.0): one row per website, with anonymous IDs (CSV). Journalists and researchers can reach Ryan Warren at hello@weare904digitalmedia.com.

Related: our small-business website cost benchmark lists published prices for building and running a small business website, from website builders to Jacksonville-area shops, each with its source.

How to cite: 904 Digital Media, “Jacksonville Service-Business Website Study,” November 2026, weare904digitalmedia.com/research/jacksonville-service-business-website-study.