Four of the Six Driving School Websites We Could Read Near Somerset Have No Analytics At All

A sixteen-year-old in Somerset gets their permit on a Thursday. That night, a parent sits down with a phone and searches "driving school near me."

They are not comparison shopping the way they would for a car. They want three things: is this place legitimate, how much does it cost, and can I get my kid into the next class before the school year starts. They will open a few tabs, and within about ten minutes one school will get a phone call and the others will not.

Here is the part that should bother every owner in that list: of the six driving school websites we could read, only two have any way of knowing that this happened. The other four have a website that is, as far as they can tell, a brochure that sits in a drawer.

What was measured, and how#

On 31 July 2026 I pulled every driving school Google Places returns for the Somerset area, in Google's own relevance order, and kept the ones inside roughly a twenty-minute drive of Somerset. That is seven businesses: one in Somerset, four in Fall River, one in Seekonk, one in Westport. Google's list also returns schools in Brockton, Bridgewater, Franklin and Rhode Island. Those are real businesses, but no Somerset parent is driving to Franklin for a road test, so they are excluded and not counted against anyone.

For each one I fetched the homepage and checked what is in the HTML: analytics tags, structured data, contact form, booking tool, whether the phone number is tappable, whether the certificate is valid. Then I ran Google PageSpeed Insights for the mobile lab score and, separately, asked whether Google's Chrome User Experience Report has any real-visitor data for that site at all.

Two honest limits, up front. One site could not be read by our crawler, so it has no HTML findings; it is marked "not measured" rather than scored badly. And one site's PageSpeed run returned an error, so it has no performance number at all. Neither of those is a failing grade. It is just a gap, and a gap is not a finding.

Everyone is anonymised. These businesses did not volunteer to be graded in public.

The table#

The letters are not in search-rank order, and I am not publishing addresses. The point is the pattern, not the pillory.

The two things this table actually says#

First: most of these owners are flying blind on purpose, without knowing it.

Four of the six sites we could read have no analytics tag of any kind. Not a light setup. Nothing. No page views, no idea which page people land on, no idea whether the traffic comes from Google or from a Facebook post or from the sign on Route 6. If one of them redesigned their site tomorrow, they would have no way to tell whether it helped or hurt. If one of them lost half their traffic next month, they would find out from a quiet phone, three months late.

That is not a technical failure. It is a business one. You cannot improve a number you have never seen.

Second: at least three of the seven are too small for Google to have measured them either.

This is the part almost nobody knows. Google's Chrome User Experience Report only publishes performance data for a site once enough real Chrome users have visited it to make a stable sample. Three of these seven sites do not clear that bar. Google has no real-visitor data on them at all. A fourth I could not check either way, because its PageSpeed run failed outright.

So for those three, every performance number in existence, including the ones in the table above, is a simulation. A robot in a data centre pretending to be a phone.

That is Google's own documentation telling you to trust the field data over the lab data. Fine. Except three of these seven schools have no field data to trust, and a fourth could not be checked. Which leaves them with the lab number, which Walton's own article exists specifically to explain is frequently wrong.

The way out of that trap is not a better test. It is your own analytics, on your own site, reporting on your own actual visitors. That is the whole reason the tag exists.

Give the top of the list its due#

Site A has 1,281 Google reviews. The next-highest in the group has 122. That is not a marketing trick. That is a very long time teaching people in this region how to drive, one nervous teenager at a time. You do not fake four figures of reviews at 4.8 stars.

Site A is also the only one in the group with a full measurement stack, the fastest real-visitor experience of the three sites Google has field data for at 1.2 seconds at the 75th percentile, and a perfect 100 on the accessibility audit. I do not think the reviews caused the analytics or the analytics caused the reviews. I think both come from the same habit: somebody there treats the business as a thing worth paying attention to.

And Site A still has no LocalBusiness schema and no way to book a lesson online. Nobody in this group is finished.

What it is costing, roughly#

Here is the honest version, because I am not going to pretend I can price this precisely for a business whose books I have never seen.

Three of the seven sites take longer than 13 seconds to render their main content in the mobile lab test. Two of those three have real-visitor data, and it tells two completely different stories. One is genuinely slow for real people: 4.3 seconds at the 75th percentile, which Google classifies as poor. The other is genuinely fast for real people, 2.1 seconds, which Google classifies as good, despite a 17.5-second lab score. That gap is exactly what Walton's article is about, and it is the reason I will not sell you a lab number as an experience.

For the third of those three there is no field data at all, so the 13.6-second figure is a simulation and I am not going to tell you it is what your customers sit through. What I will tell you is that nobody has checked, because there is no analytics tag installed to check with.

Two of the seven have a broken or absent secure connection. One serves its site over plain HTTP, so Chrome puts "Not secure" in the address bar next to a business that is asking parents to hand over a teenager. The other has a certificate that does not cover the exact address people reach it at, which means some visitors get a full-screen red security warning instead of a homepage. That one is not a slow site. It is an invisible site.

Six of six have no LocalBusiness schema, so when Google builds the little panel with your hours and your address it is inferring rather than reading.

One of the seven has no site of its own at all in Google's listing. The "website" link points at a third-party directory page. Somebody else owns the front door.

The free fix, today, sixty seconds#

Do this before you do anything else. It costs nothing and it will tell you which of the two problems you have.

1. Go to pagespeed.web.dev and put in your own homepage URL.
2. Look at the top of the report for the panel headed Discover what your real users are experiencing.
3. If it shows you numbers, Google has real-visitor data on you. Those numbers are the truth and you should act on them.
4. If it says there is not enough data, that is your answer. Google does not have a large enough sample of real Chrome visitors to report on your site. Everything below that panel is a simulation.

That is it. Sixty seconds, no account, no card. Three of the seven schools in this audit would get the second result, a fourth I could not check, and none of them appear to know it.

The fixes, cheapest first#

Install Google Analytics 4. Free. About twenty minutes. Create a property at analytics.google.com, copy the one snippet it gives you, paste it before the closing </head> tag on your site. If you are on Wix, Squarespace, GoDaddy or WordPress there is a settings field that takes the measurement ID directly and you never touch code. Two of the seven schools here have already done it. This is the single highest-value twenty minutes available to the other four, because everything after it is guesswork until it exists.

Fix the certificate. Free. About an hour. If your host offers Let's Encrypt, and almost all of them do now, turning it on is a toggle. Make sure it covers both yoursite.com and www.yoursite.com, because covering only one of the two is exactly the failure in this sample. Then force HTTPS so the old plain-HTTP links redirect.

Add LocalBusiness schema. Free. About an hour. A short block of JSON-LD with your name, address, phone, hours and price range, pasted into the head of the homepage. Six of six sites here are missing it. Google's Rich Results Test will confirm it parsed. This is the cheapest local-SEO work that exists and essentially nobody in this category has done it.

Make the phone number tappable. Free. Five minutes. Two of the six sites print the number as plain text. On a phone, that means a parent has to memorise it, leave your site, open the dialer and retype it. Wrap it in <a href="tel:5085551234">. That is the entire fix.

Put the class schedule and the price on the page. Free, but it takes a decision. Every one of these schools gets asked the same two questions on every call: how much, and when does the next class start. Answering them on the site does not lose you the call. It loses you the call you were going to lose anyway, and wins you the one where the parent had already decided before dialling.

Add online booking or a real intake form. Usually $0 to $20 a month. One of the seven has a scheduling tool. Parents research this at 9pm. A phone number is a closed door at 9pm.

What I could not measure#

I could not read one site's HTML at all, so I do not know whether it has analytics, a form, schema or anything else, and I have not counted it against anyone. One PageSpeed run failed outright, so that site has no performance number here and I also do not know whether Google holds real-visitor data on it. I did not measure social posting frequency on any of them, because there is no reliable way to verify Facebook or Instagram activity from the outside, and I would rather say nothing than say something I cannot stand behind. Review recency was only pulled for some of the group.

If you want to argue with the numbers, good. Run the sixty-second check on your own site and see what comes back.

If you run one of these schools#

You do not need an agency to do the first four items on that list. You need an afternoon and somebody who is not afraid of a settings page. Do the sixty-second check tonight and the analytics install on Monday morning, and by next weekend you will know something about your own business that you have never known.

If you would rather it was just done, and done properly, that is what I do. Avalon Partner is in Fall River. I audit and build for businesses within about thirty minutes of here, which includes every school in this piece.

774.559.8992 · Joshua.Amado@AvalonPartner.com