Fall River's auto repair shops have 4.75 stars. One website takes 20 seconds to load.

We ran five of Fall River's top-rated auto repair shops through Google's own performance tooling — the same Lighthouse engine that feeds Google's ranking signals — and looked at what a customer actually experiences on a phone.

This is the best-reviewed group of businesses we have audited so far. It is also the one with the widest gap between how good the business is and how good the internet makes it look.

First: who these shops are#

Between them, on Google, these five hold:

2,680 reviews. A weighted average of 4.75 stars. Not one rated below 4.5.

Think about what that means in this particular trade. Auto repair is an industry people walk into already braced to be taken advantage of. Every customer arrives suspicious. And 2,680 of them walked away and voluntarily said these people were honest with me.

You cannot buy that. You cannot fake it. It is years of doing right by your neighbors, one repair at a time.

Hold onto that number. It matters for everything that follows.

Now: what happens on a phone#

All five have a website. The median one takes 3.2 seconds to show a customer anything.

Google's threshold for "good" is 2.5 seconds. Over 4.0 seconds is rated "poor." Three of the five miss "good." Two are rated "poor."

And one takes 20.1 seconds.

That is not a typo. Twenty seconds is a customer tapping back to Google and calling the next shop on the list — a shop with fewer reviews and worse work. That site belongs to a business with 726 reviews at 4.8 stars.

Two of the five deserve real credit: 97/100, loading in about 2.3 seconds, with tappable numbers. Somebody did that work. It is not hard and it is not expensive. It is just unglamorous, so it usually never gets done.

Why the owner has no idea#

Here is the part that matters, and it's the part nobody tells you.

Your website feels fast to you. You open it on the laptop in the back office. You're on your own wifi. Your browser has every image on that site already cached from the last dozen times you looked at it. It snaps open. It's fine. Of course it's fine.

Your customer is standing in a parking lot on Bay Street with two bars of cellular, on a phone that has never seen your site before, downloading every single byte from scratch.

That's the twenty seconds.

There is no alarm for this. Your host doesn't email you. Google doesn't call. Your web guy from 2014 isn't watching. The site is up — it just isn't arriving. The only signal you'll ever get is a phone that rings a little less than it used to, and you'll blame the economy.

The last inch#

Getting a customer to your site is most of the work. Losing them in the last inch is the cheapest mistake in the world. Among the sites we could inspect in full:

• 1 in 3 has a phone number that isn't a tappable link. On a phone, that means your customer has to memorize your number, leave your site, open their dialer, and type it in. Some percentage of them simply won't.
• 2 of 3 have no contact form at all. Somebody decides at 9pm on a Sunday that they need their brakes looked at. If they can't reach you right then, they reach someone else.
• 1 of 3 has no LocalBusiness schema, so Google is guessing at your hours and address instead of being told.
• 1 of 3 has no mobile viewport tag — phones render a shrunken desktop page you have to pinch to read.

The fixes, cheapest first#

1. Make your phone number tappable. This is one line of HTML and it is free.

<a href="tel:+17745598992">774.559.8992</a>

That's it. That's the whole fix. Whoever maintains your site can do it in five minutes.

2. Measure your own site. Go to pagespeed.web.dev, paste in your address, and read the mobile tab. Ignore the desktop score — your customers aren't on desktop. Look for "Largest Contentful Paint." If it's over 2.5 seconds, you now know something you didn't know this morning. It's free and you don't need anybody's permission.

3. Compress your images. In nearly every slow site we measure, the cause is a handful of enormous photos — a 4MB shot of the shop floor being downloaded in full and then squeezed into a 400-pixel box. Resizing them is usually most of a 20-second problem, and nobody will see a difference except your customers, who will see the page.

4. Add a contact form. One field for a name, one for a phone number, one for "what's wrong with it." It catches the customers who find you when you're closed.

The most important thing in this report#

We are not going to publish letter grades next to these businesses' names, and we never will.

A performance score measures what a machine can see in ten seconds. It cannot see the 1,066 people who left a review. It cannot see the customer who was told by a dealer they needed a new transmission and was told by you that it was a $40 sensor. It cannot see years of a reputation built one honest estimate at a time.

These shops are not failing. They are thriving. Their websites just undersell them — and a website is a thing you can fix in an afternoon.

If any of this sounds like you, that is good news. You already built the hard part, the part that takes two decades and cannot be shortcut. What's left is the easy part, and it has been sitting there costing you calls this whole time because nobody ever told you it was broken.

If you want to know your own numbers#

I'll run the same audit on your business and send you what I find — the mobile load time, the missing pieces, and what to fix first. Free, and there is no pitch at the end of it. If you take the list and hand it to your own web person, that's a good outcome.

Joshua Amado — Avalon Partner, Fall River 774.559.8992 · Joshua.Amado@AvalonPartner.com

Method: businesses identified via Google Places. Performance measured with Google PageSpeed Insights (Lighthouse, mobile). "Mobile load" is Largest Contentful Paint — the moment a customer actually sees your page, not the moment your server responds. Two of the five sites blocked our own inspection tool, so their on-page details (tappable phone, contact form, schema) are reported only across the three we could read in full; their load times were still measured by Google directly. Businesses are anonymized; none were contacted or named. Where a metric could not be measured, we say so rather than estimate.