Every Swansea Salon With a Website Has a Booking Tool. None of Them Has Analytics.

The received wisdom about small businesses and software is that they don't buy it. That the salon on GAR Highway is still running a paper appointment book and a landline, and the job of a firm like mine is to drag them into 2026.

That is not what the data says. Not here.

On 31 August 2026 we re-measured the six hair salons Google returns for Swansea, MA. Three of them have a real website of their own. All three have online booking installed — three different platforms, all paid, all working. These owners bought the tool. They did the hard part.

None of the three has any analytics installed at all. Not Google Analytics, not a Meta pixel, not Plausible, not Clarity, not Matomo, not one line of anything. So every one of them is running a booking system into the dark, with no way to answer the only question that matters: does it actually work?

What we measured, and what we could not#

Six businesses. Two have no website at all. One uses a Facebook page as its website, which we could not measure — Facebook serves a JavaScript shell, and any speed score you get back describes Meta's infrastructure, not the salon's. We are not going to publish a number about a business that is really a number about Facebook.

That leaves three measurable sites. Small sample. We're telling you the denominator every time.

Every LCP above is a lab estimate. Google had no real-visitor field data for any of the three, which is itself a finding: these sites don't get enough traffic for Google to assemble a 28-day sample. Treat the numbers as directional, not as what a specific customer experienced on a specific Tuesday.

The review scores, ratings and "no website" flags come from our 22 July Google Places pull. The Places API returned 403 PERMISSION_DENIED for us today, so we could not refresh them. As of July these six salons held 878 Google reviews between them and averaged 4.74 stars, none below 4.4. We have no reason to think that has moved much, but we did not measure it today and won't pretend we did.

The tools are good. The instrumentation is missing.#

Across the three sites we found three separate booking platforms — GlossGenius, Square Appointments and Resurva. All three are real products with real reputations. One salon has gone all the way and made the booking platform be the website.

That is a completely defensible decision for a six-chair salon. It's fast to set up, it handles deposits and no-show fees, and it doesn't require anybody to learn WordPress.

Here is the problem. Every one of those platforms ships a reporting dashboard. Bookings by source. Bookings by day. Abandoned bookings. Repeat-client rate. And when a small business has no site analytics and has never opened the booking tool's own reports, nobody in the building can tell you whether last month was good.

Joy Hawkins has spent close to two decades inside local search results, and her point about where local businesses actually lose money is not about rankings at all:

The average small business's response times are terrible

Being found is the part these salons have already won. Four-point-seven stars across 878 reviews is not marketing. It is years of doing good work on real heads of hair. The leak is downstream of that — in the thirty seconds between a customer deciding to book and the site giving them a way to do it.

Which brings us to the second finding. Two of the three sites have a phone number you cannot tap. It's printed on the page as text. On a phone, the customer has to memorise it, leave the site, open the dialler and type it in. A meaningful share of them simply don't.

What we got wrong in July#

We ran this same category on 22 July. Re-measuring it caught four errors in our own audit tooling, and it would be dishonest to quietly fix them and say nothing.

• We reported one salon as having no online booking. It has GlossGenius. Our tool's list of booking platforms didn't include GlossGenius, so the most booking-centric site in the set scored zero.
• We reported another as running OpenTable. It doesn't. Our checker matched two Squarespace theme CSS class names that happen to contain the word "opentable". The real tool is Resurva.
• We reported one site as having Google Analytics. It has a bare gtag() stub that Square Online ships by default, with no config call and no gtag.js. It measures nothing. That means the real analytics figure isn't "3 of 4 have none" — it's 3 of 3, worse than we said.
• We counted social profiles that don't exist, because our link check was a substring match. One site's only facebook.com string was the Open Graph XML namespace declaration in its <html> tag.

We also have to retract an outbound claim: we flagged one business for an SSL certificate "expiring in 8 days." That certificate belongs to Facebook and rotates automatically. It read 8 days in July and reads 8 days today.

A measurement tool that has never been checked against reality is exactly the thing this article is complaining about. Ours had gone six weeks without a second look.

One thing we cannot verify, and neither can anyone selling you a report#

Facebook and Instagram profile health cannot be checked from a server. Facebook returns HTTP 400 for real and fake pages alike. Instagram returns HTTP 200 and a login wall for every handle you throw at it, including ones that don't exist. Any tool that hands you a green tick next to "Instagram: active" without access to your own page token is guessing.

We report those as unmeasured. If a vendor's audit scores your social presence and can't explain how, that number came from nowhere.

What a Swansea salon owner should do on Monday#

Cheapest first. Real time estimates.

1. Open your booking tool's reports tab. (Free. 10 minutes.) You are already paying for this. GlossGenius, Square Appointments, Vagaro, Booksy, Boulevard and Resurva all ship reporting. Find bookings-per-week for the last three months and repeat-client rate. If those two numbers surprise you, that is the cost of six months of not looking.

2. Make your phone number tappable. (Free. 5 minutes.) This is the giveaway. If your site's phone number is plain text, it needs to be a link. In almost every website editor you highlight the number, click the link button, and instead of a web address type:

tel:5085551234

Digits only, no spaces, no dashes, no parentheses. That's it. Now one tap dials you. If you're on Square Online, Wix, Squarespace or GlossGenius, the link field accepts tel: exactly like a normal URL. Do it on your homepage, your contact page, and your footer.

3. Install analytics. (Free. 20 minutes.) Google Analytics 4 costs nothing. So does Microsoft Clarity, which additionally records how far people scroll before leaving. Either one, once, and next month you will know how many people reached your booking page and how many finished.

4. Check that every booking link is https://. (Free. 10 minutes.) Click every "Book Now" button on your own site, on your own phone, and look at the address bar. If any of them show http:// without the s, edit the link and add it.

5. Add LocalBusiness schema and a meta description. (An afternoon, or ask someone.) Without them Google guesses at your hours and writes your own search snippet for you.

None of the first four requires a developer, a contract, or us.

The honest summary#

The Swansea salons are not technologically behind. They bought good booking software and they earned excellent reviews. What they don't have is a feedback loop — any mechanism at all for finding out that the site takes nine to sixteen seconds to paint on a phone, or that the number isn't tappable, or that five booking links throw a browser warning.

That's not a software problem. Nobody has ever sold them the habit of looking.

If you run a salon in Swansea, Somerset or Fall River and you want someone to actually check — not sell you a rebuild, just tell you what's true — that's the conversation I like having.

Joshua Amado, Avalon Partner · 774.559.8992 · Joshua.Amado@AvalonPartner.com

---

Method: mobile Lighthouse via Google PageSpeed Insights on 31 August 2026, plus a server-side fetch of each homepage. Thresholds are Google's own — LCP under 2.5s is "good," over 4.0s is "poor." You can reproduce any speed number by pasting a URL into pagespeed.web.dev. Businesses are anonymised because they did not ask to be graded in public. Anything we could not measure is reported as not measured, never as zero.