Skip to content
Up To Code

Manual 2 · Your Website · 2.15

Test your own site like a customer:
the 20-minute self-inspection

Passed final You — from the truck Takes: Twenty minutes — forty with the speed test

Your site's real job is what a stranger sees when they google your name before deciding to call. So the test that matters is walking through it the way that customer would — on a phone, before they'd dial. Twenty minutes catches the things that quietly cost you work.

All you need is your own phone. Do it on your cell signal, not the shop's wifi, so you see what a customer actually sees.

The walk — six checks, twenty minutes

Go in order. Each one ends in a yes or a no you write down. (No website yet? Run check 01 — the listing check — and pocket the rest for the day you have one.)

01Google your own business name

On your phone, search your business name. Do you come up at all? Is your Google Business Profile there — the box with your hours, phone, and reviews? If it's wrong or missing, nothing else matters yet.

Now read it the way a customer does, because each wrong detail has its own cost. Wrong hours: someone drives over or waits to call, hits a closed door, and now distrusts everything else the listing says. Old address: the map routes them to a place you left. Old phone number: the call rings somewhere that isn't you — and if a former vendor's tracking number is still on there, your calls may be ringing into a dead line. And look for a second listing of your own shop — every review that lands on the wrong twin is a review customers never see on the real one.

If you find a twin, don't start deleting. Work out which one carries your reviews, then follow Google's own duplicate-and-ownership page below. Screenshot what you file and put a calendar note three weeks out.

And if you don't show up at all — no listing, no box — that's not a failing grade, it's a missing part: your Business Profile doesn't exist yet. Creating it is free and it's Google's own flow at business.google.com. Know the second half before you start: building the profile takes minutes, getting it verified runs on Google's clock. Plan on recording a live walkthrough video — that is what most trade businesses get handed now, and Google says the review takes up to 5 business days, though owners in trades commonly report longer. Google's help pages still mention a mailed postcard code; that is rare now, so do not plan around it. When it doesn't take, Google says the signal can be nothing more than the "Get verified" button coming back, and the fix is to run verification again. No phone number is published for a free profile owner to chase it with — Google's own help list is a help center, a contact form, a community forum, and a YouTube channel. So set it up in one sitting, screenshot the confirmation, and put a date on the calendar two weeks out; while you wait, don't change your business name, address, or category. Our full walkthrough is coming in the getting-found manual. Nothing else on this page matters until that part exists.

Check: you show up, every detail is current, and there's exactly one of you. yes / no

Before you go fix what you found: an edit isn't always live when you hit save. Google says edits usually take up to 10 minutes to review, but sometimes up to 30 days, and that it might not approve a change if it can't confirm it's accurate. Make the edit once, screenshot it with the date, and check again in a week. If it comes back not approved, the doors are the ones above: help center, contact form, community forum.

02Open your site and time how fast it loads

Tap through to your website. Count how long until you can actually read and use it. This isn't a taste thing — Google publishes the yardstick: the main content of a page should load in 2.5 seconds or less to be good, and slower than about 4 seconds is rated poor. A customer on a job-site phone won't wait; they hit the back button and call the next name.

Then get the real number instead of a guess. Type your site below and the button opens Google's own free test, PageSpeed Insights, already running on your site — no login, no account, and nothing you type here goes anywhere except into that Google page's address.

Check: your page is usable in a couple of seconds, and your Largest Contentful Paint number is written down. yes / no

03Find the basics the way a panicked customer would

Pretend it's 2 a.m. and your kind of emergency just happened. Phone in one hand, thumb only, and no patience for your history or your logo. Four things have to be findable: proof you do this kind of work, proof you cover their area, a phone number, and any reason to believe you're real.

So run that test on your own site: from the first screen, can you tell what the shop does? Is the phone number visible without scrolling — and is it a button that dials when tapped, not just printed text? Can you find the service area in one scroll? Common ways this check fails, worth looking for specifically: the phone number lives only on a "Contact" page two taps deep, the number is an image that can't be tapped, or the services hide inside a slideshow that doesn't work on a phone. If you have to pinch, zoom, and hunt, so does the customer — and at 2 a.m., they don't.

Check: what-you-do, area, and a tap-to-call number are all reachable from the first screen in a few taps, and the number actually dials. yes / no

04Send yourself a message through your own contact form

Fill in your site's contact form as a customer and send it. Give it fifteen minutes before you judge it — an empty inbox at sixty seconds proves nothing. Then go check: did it actually reach an inbox you look at every day? This is the check that catches a form that's been mailing leads to an address nobody reads.

Three details make this test real instead of half-done. Check the spam folder too — a form that lands in spam is nearly as dead as one that lands nowhere, because you'll never think to look. Try replying to your test message — some forms send from a no-reply address, which means every customer you answer gets a bounce and thinks you ignored them. And know why these break silently: forms die when a site gets moved, when an email address gets retired, or when whatever piece of software runs the form quietly stops working — nothing looks different from the outside, which is why nobody notices until the slow year.

If your test lands in a dead or forgotten inbox, that's a leak, and fixing where your leads land is Chapter Zero, Door 4.

Check: your test message arrived in an inbox you monitor, not in spam, and replying to it works. yes / no

05Read it as a stranger would

Scroll your own site slowly, like someone deciding whether to trust you with their house. The question they're answering is "is anybody actually home?" — and there are specific tells they read without knowing it. Check yours for each:

  • A copyright year in the footer from three years ago.
  • Photos that are obviously stock — gleaming models in spotless hard hats — instead of your actual trucks, jobs, and people.
  • Words that could belong to any shop in any town: "we are committed to quality and customer satisfaction." A stranger's eye slides off filler; it reads as nobody-wrote-this, which reads as nobody's-here.
  • Anything half-built: "coming soon" pages, menu links that go nowhere, a blog with two posts from years ago. Half-built reads worse than small-but-finished.

Trust your gut on the overall read; it's the same gut your customers use. But check all four. The footer year takes minutes. Real photos and words written for your shop are a job you schedule.

Check: current year, real photos, words written for YOUR shop, nothing half-built. yes / no

06Check what customers touch, not the vendor's dashboard

If you pay someone monthly, they may show you a chart with impressions and clicks climbing. Ignore it. The only numbers that pay your crew are the phone ringing and the form filling. One owner looked at his own visitor numbers and realized the only clicks were his own, from testing. Judge the work by what a customer actually does, not by a graph.

And the honest catch: a call is just a call — nothing on it says where it came from. Ask every new caller "how'd you find us?" and keep a tally by the phone, starting this week. When one says "I just googled you," that's Google, not which part of Google — give it its own column.

Check: you can point to real calls or form-fills from the site in the last month — or your how'd-you-find-us tally started this week. yes / no

The truth about ranking, so no vendor can fog you

While you're inspecting, know the ground rules, because they let you catch a lying pitch. Google ranks local businesses on three things: how well your listing matches what someone searched (relevance), how close you are to the searcher (distance), and how well-known your business is (prominence). Google's own wording on reviews is careful: more reviews and positive ratings can help your local ranking. Can, not will — hold it that loosely. And Google states flatly: there is no way to request or pay for a better local ranking. Read that precisely, because there IS paid space on the page: the ads are for sale, and they're labeled as ads. What's not for sale — ever — is the ranking itself, the listing your shop earns. So a vendor selling you ads is selling something real (whether it's worth it is the price page's question). A vendor promising to buy your ranking is selling something that isn't for sale.

Print this: the 20-minute inspection sheet

The 20-minute self-inspectionon your phone, on cell data

A "no" on the form test sends you to Chapter Zero to fix which inbox your leads land in. A "no" on check 03 or 05 — a buried or untappable phone number, a site that reads abandoned — is work on the site itself: take the sheet to whoever you'd hire, and if you're weighing whether a paid rebuild is worth it at all, start at do you actually need a website. A vendor whose dashboard looks great while the phone stays quiet is a conversation for the five questions.