I build websites for trades, and the same handful of things decide whether a tradie website gets a call: a number you can tap, a licence you can check, a page that loads on a phone. So I checked them on 303 real homepages. These are the results, with the method and the limits, so you can judge them yourself and quote them with a link to this page.
The short version
- 83% of the Perth and WA electrician homepages Kovus checked in October 2026 (66 of 80) showed an EC licence number, in text or in an image, after a hand check of every site.
- 79% of the Perth and WA plumber homepages Kovus checked in October 2026 (42 of 53) showed a PL licence number on the homepage.
- 77% of 303 Perth and WA tradie homepages (233) had a tap-to-call phone link, and 45% (135) had an enquiry or quote form on the homepage itself.
- The median mobile PageSpeed Insights performance score across 297 Perth and WA tradie homepages was 60 out of 100, with a median Largest Contentful Paint of 6.9 seconds (one lab run per site, October 2026).
- 36% of 303 Perth and WA tradie homepages (109) carried LocalBusiness structured data, and 22% (67) showed an ABN.
A "didn't show" in this study means the item didn't show on the homepage when I checked. It doesn't mean the business is unlicensed or breaking a rule, and some of these businesses show the item on another page.
Licence numbers
In WA, electrical contractors, plumbing contractors, and registered building and painting contractors have a written rule that puts their licence or registration number in their advertising. Building and Energy's guidance for electricians, builders and painters names web pages among those advertisements. The guide to licence numbers on WA tradie websites quotes each rule and links the legislation. This is general information, not legal advice.
I checked every electrician, plumber and builder homepage in the sample by hand, as well as by pattern. 67 of 80 electrician homepages (84%) showed a licence number, as text or in an image, and 66 of those (83% of all 80) showed it in the EC format the electrical guideline describes. 43 of 53 plumber homepages (81%) showed a licence number, 42 of them with the PL prefix.
| Where the licence number showed | Electricians (80 sites) | Plumbers (53 sites) | Builders (5 sites) |
|---|---|---|---|
| As text on the homepage | 73% 58 of 80 | 77% 41 of 53 | Too few to report |
| Only in an image on the homepage | 11% 9 of 80 | 4% 2 of 53 | Too few to report |
| Not on the homepage, but on the contact or about page | 3% 2 of 80 | 0% 0 of 53 | Too few to report |
| Didn't show on the homepage, contact or about page | 14% 11 of 80 | 19% 10 of 53 | Too few to report |
"Only in an image" means the number was in a logo, a badge or a banner, so it is there for a sighted visitor but can't be copied into the licence search, read by a screen reader or read as text by a search engine. "Not found on those pages" means it didn't show on the homepage, the contact page or the about page. It may be on another page, on quotes and invoices, or on the van.
Builders: 5 homepages in the sample, too few to report as a percentage.
Where a number did show, the guide's advice is to put it as text in the header beside the phone number and in the footer with the ABN, and to keep it in one place in the site's code so every page changes together. Checklist item 15 covers the same ground for licences and memberships.
The pattern search alone found a number in the homepage text of 57 electrician and 41 plumber homepages. The hand check is what the published figures use: every pattern match turned out to be a licence or registration number, and the hand check found 14 more numbers inside images, in text a script added after the page loaded, or on contact and about pages.
Calls, forms and contact details
Tap-to-call phone link: 77% (233 of 303)
| Group | Had it |
|---|---|
| All trades checked | 77% 233 of 303 |
| Electricians | 83% 66 of 80 |
| Plumbers | 96% 51 of 53 |
| Pool businesses | 64% 32 of 50 |
| Carpentry and cabinetry | 62% 29 of 47 |
| Roofers | 82% 37 of 45 |
| Renderers | 64% 7 of 11 |
| Other trades | 65% 11 of 17 |
Why it matters: On a phone, a tap-to-call link is the shortest path from finding the business to ringing it. Checklist item 19
Enquiry or quote form on the homepage: 45% (135 of 303)
| Group | Had it |
|---|---|
| All trades checked | 45% 135 of 303 |
| Electricians | 57% 46 of 80 |
| Plumbers | 47% 25 of 53 |
| Pool businesses | 30% 15 of 50 |
| Carpentry and cabinetry | 28% 13 of 47 |
| Roofers | 51% 23 of 45 |
| Renderers | 64% 7 of 11 |
| Other trades | 35% 6 of 17 |
Why it matters: Someone who can't talk right now, on a job site or at work, can still ask for a quote. Checklist item 20
ABN shown: 22% (67 of 303)
| Group | Had it |
|---|---|
| All trades checked | 22% 67 of 303 |
| Electricians | 28% 22 of 80 |
| Plumbers | 36% 19 of 53 |
| Pool businesses | 8% 4 of 50 |
| Carpentry and cabinetry | 15% 7 of 47 |
| Roofers | 18% 8 of 45 |
| Renderers | 27% 3 of 11 |
| Other trades | 24% 4 of 17 |
Why it matters: A customer can type an ABN into the public ABN Lookup and see the business behind the website. Checklist item 1
Link to the Google Business Profile or map listing: 27% (81 of 303)
| Group | Had it |
|---|---|
| All trades checked | 27% 81 of 303 |
| Electricians | 35% 28 of 80 |
| Plumbers | 43% 23 of 53 |
| Pool businesses | 12% 6 of 50 |
| Carpentry and cabinetry | 17% 8 of 47 |
| Roofers | 29% 13 of 45 |
| Renderers | 18% 2 of 11 |
| Other trades | 6% 1 of 17 |
Why it matters: One tap takes a customer to the reviews, the hours and the directions. Checklist item 5
What search engines read first
A place name in the page title: 67% (203 of 303)
| Group | Had it |
|---|---|
| All trades checked | 67% 203 of 303 |
| Electricians | 65% 52 of 80 |
| Plumbers | 77% 41 of 53 |
| Pool businesses | 62% 31 of 50 |
| Carpentry and cabinetry | 55% 26 of 47 |
| Roofers | 73% 33 of 45 |
| Renderers | 100% 11 of 11 |
| Other trades | 53% 9 of 17 |
Why it matters: The title is the blue link in search results. Naming Perth, the suburb or the region tells a local searcher the business works there. Checklist item 8
Meta description written: 69% (209 of 303)
| Group | Had it |
|---|---|
| All trades checked | 69% 209 of 303 |
| Electricians | 68% 54 of 80 |
| Plumbers | 85% 45 of 53 |
| Pool businesses | 66% 33 of 50 |
| Carpentry and cabinetry | 62% 29 of 47 |
| Roofers | 71% 32 of 45 |
| Renderers | 55% 6 of 11 |
| Other trades | 59% 10 of 17 |
Why it matters: Search engines often show it under the link. When there isn't one, they pick a line from the page themselves. Checklist item 9
One main heading (H1): 55% (167 of 303)
| Group | Had it |
|---|---|
| All trades checked | 55% 167 of 303 |
| Electricians | 57% 46 of 80 |
| Plumbers | 70% 37 of 53 |
| Pool businesses | 50% 25 of 50 |
| Carpentry and cabinetry | 47% 22 of 47 |
| Roofers | 51% 23 of 45 |
| Renderers | 27% 3 of 11 |
| Other trades | 65% 11 of 17 |
Why it matters: One heading that names the service is the first line a visitor reads. With none, or several, the page's subject is harder to see at a glance. Checklist item 9
LocalBusiness structured data: 36% (109 of 303)
| Group | Had it |
|---|---|
| All trades checked | 36% 109 of 303 |
| Electricians | 46% 37 of 80 |
| Plumbers | 45% 24 of 53 |
| Pool businesses | 22% 11 of 50 |
| Carpentry and cabinetry | 21% 10 of 47 |
| Roofers | 36% 16 of 45 |
| Renderers | 36% 4 of 11 |
| Other trades | 41% 7 of 17 |
Why it matters: It states the business name, phone, area and hours in code that search engines read directly, alongside the words on the page. Checklist item 2
HTTPS and mobile basics
Served over HTTPS: 95% (288 of 303)
| Group | Had it |
|---|---|
| All trades checked | 95% 288 of 303 |
| Electricians | 98% 78 of 80 |
| Plumbers | 100% 53 of 53 |
| Pool businesses | 92% 46 of 50 |
| Carpentry and cabinetry | 85% 40 of 47 |
| Roofers | 96% 43 of 45 |
| Renderers | 100% 11 of 11 |
| Other trades | 100% 17 of 17 |
Why it matters: Chrome marks a page that isn't served over HTTPS as "Not secure", and a form on it sends what the customer types unencrypted. Checklist item 6
Mobile viewport set: 97% (293 of 303)
| Group | Had it |
|---|---|
| All trades checked | 97% 293 of 303 |
| Electricians | 98% 78 of 80 |
| Plumbers | 100% 53 of 53 |
| Pool businesses | 96% 48 of 50 |
| Carpentry and cabinetry | 94% 44 of 47 |
| Roofers | 98% 44 of 45 |
| Renderers | 91% 10 of 11 |
| Other trades | 94% 16 of 17 |
Why it matters: Without the viewport tag, a phone shows the desktop layout shrunk to fit and the customer has to pinch and zoom to read it. Checklist item 25
Mobile speed
I ran each homepage through Google's PageSpeed Insights once, on its mobile setting, which loads the page on a simulated mid-range phone over a throttled connection. 297 of 303 homepages returned a score. The median performance score was 60 out of 100, and the middle half of sites scored between 48 and 72.
| Group | Homepages |
|---|---|
| Good (90 to 100) | 8% 25 of 297 |
| Needs improvement (50 to 89) | 62% 185 of 297 |
| Poor (0 to 49) | 29% 87 of 297 |
- Largest Contentful Paint, median
- 6.9 s
- middle half 4.3 to 11.3 s
- Page weight, median
- 2.5 MB
- middle half 1.5 to 4.1 MB
- Total Blocking Time, median
- 264 ms
- middle half 59 to 610 ms
- Cumulative Layout Shift, median
- 0.006
- middle half 0 to 0.057
15 of 297 homepages (5%) had their largest image or text block in place within 2.5 seconds in this lab test, the line Google calls good for Largest Contentful Paint. Real-visitor data from the Chrome UX Report existed for only 28 of the 303 sites, which is usual for small business websites, so the figures here are lab figures from one run, not what visitors measured.
Why it matters: A customer standing in a flooded laundry won't wait for a slow page to finish drawing. Checklist items 25 and 28
Platforms and copyright years
The platform comes from the page's own code: WordPress paths, Wix and Squarespace file hosts, a generator tag. Sites built by hand or on a builder the check doesn't know are counted together.
| Group | Homepages |
|---|---|
| WordPress | 53% 160 of 303 |
| Wix | 10% 29 of 303 |
| Squarespace | 8% 24 of 303 |
| Other builders, hand-coded or not detected | 30% 90 of 303 |
Each with fewer than 10 sites, and counted in the last row: GoDaddy, Webflow, Duda, Weebly, Joomla, Astro, Next.js.
213 of 303 homepages showed a copyright year in the footer. Among those, 72% (154 of 213) showed 2025 or 2026. Many platforms update the year by themselves, so a current year shows the site is still running, not that anyone has edited it.
| Group | Homepages |
|---|---|
| 2026 | 63% 135 of 213 |
| 2025 | 9% 19 of 213 |
| 2022 to 2024 | 14% 29 of 213 |
| before 2022 | 14% 30 of 213 |
Why it matters: To a customer, an old year in the footer can read as a website nobody looks after. Checklist item 3
Results by trade
Every check, by trade, as the share of homepages that had it. A trade with fewer than 10 homepages for a check is marked "too few". The same table, with the counts, is in the CSV download.
| Check | All (303) | Electricians (80) | Plumbers (53) | Pool businesses (50) | Carpentry and cabinetry (47) | Roofers (45) | Renderers (11) | Other trades (17) |
|---|---|---|---|---|---|---|---|---|
| Served over HTTPS | 95% | 98% | 100% | 92% | 85% | 96% | 100% | 100% |
| Mobile viewport set | 97% | 98% | 100% | 96% | 94% | 98% | 91% | 94% |
| Tap-to-call phone link | 77% | 83% | 96% | 64% | 62% | 82% | 64% | 65% |
| Enquiry or quote form on the homepage | 45% | 57% | 47% | 30% | 28% | 51% | 64% | 35% |
| ABN shown | 22% | 28% | 36% | 8% | 15% | 18% | 27% | 24% |
| A place name in the page title | 67% | 65% | 77% | 62% | 55% | 73% | 100% | 53% |
| Meta description written | 69% | 68% | 85% | 66% | 62% | 71% | 55% | 59% |
| One main heading (H1) | 55% | 57% | 70% | 50% | 47% | 51% | 27% | 65% |
| LocalBusiness structured data | 36% | 46% | 45% | 22% | 21% | 36% | 36% | 41% |
| Link to the Google Business Profile or map listing | 27% | 35% | 43% | 12% | 17% | 29% | 18% | 6% |
| Copyright year is this year or last year | 72% | 82% | 77% | 67% | 55% | 68% | too few | 91% |
How I checked the sites
Where the sample came from
The sample is trade businesses in Perth and regional WA that I found between May and September 2026, 90% of them through public Google Maps listings and the rest through other public listings and searches. A business went in when its listing put it in Perth or WA (the suburb, the location or a WA postcode in the address), when it was a trade, and when it listed a website. That gave 387 unique websites. Each homepage was checked as it was published on the day, and the study reports totals only, naming no business.
84 of those websites were left out before the checks: sites that didn't load, showed an error or were parked, listings that led to a directory, social, council or franchise page instead of the business's own site, a site outside Australia, and sites that turned automated requests away. That left 303 homepages.
Of the 303 homepages checked, 230 belong to businesses with a Greater Perth address or postcode (6000 to 6214, which includes Mandurah), 46 are in regional WA and 27 gave no postcode. By trade: electricians 80, plumbers 53, pool businesses 50, carpentry and cabinetry 47, roofers 45, renderers 11, other trades 17. The other trades are air conditioning, building, building design, coatings, concreting and landscaping.
How the crawler behaved
The crawler identified itself as KovusResearchBot, with a link to this page. It read each site's robots.txt first and skipped any site that disallowed it, made about one request a second across all sites, gave up on a page after 20 seconds, followed redirects and recorded where they ended. It fetched each homepage once, and for electricians, plumbers and builders it also fetched the contact and about pages linked from the homepage. When a site's bot protection answered with a challenge page, the site was left out. Challenges were never solved or worked around.
What was checked
Each check reads the homepage's HTML as the server sent it. The same code runs the free tradie website check, so a site checked there is measured the same way.
- HTTPS: the homepage ended up on an https address after redirects.
- Mobile viewport: the page sets the viewport meta tag.
- Tap-to-call: at least one tel: link with a number.
- Enquiry or quote form: a form with a message box, or with a phone field, or with an email field and two or more other fields, or a form service's embed. Search, login and newsletter-only forms don't count.
- Licence number: EC, PL, BC or PC followed by digits, or digits after Licence No., Lic. or Reg. No., then checked by hand (below).
- ABN: "ABN" followed by 11 digits in the page text.
- Place name in the title: Perth, WA, a Perth suburb or a WA town or region in the title tag. Suburb names that are also common surnames, such as Wilson or Butler, are left out.
- Meta description: 50 characters or more.
- One H1: exactly one main heading in the HTML.
- LocalBusiness structured data: a LocalBusiness type or one of its sub-types, such as Electrician or Plumber, in JSON-LD or microdata.
- Google Business Profile link: a link to a Google Maps listing or profile. An embedded map with no link doesn't count.
- Copyright year: the latest year after ©, (c) or Copyright in the page text.
How the licence numbers were checked
The pattern search only finds candidates. A street abbreviation, a year or a phone number can look like a licence number to a pattern, and a number inside an image can't be read by one. So for every electrician and plumber homepage, and the builders, I:
- read every pattern match in context to confirm it was a licence or registration number, not a street abbreviation, a year or a phone number;
- opened each homepage without a number in its text in a browser, took a full-page screenshot, ran text recognition over it and read the header and footer by eye, because a number can sit inside a logo or a badge;
- read the contact and about pages linked from the homepage, where there were any.
I didn't look the numbers up on the Building and Energy register. The study records whether a number was shown, not whether it is current or belongs to the business.
Speed
PageSpeed Insights through Google's API: mobile, performance only, one run per homepage, on 15 October 2026 or the day before. Field data from the Chrome UX Report was recorded where it existed.
Sanity check
I also compared the tap-to-call, form and structured-data checks with what a browser showed on 20 randomly picked homepages. Tap-to-call matched the browser on all 20. Structured data matched on 19; on the other, the markup as served was valid and included a LocalBusiness type, but a script on the page changed it after loading. The form check matched on 15; on 4 more it found a working enquiry form in the page code that sat in a pop-up or slide-out the browser had not opened, which still counts as a form on the homepage; on 1 it missed a form whose email field was marked only by its placeholder, so I fixed that pattern and re-ran every site.
What this study can't tell you
- It is not a random sample of every trade business in Perth. It comes from public listings, mostly Google Maps, so it leaves out businesses with no listing and leans towards the trades I build websites for. The median business in it had 41 Google reviews when it was added.
- It is the homepage only. A business can show its licence, ABN or a form on another page. The licence check also read the contact and about pages; the other checks did not.
- It reads the HTML the server sends. Text, forms or links that a script adds after the page loads aren't seen by the automated checks. The licence hand check used a real browser, so it does see them.
- One speed test per site. PageSpeed Insights scores move a few points from run to run. The medians are steadier than any one site's score.
- A point in time. Collected 14 to 15 October 2026. Websites change, and a site that missed a check may have fixed it since.
- It measures other businesses' websites. Sites I built for my own clients are not in the sample.
- Small groups are not reported as percentages. Any group under 10 homepages says "too few to report".
Check your own site
The free tradie website check runs the same checks on any homepage and compares the result with the figures on this page. For the full list of what a tradie website needs, work through the 30-point tradie website and local SEO checklist. If you're an electrician, plumber, builder or painter in WA, the licence number guide shows where the number goes.
Download the data
perth-tradie-websites-2026.csv has one row per check and trade: the check, the trade, the number of homepages, how many passed and the percentage. It has no domains or business names. The aggregate data is free to use under CC BY 4.0: quote it with a link to this page.
To cite it: Kovus, "We checked 303 Perth tradie websites", 15 October 2026, https://kovus.dev/research/perth-tradie-websites-2026
- Data collected
- 14 to 15 October 2026 (AWST)
- Published
- 15 October 2026
- Next update
- October 2027
Questions about the method, or a correction? Get in touch.