PPC for electrical companies in Australia that books vans

A Gold Coast office manager Googles “switchboard upgrade quote” at 7.12am because the RCD tripped twice overnight. A tradie in Parramatta searches “emergency electrician near me” at 10.41pm after a storm. Same trade. Same Google Ads account if you let it be. Completely different jobs, call windows, and what a booked van is worth in AUD. That mix is where PPC for electrical Australia usually burns money: one campaign, one landing page, one “contact us” form, and a CPC that looks fine until you count how many clicks never become a truck.

Google’s own benchmarks put Australian Search CPC at roughly $2.60-$3.10 for trade services in 2024-25 (WordStream / industry Search CPC reports). Electrical queries in metro Australia sit well above that average on after-hours and “near me” terms. Google Ads still posts about 4.2x average ROAS across industries. Plenty of electrical firms never see that because 20-30% of SEM budget still dies on irrelevant queries, neglected negatives, or a homepage that was never built for the ad. This piece is about splitting the auction the way a dispatcher already splits the board, then making the click survive the page.

After-hours callouts vs solar upgrades: two auctions, one wasted budget

If you run one Search campaign called “Electrical - AU”, Google will happily spend on “how to reset a safety switch”, “apprentice electrician jobs Brisbane”, and “cheap LED downlights Bunnings”. Those queries look electrical. They do not pay for a van. High CPC pressure then tempts you to yank smart bidding back to manual on a new account. That is a symptom. The cause is mixed intent in one bucket so the algorithm cannot tell a $180 callout from a $4,800 switchboard and EV charger package.

Australian homeowners and facilities managers search in clusters. Storm and power-out language: “power out electrician”, “RCD tripping”, “no power after storm”. Compliance and upgrade language: “switchboard upgrade AS/NZS”, “safety switch install”, “three phase upgrade”. Project language: “EV charger installation”, “solar inverter electrician”, “rewire quote”. Commercial language: “office fitout electrician”, “test and tag contract”. Each cluster has a different time-to-book, a different proof set (licence, insurance, after-hours fee, photos of boards you have actually done), and a different cost per qualified job. Smash them together and automated bidding looks greedy because it is optimising for the cheapest conversion event you gave it, usually a form fill from a tyre-kicker.

Performance Max will drive bulk if you let it own the board. For electrical firms that is often a trap. PMax is fine as a support layer once Search already isolates emergency, upgrade, and commercial. It is a poor owner of brand, competitor, and “near me” terms when your conversion action is “any form”. Smart bidding with clean signals has been shown to reduce effective CPC by double digits versus manual - Google’s own case studies cite 15%+ efficiency gains - but only when conversion actions reflect real jobs, not homepage form fills. They inflate spend without improving quality when the only signal is a thank-you page from a homepage contact block.

Mobile takes 63%+ of paid search clicks, and conversion still lags desktop by 30-40%. After-hours electrical demand is almost all mobile. If your click-to-call is buried, your after-hours surcharge is missing, or the page takes too long on 4G, you paid for a storm click and handed it to the next listing. According to the Chrome UX Report (Q1 2025), roughly 56% of origins globally pass all three Core Web Vitals - electrical service pages with heavy image galleries rarely make that cut. Electrical service pages that load a 12MB gallery of “our team” rarely sit in that set.

If the account still sends every ad to the homepage, that is the first thing to stop. Dedicated pages beat a generic site every time. For paid search setup that stays tied to booked jobs rather than vanity CPL, our Google Ads and SEM programs are built around this split, not a single blended ROAS number.

The post-click leak: forms, missed calls, and slow pages in AUD terms

The click is not the expensive part. A missed after-hours call is. You will see $8-$18 CPCs on “emergency electrician Sydney” and still lose the job because the page asked for a 9-field form, the number routed to a voicemail that says “office hours 7-4”, or the map embed stalled Interaction to Next Paint on a phone. Last-click then tells you Search “did not convert”. The van that went to a competitor 11 minutes later tells a different story.

Privacy and ad blockers already strip 20%+ of conversion data. Smart bidding gets dumber when Enhanced Conversions and call events are missing. Electrical is worse than SaaS because a large share of revenue never hits a form: the customer taps Call. If that call is not a conversion with duration and source, you train Google on the wrong 40% of demand. CM360-style and in-platform numbers have drifted since late 2025. Expect Google’s CPA and your job-management software to disagree. Treat the job booked in Simpro, Fergus, or ServiceM8 as the source of truth, then feed a tighter conversion action back (qualified call over 90 seconds, or a “booked job” offline import), not “thank you page”.

Message match fails in boring ways. Ad says “after hours electrician, no call-out if we cannot attend tonight”. Page headline says “Quality electrical services since 1998”. Ad says “switchboard upgrades from $X, licensed, photos of recent boards”. Page leads with a stock photo of a smiling sparkie and a “get a free quote” box with no postcode field. You did not have a CPC problem. You had an offer problem. Session behaviour makes this obvious: rage taps on the phone number that is not tappable, scroll death above the licence numbers, form abandon on “how did you hear about us”.

That is where HeyLead Insights belongs on an electrical site: recordings and heatmaps on the emergency and upgrade landers, not a monthly PDF of blended CPL. You want to see whether storm traffic ever reaches click-to-call, and whether upgrade traffic ever reaches the photo proof of a three-phase board. Do not guess. Watch the session.

Core Web Vitals still decide who keeps the click on mobile. Compress the hero, lazy-load job photos, keep the number sticky, and drop third-party chat widgets that steal INP during a blackout search. Check the live URL with a Core Web Vitals checker before you scale spend. A pretty ROAS in Google Ads with a slow lander is just an expensive way to train competitors’ brands.

Playbook: split campaigns, negatives, and landing proof for Australian electrical firms

Build the account around jobs a dispatcher would recognise, then give each job a page that could close the call without a human explaining the ad. Keep structure stable. Constant rebuilds reset learning. Broad match can work once negatives and conversion quality are tight. It wastes 20-30% of budget when they are not.

A pattern we see repeatedly: a Melbourne-based operations lead running a $4,200/month blended campaign. Search terms were full of “lighting design course” and suburbs an hour outside the SLA. They split emergency from upgrades, added 140 negatives in two sittings, and moved storm ads onto a tap-to-call page. Over 90 days, form volume dropped 22%. Booked after-hours jobs rose because the remaining clicks were people who wanted a sparky tonight, not a PDF. That is the mechanism: stop paying for the wrong auction, then stop losing the right one on the page.

A pattern we see repeatedly: a Brisbane owner kept PMax on because “Performance Max drives the bulk now”. It did. Bulk was ceiling-fan tyre-kickers and a competitor’s brand. They carved brand and emergency back to Search, fed 90-second calls as the primary conversion, and left PMax a thin upgrade asset group. CPC did not magically fall. Cost per booked board did. Realistic timelines still sit in the 3-6 month band if tracking was a mess on day one. Anything that promises a fixed ROAS before seeing the account is a pitch, not a plan.

What marketing leaders are seeing

HeyLead works with Australian trade businesses on exactly this split - emergency, upgrade, and commercial, each in its own campaign.

FAQs

Is broad match worth turning on for an electrical Google Ads account?

Yes, after negatives, geo, and a qualified-call conversion exist. Broad without those is how 20-30% of budget becomes DIY and jobs queries. Signals over keywords only works when the signal is a real job.

Is Performance Max enough on its own?

No. Use it as a support layer. Keep emergency, brand, and high-intent upgrades in Search so you can still read search terms and write RSA copy that names after-hours fees and licence numbers.

Why does in-platform CPA disagree with booked jobs?

Call-heavy trades lose data to ad blockers, missed call tracking, and last-click double counting. Import booked jobs from your field software and treat Google’s number as directional.

What should we spend before it is “enough” data?

Enough to cover your top two job families in the suburbs you actually service, with room for after-hours. A tiny budget across every service resets learning and makes automated bidding look worse than it is. Do not expect a guaranteed CPL on week one.

Do we still need keywords if AI bidding is the default?

Google still needs keywords and negatives, even as delivery leans on signals. Intent clusters belong in campaign structure. “Run broad” is not “run unsupervised”.

Putting it to work

This week

• Pull 30 days of search terms and mark every query that never becomes a van.

• Split emergency from upgrades if they still share a campaign.

• Put tap-to-call on the after-hours lander and kill extra form fields.

• Switch the primary conversion to 90-second calls or booked-job import.

• Run the emergency URL through the Core Web Vitals checker and tag ads with the UTM builder.

Next 30 days

• Build the switchboard / EV lander with real board photos and suburb proof.

• Carve brand and competitor out of PMax.

• Reconcile Google CPA against Simpro, Fergus, or ServiceM8 booked jobs once a week.

• Pause suburbs and hours you cannot staff. Raise budget only on the family that books.

Start by exporting search terms and counting how many paid clicks were never a dispatchable electrical job in your Australian service area. If you want that split, the landers, and the call-to-job tracking taken off your plate, HeyLead can run the PPC program so emergency auctions and upgrade auctions stop sharing a budget. Chat with us on WhatsApp