How Much Do Google Ads Cost for Contractors? Clicks, Leads and Jobs

Google Ads costs come in three layers: cost per click, cost per lead and cost per job. Published benchmarks, billing rules, and how to check your own trade.

By Gavin Sevastian · Updated August 4, 2026 · 15 min read

When a contractor asks what Google Ads cost, they are usually asking three questions at once. What does a click cost? What does a lead cost? What does a booked job cost? Those are three different numbers, they move independently, and only one of them lands in your bank account.

The click price is the one everyone quotes, and it is the one you control least. The job price decides whether any of this was worth doing, and most of it has nothing to do with your ad account.

Below is what a click costs by trade in published US data, how Google actually meters your budget, how to look up your own trade in your own city in Canadian dollars, and how to work out whether the whole thing clears your margin.

How much do Google Ads cost? You are asking three questions

A click is priced by an auction, not a rate card

There is no price list. You tell Google the most you are willing to pay for a click, and what you actually get charged is decided in a live auction against everyone else bidding on that search, in that city, at that moment. Two roofers in the same town, bidding on the same word, on the same day, can pay meaningfully different amounts.

Any published cost per click figure is therefore a market condition rather than a quote. It tells you the neighbourhood you are shopping in. It does not tell you your price.

A lead is a click divided by your conversion rate

This is the arithmetic that catches people out. If a hundred clicks cost you a thousand dollars and five of those people call, your leads cost two hundred dollars each. If ten of them call, your leads cost a hundred.

The auction did not change. Nothing about your bidding changed. Your page did. Halving the conversion rate doubles the lead cost with no movement in the click price at all.

Almost everything on the conversion side is yours to fix: whether the page matches the ad, how fast it loads, how long the form is, whether the phone gets answered, and whether the traffic arriving was ever going to hire anybody.

A job is a lead divided by your close rate

Then leads become jobs, or they do not.

Which brings up the most expensive misunderstanding in this whole category. A contractor comparing cost per lead quotes between two vendors, without knowing which of these three numbers each one is quoting, is comparing nothing at all. Ask which number it is before you compare it to anything.

What a click costs, trade by trade

Published US benchmarks by trade

The figures below come from LocaliQ's 2025 search ad benchmarks for home services, drawn from 3,211 US search campaigns running between April 2024 and March 2025, with a minimum of 103 active campaigns behind every category. All values are USD, and the source states its averages are technically medians so that outliers do not distort them.

Read down the last two columns before you read anything else. The most expensive click on that table is roughly two and a half times the cheapest one. The most expensive lead is roughly five times the cheapest one. The click price is not where the money separates.

Roofing shows this most clearly. An ordinary-looking click price near the top of the range, and by far the highest lead cost on the table, because fewer than four in a hundred visitors convert. Getting roofing leads is a conversion-rate problem long before it is a bidding problem.

Plumbing makes the contrast. Its click price is a shade below roofing's at $10.49 USD, but its conversion rate is roughly double, which drops the lead cost to $129.02 USD against roofing's $228.15 USD. Plumbing searches tend to carry more hire-now urgency, and the lead cost reflects it. Same neighbourhood on the click, very different outcome on the lead.

Where home services sits against every other industry

The wider WordStream 2026 all-industry benchmark set, drawn from 13,474 US campaigns running from April 2025 to March 2026, puts home and home improvement at $8.33 USD per click. That is behind only legal services at $9.87 USD, against an all-industry average nearer $5.42 USD.

Home services is a premium vertical, and it is priced that way because the jobs behind the clicks are worth real money. If you have ever read a general "average cost per click" article and thought the figure looked nothing like your trade, that is why.

One caveat the report states about itself matters if you compare figures across sources. Those tables blend Google Ads and Microsoft Ads together rather than reporting Google alone, so treat the ranking as directional.

Why nobody publishes a real Canadian benchmark

Everything above is US data in US dollars, and that is not an oversight. There is no independent, methodologically disclosed Canadian study of contractor click costs.

Every Canadian range circulating online traces back to an agency blog with no sample size, no date window, and no stated method. Some of them are probably close to right. There is no way to tell which ones, which makes them useless for a decision.

That gap is worth naming rather than papering over, because the alternative is worse. A confident-looking Canadian number with no method behind it is not more useful than no number. It is less useful, because you will act on it.

There is a better answer than a benchmark table anyway. You can pull a Canadian figure yourself, free, from Google's own planning tool, in Canadian dollars, for your own city. That matters more than it sounds: the auction in Toronto prices differently from the auction in a town of four thousand, and a provincial or national average is the two of them blended into a number that describes neither. There is a walkthrough further down this page.

What counts as a good cost per click

The honest answer is that the question is slightly off target.

Two contractors bidding the same word pay different prices

You are not charged your maximum bid. Google's documentation on actual cost per click states that you pay only what is minimally required to clear the Ad Rank thresholds and beat the advertiser immediately below you, and that where no competitor below you has cleared their threshold, you pay only the reserve price.

The number you set and the number you pay are different numbers, and the gap between them is decided by everyone else in the auction. Bidding higher does not automatically mean paying more per click, and a competitor running a weak account can make your clicks cheaper without you doing anything.

Two caveats so this does not get oversimplified. It is not a clean second-price auction, because minimum quality thresholds and a reserve price sit in the middle of it. And the same Google page notes that your actual cost per click can exceed your maximum in some configurations, such as when bid adjustments are in play.

Quality is a discount, and Google says so

Google is unusually direct about this one.

Google makes the same point from the opposite direction in its glossary. Its definition of a top of page bid estimate says the estimate is based on the keyword's Quality Score and competition from other advertisers, and that if your estimate is very high, it means the keyword's Quality Score is poor and could be improved. A high price is treated as a symptom, not a fixed cost of entry.

The 1 to 10 score in your account is a diagnostic, not the auction input

Two things get blurred together here, and separating them saves a lot of wasted effort. Ad quality changes what you pay, as above. The 1 to 10 Quality Score number sitting in your keyword table is a different object.

Google says so directly. Its documentation on Quality Score states that "Quality Score is not an input in the ad auction. It's a diagnostic tool to identify how ads that show for certain keywords affect the user experience", and adds that it "is not a key performance indicator and should not be optimized or aggregated with the rest of your data".

The useful part is the component ratings underneath it, each marked above average, average or below average. Google says that judgement is "based on a comparison with other advertisers whose ads showed for the exact same search over the last 90 days". So a below average rating is not an abstract grade. It is you against the advertisers you were actually up against, recently.

Anyone selling you an average Quality Score as a headline result is reporting on a diagnostic tool.

Cheap clicks are usually the expensive ones

Now the part most people get backwards. Cheap traffic is cheap for a reason. Research searches, DIY searches, parts shoppers and people looking for a job in your trade are all inexpensive to reach, because nobody who is about to hire someone is typing those words.

A cheap click that never converts costs exactly as much as a good one and produces nothing. Fill an account with them and your click cost drops while your lead cost climbs, which reads as progress on the report most people check first.

A good cost per click is one that produces a cost per lead your job margin can carry. You cannot judge it at all until you know the other two numbers.

Cost per lead: the number the click price hides

The click price is the part of this you can shop around. The lead price is the part you can actually change.

Two published sources disagree by more than two to one

WebFX publishes its own home services marketing benchmarks and puts the average home services search cost per click at about $3.50 USD. The LocaliQ table above puts it at $7.85 USD. Same industry, overlapping periods, two published sources, and a gap of more than double.

Neither one is lying. WebFX draws on its own client database and does not publish a sample size or a method, so it belongs on this page as a contrast rather than a correction. LocaliQ publishes both.

The disagreement itself is the useful part, because it settles what a benchmark is for. A benchmark tells you the order of magnitude. It does not forecast your business, and anyone quoting one as a guarantee is selling something.

You cannot multiply your way from a click to a lead

LocaliQ states that its "averages" are technically medians, which stops a handful of extreme campaigns dragging the numbers around. That is the right call statistically. It also means the columns do not compose.

Divide a median click cost by a median conversion rate and you do not get the published median lead cost. You can check this on a calculator in thirty seconds.

Across every row in that table the chained figure lands anywhere from about 8 percent below the published lead cost to about 43 percent above it, with the middle of the pack sitting around 15 percent high. Most trades land reasonably close. A few are a long way out, and you have no way of knowing in advance which kind yours is.

What actually moves the conversion rate

Look at the two ends of the trade table. Cleaning services convert at nearly eighteen in a hundred. Roofing converts at under four. That is not mostly a difference in skill between advertisers. A blocked drain and a full roof replacement are different purchases with different urgency, and one of them involves a spouse, three quotes and a two-week think.

What you can move sits on your side of the click:

• Whether the page says the same thing the ad said.
• Whether it loads before the person gives up.
• How many fields the form asks for.
• Whether the phone gets answered on a Saturday, which is a large part of what separates HVAC accounts that produce work from ones that do not.
• Whether the person was in your service area to begin with.
• Whether you are paying for searches that were never going to hire anyone, because excluding the wrong traffic lifts the conversion rate without touching a single bid.

Cost per job: the only number that touches your bank account

Leads are an intermediate step. Nobody deposits a lead.

Two identical accounts, double the customer cost

As a worked example with made-up round numbers: say you spend CAD 4,000 in a month and it produces 25 leads, which puts your lead cost at CAD 160. Close eight of them and each customer cost you CAD 500 to acquire. Close four and each one cost you CAD 1,000.

Same spend. Same leads. Same ad account, running identically. The acquisition cost doubled inside your office. This is how two contractors run comparable campaigns and reach opposite conclusions about whether Google Ads works, and it is why the sales side deserves at least as much attention as the bidding side.

Speed of response is the cheapest lever you own

You paid the same price for that lead whether you called back in four minutes or four hours. One of those versions costs you far more per booked job than the other, and it does not appear anywhere in your ad reporting.

Not every lead is a job, and not every job is the same job

If your trade spans a CAD 300 service call and a CAD 18,000 install, a single blended cost per lead tells you almost nothing. Half your leads could be profitable at that price and half underwater, and the average would look fine.

Compare cost per job to job margin rather than to job price, because the price includes materials and labour you do not keep. And when you compare Google Ads against referrals, compare cost per booked job on both, because referrals carry a real acquisition cost that is simply invisible: the discount, the follow-up, the years of goodwill.

Most of this section is not the ad account's job. That is worth being blunt about, because an account can look broken when the real gap is response time.

Look up your own trade, in your own city, in CAD

You do not have to take anyone's word on price. Google publishes the estimate itself, free, and it takes about ten minutes.

What Keyword Planner is actually showing you

The column worth reading is the top of page bid estimate. Google defines it as "the bid you likely need to set for your ad to be shown among the ads at the top of the first page of search results", and adds that it "approximates what cost-per-click (CPC) bid is needed for your ad to show on the top of the first page of search results when a search query exactly matches your keyword".

Three things in that definition do real work.

It is a bid, not a price paid. It tells you roughly what you would have to be willing to pay to compete at the top of the page. What you actually get charged per click is the auction mechanic from earlier on this page, and it is usually less.

It is not a floor. Google says plainly that your ad "can still appear if your bid doesn't meet this estimate, but it's less likely to appear on the top of the first page". You are not locked out below the number. You are just less visible.

It is partly about you. The estimate is based on the keyword's Quality Score alongside competition from other advertisers, so a very high estimate is Google telling you the relevance is weak, not only that the market is expensive. Google also says outright that "the estimate isn't a guarantee".

That is what turns a mystery number into a usable one. It is a planning figure for a bid, with your own account quality baked into it, produced by the company selling the clicks.

How to run it for your trade and your city

1. 1

   Create a Google Ads account and stop before you launch a campaign

   You need an account to reach Keyword Planner. You do not need a live campaign, so skip past the setup prompts rather than completing them. Keyword Planner sits with the other planning tools in the main menu.

2. 2

   Set the location to your actual service area

   Do not leave it on Canada. Set it to your city or your region. A provincial or national average blends downtown Toronto with towns of four thousand people, and neither number is yours.

3. 3

   Check the currency shown is CAD

   If it reads USD, every figure in front of you is in the wrong currency and none of it is your number. Switch it before you read anything else. If you cannot switch it, convert at the day's rate rather than reading a US dollar figure as though it were Canadian.

4. 4

   Enter three or four things a customer would actually type

   Not your service names. The words a worried homeowner types with one thumb. Furnace not working, roof leak repair, garage door spring broken. Add your city to a couple of them and leave it off the others so you can see the difference.

5. 5

   Read the top of page bid columns rather than the forecast

   The low and high top of page bid estimates alongside monthly search volume are the useful output. The forecasting tool moves with recent data and with the season, so a figure pulled in January will not match one pulled in June for a seasonal trade.

6. 6

   Do it again for a second set of words

   Run the emergency phrasing and the considered-purchase phrasing separately. In most trades they price differently, and knowing which half of your work is expensive to buy changes how you would structure a budget.

What the number does not include

What it costs to have somebody run it

Three separate costs, usually quoted as one

There are three live costs here, and quotes that bundle them are almost impossible to evaluate.

Ad spend goes to Google and buys clicks. The management fee goes to whoever runs the account. Build cost covers the thing the ads point at: a landing page that converts, conversion tracking that works, call tracking if you need to know which calls came from where.

That third one is where most quotes go quiet, and it is also where most underperforming accounts have their real problem. When someone quotes you CAD 1,500 a month, ask how much of that Google receives. If the answer is vague, you cannot compare the offer to anything.

Percentage of spend, flat fee, and what each one rewards

The most commonly cited range is 10 to 20 percent of monthly ad spend, with flat monthly fees and hybrid structures alongside it.

Grade that source before you lean on it. It comes from a reporting-software vendor surveying its own agency users. No regulator, trade body or independent survey publishes PPC fee data that I have been able to find, so treat the range as what the industry says about itself rather than as researched fact.

Neither of the first two is dishonest. They pull in different directions, and you should know which incentive you are buying. For the longer version, including what the fee is supposed to cover, here is what Google Ads agencies charge.

What Google requires anybody managing your account to tell you

You do not have to rely on a manager's own account of what is fair. Google publishes policy for third parties who resell or manage ads, and it reads like a buyer's checklist.

On the fee itself: "If you charge a management fee (separate from the cost of Google Ads or AdWords Express), you must clearly inform customers. At a minimum, inform new customers in writing before each first purchase and disclose the fee on all customer invoices."

On what gets reported back to you: "When sharing Google advertising cost data with customers, report the exact amount charged by Google, exclusive of any fees that you charge." The minimum report has to "include data on costs, clicks, and impressions at the Google advertising account level". The spend number you are shown should be Google's number, with nothing folded into it.

There is also a trigger sized almost exactly for contractor accounts. Where "80% or more of your customers spend less than $1,000 USD (or local currency equivalent) per month", the manager must share Google's disclosure notice with every customer who buys, and "The notice must be linked in a clearly visible location on your website, such as the footer of your homepage, the advertiser section of your site, your advertiser reporting dashboard, and the products or services section of your site".

Read that list before you go looking. Four locations qualify, and Google says "such as", so a homepage footer with no link in it does not prove anyone is out of compliance. The stronger check is the other half of the same rule: the notice has to reach you directly at a new sale or a renewal, "by either emailing them a soft copy or mailing a printed copy". If you never received one, ask for it.

The account should also be yours alone. Google requires "that you use a separate account for each end-advertiser that you manage", because combining advertisers "can result in Quality Scores that inaccurately represent any one advertiser's performance".

One more from that page, because contractors try it constantly. On splitting a budget between two agencies to see who wins: "You're welcome to do this, but we recommend that you test them one at a time. Google doesn't display more than one ad for a business on a page of search results, so you'll find it difficult to test multiple third parties at once."

What the Google Partner badge is scored on

The badge tends to get presented as proof of competence, so it is worth reading the published criteria.

Partner status requires a registered manager account to "maintain a 90-day ad spend of $10,000 USD across managed accounts", a "minimum optimization score of 70%", and a "minimum of 50% of your account strategists certified in Google Ads".

Premier Partner is the "top 3% of participating companies within a given country", and Google's stated factors include "Existing client growth: Ability of partners to grow their existing clients, measured by year-over-year ads spend growth and ads spend among existing Google Ads clients" and "Product diversification: Demonstrated investment in product mix beyond Search".

Those are Google's words rather than a criticism of anyone holding the badge. Read what the criteria measure, and decide for yourself what a badge scored on spend, certifications and product mix tells you about whether your phone rings.

Why cheap management is usually not management

The work does not scale down with the budget. Conversion tracking takes the same work to set up correctly at CAD 1,000 a month as at CAD 10,000. Search terms need reviewing at the same frequency. Negative keyword lists need the same maintenance.

So on a small account the fee is a large share of total cost, and the only way to make it small is to do less work. That is what a very cheap fee is quietly telling you.

To say the obvious thing against my own interest: everything above this line, you can do yourself. The lookup, the arithmetic, the go or no-go decision. Plenty of contractors run their own accounts perfectly well.

How Google actually charges you

This is the part that generates panicked phone calls in week one.

Google publishes no minimum spend

Google's own budget page, linked in the next section, never names a floor. You set an average daily budget yourself, per campaign, and a very small one will run.

Read that as what Google publishes rather than as a term in a contract, because Google does not state it in those words on any page I can point you at. It still settles the practical question: anyone telling you Google requires a minimum spend is describing their own fee floor rather than the platform's rules.

What comes off the card is metered against that daily budget, though not in the tidy way most people expect.

Two times daily, 30.4 times monthly

Google's budget documentation states the two limits directly, and both are worth committing to memory before you turn anything on.

The number that surprises people is the day. The number that matters is the month. If you set CAD 100 a day and see CAD 187 on a Tuesday, nothing has gone wrong, and the month will still land at or under 30.4 times your daily figure.

Working backwards from a monthly number

Most contractors think in monthly budgets, so run it the other way. Google gives the arithmetic on the same page: divide the monthly figure by 30.4, the average number of days in a month.

CAD 3,000 a month divided by 30.4 gives an average daily budget of about CAD 98.70. That is the number you type into the campaign, not the 3,000.

It is worth checking this rather than assuming, because the two numbers look similar enough that people enter the monthly figure into the daily field and only notice when the first invoice arrives.

A weekday-only schedule does not reduce the monthly limit

Nearly every trade runs some kind of schedule. Weekdays only, business hours only, nothing overnight. It is natural to assume that shrinks the bill. It does not.

Google's documentation on spending limits is explicit: "If your campaign uses ad scheduling, your monthly spending limit will continue to be 30.4 times your average daily budget. The system will pace to reach this full monthly limit regardless of how many days the campaign is scheduled to be active."

The month stays the same size. It gets compressed into fewer serving days, so the days you do run will spend harder than the daily figure looks like it permits. Editing the schedule partway through changes nothing either: "If you adjust your ad schedule during the month but don't change your budget, your monthly pacing goal will not change."

The one thing that genuinely shrinks a month is a late start. "If your campaign starts during a calendar month, we'll only take into account the days the campaign was running." A start date prorates. A schedule does not.

Changing the budget mid-month does not restart the month

Raise your budget on the 24th and you do not get 30.4 times the new number for what is left.

Google's page on budget changes states it: "When you change your budget, your spend for the rest of the month won't exceed your new average daily budget multiplied by the remaining number of calendar days in the month. Note that this applies to campaigns using ad schedules also."

Google's own worked example: $103 USD already spent by 24 November on a $5 daily budget, raised to $10, gives "$103 USD spent so far + ($10 USD/day * 7 days remaining in November) = $173 USD monthly spending limit". What you have already spent stays counted against the month.

Two details on that page catch people out. On a day you edit the budget, the daily limit is based on "the highest average daily budget that you chose for that day". And "changing your campaign end date is considered a budget change", which is not obvious from anywhere in the interface.

Served costs can go over. Billed costs cannot

The spending limits documentation also covers what happens when delivery runs ahead of the limit, which is worth knowing before you go looking for someone to blame on a heavy day.

So a report showing a figure above the limit is not the same as an invoice showing one. Check the billed number before you escalate.

One campaign type sits outside the daily half of the rule altogether. Google notes on that page that with a pay for conversions campaign, "there is no daily spending limit but the campaign will always stay within the monthly spending limits".

Why the card gets charged on odd dates for odd amounts

Budget metering and billing timing are two separate systems, and the second one is what makes a first statement confusing.

Google's billing documentation says charges "don't usually happen once a month or at the end of the month. They can happen multiple times throughout the month, and are based primarily on payment thresholds", meaning a cost amount that triggers a charge once your account reaches it. If you never reach it, "you'll get automatically charged on the first date of the month", and that first of the month charge "may sometimes exceed your payment threshold".

The threshold itself moves as you spend, which a separate Google page on changing how often you are charged states directly: "Your payment threshold might rise if you tend to reach that limit before your next automatic charge. For example, if your threshold is $50 USD and you reach that amount multiple times, the threshold might raise to $200 USD or more." Charges getting larger and less frequent over time is normal, and is not a change to your budget.

Local Services Ads bill on the same mechanic. Google states you are billed either "30 days after your last payment, or whenever your account reaches a predetermined amount known as your payment threshold", whichever comes first.

The floor: spending too little is its own failure mode

Google publishes no minimum. Your data has one

The real floor is statistical. Below a certain number of clicks a month, nobody can tell a good account from a bad one, because there is not enough happening to separate signal from noise.

At very low volume, one good week and one dead week look like a trend and are actually a coin flip. Every decision made off that data is a guess wearing a spreadsheet. That applies to you, to a hired manager, and to Google's own bidding algorithms equally.

How to work out your own floor

You already have everything you need to compute this for your own trade and your own city.

1. 1

   Take your trade's top of page bid range

   Run the Keyword Planner lookup for your own city as described earlier. Take a number between the low and high estimate as a working figure, and remember it is a bid rather than a price paid.

2. 2

   Divide your monthly budget by it

   That gives you roughly how many clicks a month the budget buys. A CAD 1,500 month against a CAD 12 click is about 125 clicks. Against a CAD 5 click it is about 300.

3. 3

   Apply a published conversion rate for your trade

   Use the conversion rate column from the benchmark table above as a rough guide. Multiply it by your click count to get an expected monthly lead figure.

4. 4

   Ask the only question that matters

   Look at that lead number and ask whether it is enough to tell a good month from a bad one. If one lead landing or not landing would swing your read of the whole account, you are below your floor.

There is deliberately no threshold number here, because the honest answer changes by trade. A trade with a two percent conversion rate and a twelve dollar click needs far more budget to produce a readable month than one converting at fourteen percent on a six dollar click. Work out your own.

What Google publishes instead of a minimum budget

Google will not name a budget floor. It does publish minimums for reading results, which is the same problem seen from the other end.

For evaluating performance, Google recommends a window covering "at least 2 full conversion cycles", and says it "can be helpful to assess over longer periods of time, like a month, or at least 50 conversions, for a clearer perspective". For a Target CPA read specifically, it says to "measure performance for the last 30 days, including at least 30 conversions".

Take those as reading thresholds rather than entry requirements. The same Target CPA page says plainly that "Advertisers can start using Target CPA with no conversion history, and Target CPA is effective for campaigns of all sizes". You are allowed to start small. You are just not able to judge it yet, and knowing the difference is what stops a premature verdict.

Google also tells thin accounts to bid differently while they build up data: "If you're using automated bidding like 'Maximize conversions' without enough historical data, switch to 'Maximize clicks' or 'Manual CPC' to build conversion data first." And it warns off the metric most people reach for while waiting: "Don't look at trailing, dependent metrics like cost-per-click and impressions."

Two things stretch the wait further. Google says a bid strategy change can take "up to 3 weeks or 1-2 conversion cycles" to calibrate, and that its algorithms "continue to learn even when the bidding status no longer shows 'Learning'". Below a certain volume the diagnostics go blank entirely: Google says a dash in the Quality Score column means there are not enough searches exactly matching your keywords to determine a score.

Read those together and the floor stops being a matter of opinion. A budget that cannot produce enough conversions to fill an evaluation window is a budget nobody can grade, including Google.

What a too-small budget actually looks like

The failure mode is not dramatic. It is three months of spending, a handful of leads, no way to tell whether the account is broken or the market was quiet, and a decision to quit made on no evidence. The money is gone and nothing was learned, which is the worse half of that.

Seasonality belongs in the same calculation. It is real in almost every Ontario trade, and an annualized monthly average hides it completely. Furnace demand and deck demand do not arrive in twelve equal instalments, so a budget that reads as adequate on paper can be far too thin in the eight weeks that actually matter.

Run the numbers against your own job value

Work forwards from a job to a lead ceiling

1. 1

   Start with your average job value

   Use your real average, not your best month and not your biggest job. Say CAD 6,000 for the sake of the example.

2. 2

   Multiply by your gross margin

   The share left after materials and labour, not the invoice total. At 35 percent that is CAD 2,100 of gross profit per job.

3. 3

   Multiply by your close rate

   The share of quotes that turn into work. At one in four that is CAD 525 of gross profit per lead.

4. 4

   That figure is your ceiling

   It is the absolute most a lead can cost before the program stops making money, before a single dollar of overhead or management fee. Your target sits well below it.

Now compare that ceiling against the published lead costs in the benchmark table, remembering those are USD and US market conditions. If your ceiling is CAD 525 and your trade's published lead cost sits near USD 200, there is room. If your ceiling is CAD 90 and your trade's published lead cost is USD 228, you have your answer and you did not need to spend anything to get it.

Work backwards from a lead target to a budget

1. 1

   Start with the jobs you want

   Say four extra jobs a month. Use a number you could actually service, because a budget sized for work you cannot deliver creates a different problem.

2. 2

   Divide by your close rate

   At one in four, four jobs means 16 leads a month. This is the step where an optimistic close rate does the most damage, so use the rate you can evidence.

3. 3

   Multiply by a realistic lead cost

   At CAD 200 a lead, 16 leads is CAD 3,200 a month in ad spend, before any management fee or build cost.

4. 4

   Ask whether you can sustain it

   Not for one month. Long enough to read the result through a slow season as well as a busy one. If the answer is no, narrow the scope rather than running the same plan on a third of the money.

Narrowing means fewer services, a tighter geography, or one trade instead of four. A small budget aimed at a small target can work. The same budget spread across everything you do cannot.

Where this calculation lies to you

The single most expensive assumption in this exercise is the close rate. Someone who has never bought leads tends to assume they become jobs at close to one for one, because that is roughly how referrals behave. Paid leads do not work that way.

A close rate borrowed from referral work will overstate the outcome badly, and every number downstream inherits the error. The program then gets judged a failure at the point where it was performing normally.

What Canadian contractors actually spend on advertising

For context, Innovation, Science and Economic Development Canada reports that specialty trade contractors spent about 1.1 percent of total revenue on advertising and promotion in 2024, across 164,896 businesses, with about 70 percent of them reporting the expense at all.

Three caveats before that number goes near a decision. It covers total advertising, not Google Ads. It is Canada-wide rather than Ontario. And the publisher rates the figure E on its own reliability scale, which is its use-with-caution band.

The same table splits that spending by profit-margin quartile, from 1.3 percent in the least profitable quartile down to 0.8 percent in the most. Resist the obvious conclusion. It is a percentage of revenue, so a business with lower revenue shows a higher percentage for identical dollars. It is not evidence that advertising less makes you more profitable.

Whatever the figure, it is money spent on the business rather than in it. Advertising is a deductible business expense on line 8521 of form T2125, and the restrictions the Canada Revenue Agency names on that page concern advertising in periodicals and advertising with foreign broadcasters. That page does not address online advertising specifically, so treat the detail as a question for your accountant.

What makes the number worse than it needs to be

Everything below raises your real cost per job without touching the auction, and most of it is invisible on the report you are looking at.

The settings worth checking before you raise a budget

Several Google settings widen the number of billable events, which is reasonable for Google to want and costly for you to leave unexamined. None of these are wrong by definition. They are just decisions somebody should make deliberately.

Location targeting can include people merely interested in your area rather than only people located in it. Check which setting yours is on, and whether it matches the way you actually take work.

Search partners and display inventory are separate surfaces a search campaign can be extended onto. Check whether yours are switched on, and whether you meant them to be.

Auto-applied recommendations let the platform change your account without you approving each change. Check whether they are enabled and what they have already changed.

A keyword cannot tell you the size of the job

This one costs people real money. The idea is that if you bid only on words like luxury, high end, premium or custom, you will only get big jobs.

It does not work, and the reason is structural. A keyword tells you what somebody is looking for. It does not tell you the size of what they need. Two people can type the same three words and want a CAD 400 repair and a CAD 20,000 replacement.

Job size gets qualified on the page, in the form, and on the phone. Not in the keyword list. This is a different thing from separating genuinely different services into their own campaigns, which is ordinary and correct.

Set and forget is the failure mode

Search terms drift as match types expand. Competitors enter the auction and leave it. Seasons turn and the same keyword changes value. Performance decays without ongoing search term review, negative keyword maintenance and budget management, and none of that is dramatic enough to notice week to week.

That is a fact about how the platform works, and it is the honest reason management is a recurring cost rather than a setup fee.

More budget on a structurally broken account amplifies the waste in exact proportion, so fix the destination and the tracking before raising anything. And treat platform recommendations as suggestions with an interest attached: they tend toward broader matching, higher budgets and additional inventory, so test each one against booked jobs rather than against clicks. If you would rather have someone else carry that maintenance load, that is what this business does, but the work itself is the same either way.

Google Ads versus paying per lead

Plenty of contractors weighing this are also weighing Local Services Ads. The structural differences matter more than the headline prices.

Paying per lead instead of per click

Google's Local Services Ads charge you "for each valid lead you receive", which means a call, a message or a booking rather than a click. Google states that lead prices "may vary depending on your location, the job type, the type of lead, or your bidding mode".

Budget works differently too. It is set as an average weekly figure rather than a daily one, and leads "determined to be invalid or low quality are not charged". Charged leads get reassessed over time and "may be issued credits automatically if determined to be low quality", which has no equivalent in a search campaign.

Worth knowing that the shape of this option is changing. Google has published a transition of Local Services Ads toward Performance Max campaigns with pay-per-lead goals, so anything you read about how it works, including this page, should be checked against a recent date. The deeper version of the comparison is in the guide on Local Services Ads for contractors.

Where the two budget systems behave differently

Two Local Services mechanics catch contractors mid-month, and both are the opposite of how a search campaign behaves.

The monthly maximum is a hard stop rather than a pacing target. Google states that "the exact formula for the monthly max is average weekly budget x 30.4 ÷ 7. Once you have hit your monthly max, your ad will no longer appear for the remainder of the month unless you adjust your budget." Search paces toward its monthly figure. Local Services stops dead at it.

And a budget change ignores what you have already spent: "When you change your average weekly budget, your new budget will be used to determine your ability to spend for the remaining days in the month, regardless of how much you've spent so far." Google's example has a weekly budget raised from $100 to $140 on 18 January, making the account "eligible to spend $280 ($140 weekly budget ÷ 7 days per week x 14 days left in the month) for the remaining days in January". A search campaign counts your prior spend against the month. This one does not.

Local Services also carries a published budget recommendation, which search ads never get. On automated bidding Google says "it is recommended to set a minimum budget of 10 leads per week for optimal results", and that "it takes about 2 weeks to see accurate results as the model adjusts and learns about your business". Multiply ten leads by the lead price you are quoted locally and you have a floor figure in your own currency, which is more than the search side hands you.

Read the credit rules before you count on them

Credits are a real advantage of the model and narrower than they sound.

Google no longer supports credits "for 'job type not serviced' and 'geo not serviced' leads", which are the two categories a contractor would most expect to claim. Credits are also unavailable "for health care verticals, tax specialists, or advertisers in EMEA".

Timing matters for your bookkeeping. "In most cases, credits will be applied to your account balance within 30 days", and "the original lead charge will still appear on your invoice", so the raw invoice total overstates what you actually paid. Google adds that despite automatic screening, "you may still notice some poor quality leads processed via auto credits within the first 24 hours".

None of that makes the credit system worthless. It means a lead price quoted net of credits you have not checked is an estimate.

If you are also weighing a shared-lead marketplace

Some platforms sell the same enquiry to several contractors at once. Before comparing any of them on price, ask the platform directly whether the lead is exclusive to you, and get the answer in writing.

That single question decides whether a lead price is even comparable across channels. Then compare the channels on cost per booked job rather than cost per lead, because those two numbers only track each other when the leads behind them are alike.

They are not mutually exclusive

The two answer different searches, so running both is not a contradiction. Local Services Ads are bought by service category with the lead as the billable unit, which suits work that is booked quickly and described simply.

Search ads let you write the ad, choose the exact wording you bid on, and send the click to a page you control. That matters most for the considered purchases, where somebody needs persuading before they will pick up the phone.

What to do this week

You now have everything you need to reach a defensible go or no-go without spending a dollar.

1. Open Keyword Planner, set your city and CAD, and pull the top of page bid range for three or four terms your customers actually type.
2. Divide the monthly budget you could sustain by that figure to get a monthly click count.
3. Apply your trade's published conversion rate to get an expected monthly lead count, and check it is enough to read.
4. Work out your gross profit per lead: average job value, times gross margin, times close rate.
5. Compare that ceiling against your trade's published lead cost. If there is comfortable room, this is worth testing. If there is not, fix margin, close rate or job value first, because no amount of bidding skill will rescue arithmetic that does not work.
6. Decide whether you are running it yourself or paying somebody, and price those separately rather than as one bundled number.

The expensive version of this is guessing, spending for three months, and still not knowing which of the three numbers let you down.

Everything on this page is yours to run without hiring anyone, and a contractor who works through those six steps carefully will land in roughly the right place.

What changes when somebody does it daily is the maintenance: the search terms that drift, the wrong-intent traffic that creeps back, and the tracking that quietly breaks. Here is how that works if you would rather not carry it.