How Much Should a Contractor Spend on Google Ads?

There is no single right number. Work backwards from the jobs you need, through your close rate, to a monthly budget that actually clears your click price.

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

There is no single right number, and anyone who hands you one without asking about the work you need and the economics of that work is quoting somebody else's budget. A defensible number connects the job gap to your close rate and the local cost of acquiring a lead. Google expresses the result as an average daily budget, but the business decision belongs at the monthly level.

Two things that calculation cannot see will move the answer. There is a floor, set by your click price: budget divided by cost per click is how many clicks you buy, and in some trades a small budget buys too few clicks to produce a lead in a month, never mind a decision. There is a ceiling, set by demand: only so many people search for your service in your area, so past a point more money buys worse traffic rather than more of it. This page covers all three.

Work Backwards From the Jobs You Need, Not Forwards From a Budget

Most contractors start with a number they can stomach and ask whether it is enough. That question has no answer, because "enough" is defined by the work, not by the wallet. The useful inputs are the number of jobs ads are expected to add, the share of paid enquiries that become jobs, and a realistic lead cost for the trade and market. Their relationship matters more than a universal minimum.

Start with the real job gap

Pick the real constraint first. If you want another $180,000 in revenue next year at an average job of $2,500, that is six jobs a month. If instead you have a two-person crew with eight open days, the constraint is capacity, and the job count comes from the calendar.

Then subtract. Repeat customers who call you every spring are not leads you have to buy, and neither are the referrals that arrive without you doing anything. Ads only have to cover the gap between what already walks in the door and what you need. That gap is usually smaller than contractors expect, which is good news, because a smaller gap clears with a smaller budget.

Close rate changes how much demand you have to buy

The same job target can require very different lead volume at different close rates. A business converting half its paid enquiries needs far less traffic than one converting a quarter of them, even when both sell the same service in the same city.

This is where most budget estimates quietly fall apart, because most contractors have never measured the number and reach for a feeling instead. A feeling that is off by fifteen points changes the recommended budget by thousands of dollars a month. If you are going to be approximate about one input in this chain, do not let it be this one. The next section covers where to find the real figure.

Cost per lead anchors the scale

If you have run ads before, your own cost per lead from a representative period is the strongest anchor. If you have not, a published trade figure can establish an order of magnitude until local results replace it. The spread between trades is wide enough that an all-industry average is close to useless, so start from our cost per lead by trade breakdown instead.

Google does not ask for a monthly budget. It asks for an average daily one and uses 30.4 days to connect the two, as Google Ads Help documents. That interface detail should not distract from the more important point: an apparently precise daily number is only as sound as the business inputs behind it.

The Two Numbers Only You Have

Two inputs move the answer more than everything else combined, and neither of them is a marketing number. They are average job value and close rate, they live inside your business, and nobody outside it can supply them. Any budget recommendation that arrives without asking for both is a guess wearing a suit.

What a job is worth to you, after costs

Use gross profit per job, not revenue per job. A $12,000 roof and a $12,000 landscaping build are not the same input if one carries 20% margin and the other 45%. Revenue tells you how big the invoice is. Gross profit tells you how much is still there after materials, labour and subs, and that is the pool an ad budget comes out of.

Work it out for the specific service you plan to advertise, not for the business as a whole. Most contractors sell a mix, and the blended average hides the fact that one service pays for the marketing while another barely covers the truck.

Your close rate, and what it needs to include

For acquisition planning, the denominator needs to include paid enquiries that never reached an estimate as well as the quotes you sent. Otherwise the figure describes sales performance only after qualification while ignoring leads the advertising already paid for.

Published data is a sanity check here and nothing more. Invoca's home services benchmarks report that roughly half of callers to home services businesses reach a person at all, that a bit over a third of calls from digital marketing are genuine leads, and that most businesses never ask the caller to book. Useful for context, useless as your number. Your own CRM beats every published figure on this page.

If you don't know your close rate, that is the first job

The unsatisfying answer is that you have to measure it, and that takes about a month of writing things down. There is no shortcut and no benchmark that substitutes, because the figure varies by two or three times between contractors selling the same service in the same city.

That factor of two or three flows straight through to the budget. Guessing 45% when the truth is 20% does not produce a slightly optimistic plan, it produces a plan that needs more than double the spend to deliver what you promised yourself. Spending a month counting is cheaper than spending a quarter finding out.

What Published Cost Per Lead Data Says by Trade

You still need a cost per lead to finish the calculation, and if you have never advertised you have to borrow one. The most useful public set comes from LocaliQ's home services search benchmarks, drawn from 3,211 US search campaigns running between April 2024 and March 2025, with at least 103 campaigns behind every category. Their broader Google Ads benchmarks for 2026 covers a later window, April 2025 to March 2026, and puts Home and Home Improvement at $8.33 USD per click and $90.92 USD per lead.

Two things to hold onto before you read the table. WordStream and LocaliQ are brands under the same corporate publisher, so their reports are not independent corroboration. The reports use different samples and periods, however, rather than one shared dataset. Their methodology notes also say the averages are technically medians, chosen to keep outliers from distorting the picture, which is worth knowing when you compare yourself to one.

Three rows that show the spread

A five-fold spread is the point

Roofing sits at $228.15 USD per lead and pools sit at $45.15 USD. Same platform, same auction mechanics, same country, five times the cost. Plumbing lands in between at $129.02 USD. The cost-per-lead guide carries the complete 16-trade table and its source caveats.

That spread is the strongest argument on this page that there is no universal budget answer. A "you need at least $X a month" rule quoted without a trade attached is a rule about somebody else's market. A budget that produces steady pool enquiries would produce a trickle in roofing. The trade you are in moves the number more than any tactic will.

Why these columns don't tie out, and how to use them anyway

If you multiply and divide across a row you will not reproduce the published cost per lead, and you should hear that from us before you find it yourself. Take the 2026 Home and Home Improvement figures: $8.33 USD per click divided by an 8.05% conversion rate implies $103.48 USD per lead, against a published $90.92 USD. That is not an error. Each column is an independently calculated median across thousands of campaigns, and medians do not survive being multiplied together.

So use these numbers for what they are good for. They tell you whether your own cost per lead sits in a sane neighbourhood, and what order of magnitude to plan around before you have data of your own. They do not set your budget, and being under a benchmark is not automatically good news. Cheap leads from the wrong searches lose money faster than expensive leads from the right ones.

Cost Per Job Is the Number That Decides, Not Cost Per Lead

Cost per lead is the number everyone quotes and the wrong one to decide on. A lead is not a job. What matters is the acquisition cost of a booked job after the paid-lead close rate is accounted for.

Add close rate and the picture changes

At $130 per lead and a 20% close rate, a booked job costs $650. At the same $130 and a 50% close rate, it costs $260. Identical ads, identical spend, identical cost per lead, and a business that is either comfortable or underwater depending entirely on what happens after the phone rings.

It is also why two contractors in the same trade can look at the same cost per lead and reach opposite conclusions, both honestly. Neither is wrong about the ads. They are running different businesses on the other side of the click.

Compare it to gross profit, not to revenue

Once you have cost per booked job, hold it against gross profit per job. The published $228.15 USD roofing lead sounds expensive right up until you notice it is buying a shot at a job worth several thousand dollars in margin. The published $46.99 USD cleaning lead sounds cheap right up until you work out what one cleaning job actually leaves behind.

Revenue is the wrong comparison because it flatters everything. A $12,000 job with $1,500 of gross profit cannot fund the same acquisition cost as a $12,000 job with $5,000, and the invoice looks identical from the outside.

The calculation is allowed to come back "no"

Sometimes the arithmetic says do not run this. If leads needed multiplied by cost per lead is more than the resulting jobs are worth in gross profit, the answer is not a smaller version of the same campaign. It is a different service, a better close rate, or a different channel.

Better to know that before you spend than after. A campaign funded past what the work is worth does not fail loudly, it fails slowly, and it usually gets blamed on the platform rather than on the arithmetic. The longer version of this test is in are Google Ads worth it for contractors.

The comparison can still come back no. A low-ticket service with thin gross profit may fail even with inexpensive leads, while a higher-cost lead can be comfortable against a high-margin replacement. That relationship, rather than a generic "good CPL," is the decision.

The Floor Is Set by Your Click Price, Not by a Rule of Thumb

Everything so far has been about what you want the campaign to produce. This is about what the auction will let you buy. The two do not always meet, and when they do not, no amount of careful planning fixes it.

Click price determines how much evidence the budget buys

An expensive auction buys fewer visits from the same budget, and fewer visits mean a slower, noisier read on whether the campaign works. Nothing else in the machine can conjure a lead out of a click you did not pay for.

That is why a fixed dollar minimum, quoted with no click price attached, tells you nothing. Fifteen hundred dollars is a serious budget in a trade where clicks cost four dollars and a rounding error in one where they cost thirty. The question is never "is this enough money," it is "how many clicks does this money buy in my auction, and is that enough clicks to produce a lead worth having." I go deeper on the click side of this in how much Google Ads cost for contractors.

What clicks cost in Ontario right now

You do not have to guess at your click price, and you do not have to trust anyone's blog post about it. Google publishes estimated bid ranges for keywords and locations through Keyword Planner, which becomes available after you complete Google Ads account and billing setup. Here is what it returned for a set of contractor search terms targeted at Ontario.

The range matters more than any single figure in it. Window replacement runs from twelve dollars to fifty for the same search term, because what you actually pay depends on the auction and your ad quality. Use the range for scenario planning rather than assuming the low end will be your click price.

The same budget is workable in one trade and useless in another

The Ontario ranges above show why the same monthly amount can create a useful test in one trade and little more than noise in another. A budget buying expensive window-replacement or emergency-plumbing traffic accumulates evidence far more slowly than the same amount in a lower-cost auction. The lead rate then widens that gap again.

A Budget Too Small to Learn From Is Worse Than Not Advertising

There is a version of this argument that agencies use to talk people into spending more, and it deserves the suspicion it gets. This is not that. The point is narrower and it is mechanical: below a certain volume, nobody can tell you whether the campaign is working, including the people running it.

Automated bidding is a prediction system, and predictions need data

Conversion-focused Smart Bidding strategies decide what to pay in each auction partly by predicting conversion likelihood from the signals available to the account. With little conversion history, performance takes longer to evaluate. Our bid strategy guide covers which strategy suits which stage.

Google is fairly direct about the volume it wants for evaluation. Its own Target CPA documentation says: "For evaluation, we recommend you measure performance for the last 30 days, including at least 30 conversions."

The same page also says, and it would be dishonest to quote one without the other, that "Advertisers can start using Target CPA with no conversion history, and Target CPA is effective for campaigns of all sizes." Both are true. You can start small. You just cannot evaluate small, and evaluation is what you are paying an ad budget to buy in the early months.

Readable evidence costs more in some trades than others

The published trade data implies a wide difference in what it costs to accumulate the same amount of evidence. Roofing sits at the expensive end and cleaning near the other, but combining independently calculated median columns does not produce a trustworthy budget quote. The defensible conclusion is narrower: lower conversion volume takes longer to evaluate, and higher-cost trades pay more for that learning.

At low volume, zero leads is a normal month

This is the part contractors are almost never told, and it costs them real money. When a campaign gets a handful of clicks a week, a month with no leads at all is an ordinary outcome for a perfectly healthy account. Not a warning sign. Ordinary. And a month with two leads is not proof anything works either.

Both readings get acted on. The zero month gets the campaign paused a fortnight before it would have produced something, and the two-lead month gets the budget doubled on the strength of a coin flip. Neither decision had any information behind it.

The practical consequence is that a thin budget has to be judged over a much longer window before it says anything, and those extra months of unreadable data are a genuine cost that belongs in the budget conversation rather than being discovered later. How long Google Ads take to work goes further into what that timeline really looks like.

If the Floor Is Above Your Budget, Shrink the Scope Instead

When the available budget cannot support the full plan, spreading it thinly across every service and city leaves each part with too little evidence to judge. A narrower service mix, geography and campaign structure can make the same budget more interpretable, but the exact choice depends on margin, demand and capacity rather than a universal order of operations.

Fewer services, not less money per service

The strongest service is not automatically the one with the largest invoice. Margin, local demand, operational capacity and lead quality all belong in the choice. The important principle is concentration: funding several thin service lines can leave no reliable evidence about any of them.

This feels like leaving money on the table and it is the opposite. Covering all four at a quarter of a workable budget produces four campaigns that each fail quietly and no evidence about any of them. One funded service produces leads, a real conversion rate, and a number you can use to argue for the next one. Our Google Ads for small business guide walks through the same logic for smaller operators.

A tighter service area beats a wider thin one

Same principle applied to geography. Three postal areas you can reach in twenty minutes will beat the whole region at a third of the click volume, and the leads are better besides, because the drive time is short and you can quote faster than whoever is coming from an hour away.

There is a second benefit that is easy to miss. A tight area concentrates your spend on fewer auctions, so you appear more often to the people who are actually in it. Wide and thin means you are invisible everywhere in equal measure.

One campaign, not five

Segmentation is genuinely useful. Separate campaigns let you set separate budgets, see performance cleanly and stop one service eating another's money. All of that is real, and none of it helps when the budget cannot fill one campaign, let alone five.

Treat structure as something budget buys rather than something that substitutes for it. A constrained account often benefits from consolidation; a larger or more diverse one may need separation to prevent one service from consuming another's budget. A painting contractor is generally better off being meaningfully present for a valuable service area than being a rumour across the whole region.

The Ceiling Is Set by Demand in Your Service Area

The floor is about whether you can afford to be in the auction. The ceiling is about whether there is enough auction to be in. Only so many people search for your service in your area each month, and past a certain point more budget stops buying more of the same traffic and starts buying worse traffic.

How to size local demand for free

Keyword Planner does this too. Set the location to your actual service area, not the province, and enter your core service terms. What comes back is an estimate of how many people search those terms in that area each month.

Then read Google's own warning about what you are looking at. Its Keyword Planner forecasts documentation says plainly: "You can expect results to be less accurate if you're targeting a small geographic location, simply because there is less data on which to base predictions." That is the vendor telling you the forecast is least reliable exactly where a contractor needs it most. Treat the number as an order of magnitude, not a figure to plan against. Sizing a market inside a dense area like the GTA is a different exercise from sizing a town of eight thousand, and the tool is far more useful for the first.

Impression share tells you which wall you're against

Once you are running, Google reports how much of the available auction you are actually capturing, and why you are missing the rest. This is the honest answer to "should I spend more," and the reason it is honest is that it distinguishes two problems that call for opposite responses.

The trap is treating lost impression share to budget as an instruction. It is a prompt to look, nothing more. If the campaign is not exhausting the budget it already has, that number is telling you about auctions you would not have won anyway.

The next lead costs more than your average lead

Scaling is not linear. Extra budget reaches additional auctions whose search mix, competition, and intent can differ from the auctions already producing your average.

So when you consider raising the budget, the right question is not only what your average lead costs today. It is what the next block of leads is likely to cost. As a campaign reaches further beyond its strongest current demand, marginal cost often rises. A campaign at $130 a lead should not be assumed to double at the same $130 when its budget doubles.

There is a branch here that almost nobody publishes, so here it is. In a small enough market, the total monthly search volume for your service sits below what a readable campaign needs, no matter what you spend. That is a legitimate finding rather than a failure, and knowing it is worth more than another quarter of trying harder.

What Actually Happens to Your Money After You Set the Number

You put a daily number in a box and Google spends it in ways that surprise most people the first month. None of it is hidden, but almost nobody reads the documentation, and the misunderstandings cost real money and a lot of unnecessary panic.

The daily budget is an average, not a cap

Google's daily spending limits documentation is explicit about both limits. Monthly: 30.4 times your average daily budget. Daily: "For most campaigns, the daily spending limit is your average daily budget multiplied by 2. On a given day, your campaign might spend up to twice your average daily budget to take advantage of fluctuations of traffic."

So a $35 daily budget can produce a $70 day, and that is the system working correctly rather than a billing error. Google is deliberate about the unevenness: it says it "will optimize your campaign spend for days of the month when you're more likely to get clicks and conversions," which means some days come in under and others come in over. There is also a genuine protection in there. If served costs run past the limit, Google covers the excess, and you are never billed above your spending limit.

The practical instruction is short. Check spend against the month, not the day. Watching a daily figure and reacting to it is one of the more common Google Ads mistakes contractors make, and it usually ends with someone lowering a budget on a Tuesday because Monday was busy.

Running ads fewer days does not save money

This one is genuinely counterintuitive and it costs people money every month. Restricting your ads to weekdays, or to business hours, does not reduce what you spend. Google says so directly on the same page: "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."

Read that again with a Monday to Friday campaign in mind. The monthly total is unchanged, but it now has about 21 serving days to fit into instead of 30, so the campaign runs at roughly 1.4 times your stated daily figure on every active day. You have not saved anything. You have concentrated the same money into fewer days.

Which is fine, if that is what you wanted. Ad scheduling is a good tool for lead quality and phone coverage, because a lead that arrives when somebody can answer is worth more than one that goes to voicemail. It is not a tool for spending less. If you want to spend less, lower the budget.

Limited by budget is a prompt, not an instruction

Sooner or later a campaign gets flagged as limited by budget, usually with a helpful suggestion attached. Google's own explanation of the label says there are two cases that trigger it: the campaign is underperforming because of a limited budget, or the campaign uses Maximize Clicks bidding and could get more traffic with more money.

Note what neither case establishes. It does not say the extra traffic would be profitable, and it does not say the money would come back. The same page also notes that Google will not show a recommended daily budget at all if you rarely meet the budget you already have, or if the campaign has limited data, which is a useful tell in itself.

Our reading, not Google's: treat the label as a reason to open the account, not as an instruction to act. Check whether the campaign is actually exhausting what it has, and whether the leads you are already buying are clearing your cost per booked job test. If both hold, more budget is a reasonable move. If either fails, the label is pointing at auctions you were better off missing.

One Number for the Whole Year Is the Wrong Answer

You now have a monthly figure. It has a shelf life. Most of what feeds the calculation moves through the year, and a budget set once in February is quietly wrong by June in both directions at different times.

Seasonality moves the right number in both directions

Almost every trade has a season. Furnaces sell in October and November, air conditioning in June and July, roofing in the dry months, landscaping the moment the frost lifts. Search demand follows the same curve, and so do click prices, because your competitors are all bidding harder in the same weeks.

A flat monthly budget can be underfunded when demand is there to capture and overfunded when it is not. Plan the year as a shape rather than a line, even a rough one: heavier through the season and lighter through quieter months, with the actual curve checked against demand and capacity.

Budget has to match who is answering the phone

Spend that produces more leads than you can quote is not growth, it is waste with extra admin. The reverse is worse and much more common. An unanswered call is billed exactly like an answered one, so every hour nobody picks up makes your true cost per lead higher than the figure in the dashboard, without the dashboard ever showing it.

This is the reason the answer rate context from Invoca, referenced earlier, matters more than most benchmark figures. If a meaningful share of calls to home services businesses never reach a person, and your ads are running at times when yours will not either, you are funding a leak. Match your ad schedule to your actual coverage before you raise the budget, and if calls are going missing, why your Google Ads are not getting calls covers where they usually go.

Changing the number too often has its own cost

There is a real penalty for restlessness. Significant budget and bid strategy changes can trigger a new learning or recalibration period, and during that stretch short-term performance is harder to interpret. Repeated changes can keep moving the baseline before enough data accumulates to judge the previous one.

So set the budget at what you can fund through a full evaluation cycle, not at what you can afford in the best month you have had this year. A smaller number you can hold for six months is worth more than a larger one you will pull in week three, because only the first one ever produces an answer.

Why the Number Doesn't Stay Solved

The arithmetic on this page is not hard. You can do all of it in fifteen minutes with a calculator and last year's quotes. What makes this a job rather than a task is that none of the inputs hold still, and the answer has to be rebuilt against live data rather than remembered.

Every input in the calculation drifts

Click prices climb as more contractors advertise. The same publisher's benchmark set shows Home and Home Improvement clicks rising between consecutive editions, from $7.85 to $8.33 USD, and its own long view is blunter still: "CPC is over twice what it was 10 years ago ($2.32 versus $5.42)." Nothing about your business has to change for your cost per lead to get worse.

Conversion rates move too, with the landing page, the offer, the season and the competition beside you in the results. Close rates move with who is doing the quoting. Every one of those sits inside the chain from jobs needed to monthly spend, so every one of them changes the correct budget when it moves. Doing the calculation once and treating the output as settled is how contractors end up funding last year's answer.

The gap between a working and a non-working account at the same budget

Two contractors in the same trade and the same city, spending the identical amount, routinely get results that are nothing alike. The budget was never the variable. What separates them is a long run of ongoing decisions about where that money lands.

Which searches get bought and which get blocked. What the ad says to the person who typed the search. What the page does with the click once it arrives. Whether the conversions being counted are actually leads. None of that is exotic knowledge, and all of it needs revisiting continuously, because search behaviour changes and the account drifts if nobody is watching. If you already have someone doing this, is your Google Ads agency doing a good job is a fair way to check.

Structure eats budget, so this is often the wrong question

Here is the unsatisfying part. More money on a badly built account buys proportionally more waste, and the waste scales just as reliably as the results would have.

The sharpest version: if conversions are not being tracked, none of the inputs in the first section of this page exist. You cannot know your cost per lead, so you cannot know your cost per booked job, so there is nothing to compare to gross profit and no way to tell whether the budget is right at any level. In that situation the correct answer to how much should I spend is fix the measurement first, then ask again. It is not the answer anyone wants, and it is usually the one that saves the most money.

The inputs a defensible budget needs

A defensible budget needs your real business inputs, not a universal spend floor: the work you need, the share of paid enquiries that become jobs, gross profit on the service being advertised, and the cost of reaching demand in your area. Missing any one of those turns a calculator into a confident guess.

If the available budget spreads those inputs too thin to evaluate, narrow the service or geography before deciding the channel failed. If you would rather have someone assess the numbers against your market and tell you whether the case holds, that is a conversation worth having.

Common Questions About Google Ads Budgets