How to Get Landscaping Leads (And More Hardscaping Jobs)

The problem with landscaping ads is not lead volume, it is lead mix. How to buy design and build work instead of mowing calls, and what controls it.

By Gavin Sevastian · Updated August 4, 2026 · 12 min read

The ads are running. The phone is ringing. It is the wrong phone call every time.

If you are already advertising, the complaint is rarely that the leads stopped. It is that every one of them is a weekly cut, a spring cleanup or somebody asking what you charge to mow, while the patio and retaining wall work you built the business around is nowhere.

Most advice on how to get landscaping leads answers the volume question. This one answers the mix question, and the claim it makes is narrow. You cannot buy job size. Nothing you type into a keyword list makes a homeowner's project bigger. You can buy job type, and there is a short list of things that control which type shows up.

No magic keyword. Just the mechanism, and what it costs you to hold it in place.

How to get landscaping leads that are not mowing calls

If you read one section, read this one.

1. 1

   Split the campaigns and the budgets by service line

   Maintenance in one campaign with its own budget. Design, build and hardscape in another with its own budget. This is the boundary spend cannot cross.

2. 2

   Buy the service words, not the scope adjectives

   Installation, contractor, company, replacement, cost to install. Those are the words people use when they are hiring. Luxury, high end and custom are the words people use when they are looking at pictures.

3. 3

   Build the maintenance exclusion list by hand

   Negative keywords match literally. Every variant of every word you want blocked has to be typed in on purpose, and the list is never finished.

4. 4

   Send each service line to its own page

   A patio inquiry and a mowing inquiry should not land on the same page, and neither of them should land on your homepage. The page is where scope gets qualified, not the keyword.

5. 5

   Send closed job values back into the platform

   Record what each lead turned into and report the value back. This is what teaches the bidding system which searches produce real projects.

Split the budget before you touch anything else

One campaign covering mowing and patios drifts toward mowing. Not because anyone decided that, but because a maintenance inquiry is cheaper and easier to produce than a build inquiry, and a budget-capped campaign that is optimizing for conversion count goes and gets the cheap one. Money in one pot gets spent on whatever is cheapest in that pot.

Two pots, two budgets, and the spend cannot cross the line. That is a structural fix rather than an organizational preference. If you are running the account yourself, get the structure right before you try anything clever, because most of the value in this trade sits there and not in tactics.

Send job value back, or none of the rest holds

Keywords, negatives, budgets and page copy are all filters applied before the auction. They control what you are eligible to show for. None of them tells the system that one inquiry was worth many times another.

Only conversion value does that, and it means knowing which lead became which job and reporting it on an ongoing basis. That is the genuinely hard part of this trade, and it is the part most accounts never start.

Why landscaping leads are harder to control than most trades

A roofer sells roofs. A furnace company sells furnaces. Landscaping is two businesses that happen to share a trade name, and almost every problem with landscaping leads comes back to that.

A weekly cut and a backyard build are the same trade on paper only

The maintenance side is recurring revenue with small tickets, short sales cycles and capacity limited by route density and drive time. The build side is project revenue with large tickets, a sales cycle measured in weeks or months, and capacity limited by how many crew slots exist in a season. Lawn care leads are cheap, quick to settle and repeat all season. A hardscape lead is dear, slow and singular.

Two sets of economics, one brand, and usually one budget.

None of that is an argument for dropping maintenance. The recurring line is the predictable base that funds the season, keeps crews working in the shoulder months and builds the review base the build side later needs. The goal is a mix you chose, rather than the mix the auction handed you.

A blended cost per lead hides the entire problem

If you report one cost per lead across both service lines, you are reporting a weighted average of two things that should never have been averaged. It moves month to month when nothing about your performance changed, purely because the mix of searches changed.

Worse, it flatters the wrong side. The maintenance line looks efficient because those leads are cheap. The build line looks expensive because those leads are dear. On a margin basis that reading is backwards, and it is a common reason a contractor switches off the campaign that was working.

Almost anyone can call themselves a landscaper here

Trade certification for landscape horticulturists in Ontario is available but voluntary, which means nothing stops an operator with a truck and a trailer from competing with you on the same search.

That has two consequences you feel directly. The auction fills with operators carrying no overhead, no insurance and no payroll, which drags the apparent going rate down. And trust has to be demonstrated on your page rather than assumed, because a homeowner has no licence number to look up the way they would for a regulated trade.

One exception is real. Applying chemical fertilizers, fungicides, herbicides and pesticides may require a provincial licence, so landscaping is not unregulated and you should not tell people it is. If you hold that licence, name it rather than burying it inside a generic line about being licensed and insured.

Because there is no general licence, third-party credentials do real work. Depending on the membership category, Landscape Ontario can require references from two current members, liability insurance, WSIB coverage, job site photos and board accreditation. The entry-level category does not carry all of that, so membership on its own is not proof of every item on the list. If you are a member, say which category you hold and what it took to get there.

Buy the job type, not the job size

Open a keyword tool for the first time and the same idea occurs to nearly everyone running Google Ads for landscapers. It is the most intuitive move in this trade and it is the wrong one.

Why luxury, high end and custom bring browsers

Scope and premium adjectives describe aspiration, not budget. Anyone can type them. A homeowner with a modest budget and a Pinterest board types luxury backyard design as readily as somebody about to sign for a full rebuild.

Nothing in a search reports a bank balance. The adjective tells you what somebody wants their yard to look like. It does not tell you whether they are ready to pay for it.

The words that signal hiring

What a searcher does reliably state is the service they want and the role they want filled. Installation. Contractor. Company. Replacement. Cost to install. Near me, attached to a specific service rather than a mood.

The role noun and the service noun carry the intent. The adjective in front of them does not. Retaining wall contractor is a person looking for somebody to build a retaining wall. Beautiful retaining wall ideas is a person looking at photographs. Same subject, completely different customer, and the difference is visible in the words themselves.

What other advertisers pay tells you more than the words do

You do not have to take that on faith. Keyword Planner publishes a top of page bid range for every term, and those ranges are set by advertisers spending their own money against real intent. Hiring language and inspiration language do not carry the same price, and the gap is visible to anyone who looks at both.

That tells you which words buyers use. It does not tell you how big their project is, and nothing else in a keyword tool does either. It helps to know roughly what those clicks cost before you commit to either group.

Separate budgets are the only hard control you have

Budget separation looks like housekeeping and works like a wall.

What a conversion count strategy is told to do

When you run an automated strategy aimed at conversion volume, you are instructing the system to produce as many conversions as your budget allows. That is the whole instruction. It is not told that some conversions are worth more than others, because a standard lead form communicates no value at all.

Which means the cheaper, more frequent conversion is the one it is defined to prefer. In landscaping that is the maintenance inquiry: more of those searches exist, they cost less to win, and the person searching decides faster.

One budget means the build work never gets the money

Put both service lines in one campaign and you have created a contest your build work cannot win. The budget is finite, the system reallocates toward what performs on the metric it was given, and the metric is a count.

This is why adding hardscaping keywords can change nothing at all. The keywords were eligible. They never got the money, because every hour of every day they were losing an internal contest inside the same budget.

What separation looks like in practice

Separate campaigns, each with its own daily budget, so spend cannot physically cross the boundary. Separate ad groups by service inside each one. Separate landing pages. Then set the bid strategy for each campaign against what that campaign is for, rather than running one setting across two businesses.

Say this out loud once, because it is the part people resist. The build campaign will show a worse cost per lead than the maintenance campaign, forever, and that is correct. You are not trying to make those two numbers match. You are trying to stop one of them eating the other's budget.

Local Services Ads in Canada cannot bring you hardscaping work

If you are advertising in Canada and you want more design and build work, Google's Local Services product cannot deliver it. The reason is structural, and it comes down to what a Canadian landscaper is allowed to sign up for.

The Canadian category list has lawn care, and no landscaping

Local Services runs off a fixed list of service categories, and the list differs by country. Checked on August 4, 2026, the Canadian list includes lawn care services. It does not include landscaping services. The American and British lists both do.

Google's own definition of lawn care is maintenance work

You might hope lawn care is a loose label a design and build operator could sit under. In its screening documentation for Canada, Google describes lawn care professionals as maintaining lawns, including installation and maintenance, such as weed control, mowing and seeding, among other services.

That last clause is open-ended, and it does not rescue you, because the constraint is not the wording of the definition. It is the category list. Lawn care is the only entry on it you could plausibly select, so lawn care is what you would be buying, and lawn care leads are what it would send you.

One trap catches people out, because the wording invites the mistake. The general prose on that same Canadian page describes home services as including providers like locksmiths, cleaners, painters or landscapers. The word does appear. It describes the kind of business the product serves rather than a category you can select, and a contractor skimming that sentence would reasonably come away believing they are covered.

Where that leaves you, and how it differs in the US

For a Canadian landscaper, Local Services structurally delivers the exact inquiries this guide is about escaping.

It is still a good product for the right business. If you run a maintenance-heavy operation and you want route density, it is a strong channel, and how the product works in general is a fair thing to read before dismissing it. But you cannot point it at a patio. There are no keywords, no ad copy and no landing page inside it. You pick a category and the category picks your work.

A US reader is in a different position and can run under a landscaping category. Even then, expect category-level control and nothing finer. Choosing landscaping instead of lawn care is a much blunter instrument than choosing one search over another.

What a landscaping lead costs, and why published numbers will not fit you

Everybody wants the number. Here is the closest one available, and then the reasons it cannot answer your question.

The best published number covers home improvement as a whole

LocaliQ publishes annual search advertising benchmarks drawn from thousands of its own customers' campaigns across Google Ads and Microsoft Ads. The nearest category to your business is Home & Home Improvement, which reports an average cost per lead of $90.92, an average cost per click of $8.33, an average click-through rate of 6.47% and an average conversion rate of 8.05%. Across all industries it reports $66.69 per lead and $5.42 per click.

There is no landscaping category in that report. Home & Home Improvement blends roofing, HVAC, remodelling and everything else into one line, so it averages trades with very different economics.

The report does not name a country or a currency anywhere. Assume US dollars, and do not convert them into Canadian dollars and treat the result as a target. That is false precision on a number that never had any.

It is also a vendor benchmark built from one company's customer base rather than an independent panel of the market. It is a real, dated, published methodology, which is more than most benchmark numbers can say. It just means you read it as a shape and not as a score.

Why a blended figure is the wrong number for this trade

Suppose somebody did publish a landscaping-specific cost per lead. It still would not help, because it would be an average across the same two businesses your account contains, in whatever proportion that dataset happened to hold.

Two landscaping accounts can run identically well and report completely different costs per lead, purely because one sells more patios. So when you compare yourself to a published figure and come out looking expensive, the likeliest explanation is a different mix rather than worse advertising. That is the moment most owners decide something is broken and cut the thing that was working.

The number to manage

Cost per booked job, worked out separately for each service line.

That figure survives a mix change, and it settles the argument about advertising being expensive without any special pleading. A build lead that costs several times a maintenance lead and turns into a job worth many times more is not an expensive lead. It looks like one only when you compare it to something it is not.

The catch is that you cannot calculate it from anything inside the ad account. It needs you to know what each lead turned into.

Negative keywords are a filter you build by hand

Almost every contractor account gets this wrong in the same way. You believe you are protected, you are not, and the search terms report is where that shows up.

Negatives match literally, keywords expand

Your positive keywords expand. Google reads intent into them and matches synonyms, close variants, implied words and phrasings you never typed. That is the point of modern match types and it mostly works in your favour.

Negative keywords do none of that. They match what you typed and nothing else. No synonyms. No stemming. No automatic bridge between singular and plural. If you block one word, you have blocked exactly one word, and the ten other ways a homeowner might phrase the same request are all still live.

This is why adding hardscaping keywords is not protection. Your keywords are reaching further than you wrote them, and your exclusions are not reaching at all.

Exact match has not been exact for years

Any advice built on the idea that exact match shows your ad only for that precise phrase is unsafe. Match type controls how far your eligibility stretches. It does not give you a literal filter.

What controls what you are buying is the search terms report, read regularly, with new negatives added from what you find. That is recurring work rather than a launch task. Google also does not report every search term, so the report is not a complete inventory of what you paid for.

Blocking a service does not save you money

On a budget-limited campaign, adding negatives generally redirects spend rather than removing it. The budget still gets spent. It just gets spent somewhere else.

That is still worth doing, because where the money goes is the entire question. But the freed budget does not automatically find higher-value searches, and it will happily go and buy more of a different cheap query if one is available. Add negatives, change what you are buying, then check what it bought instead.

If you are paying for traffic that never turns into a call, negatives are one of several things to look at, and rarely the first.

Make the landing page do the qualifying

Search and job are loosely coupled in both directions. Somebody searching a maintenance term sometimes wants a build. Somebody searching an install term sometimes wants a repair. Homeowners search at the level of the trade, not at the level of your job ticket.

Which means the keyword can only get you close. The page has to finish it.

One page per service line, and never the homepage

Sending ad traffic to your homepage is a routing decision, and for a trade with two radically different service lines it collapses the exact distinction the whole strategy depends on. A homepage speaks to everyone, so it speaks to nobody in particular, and the visitor has to go and find the thing they searched for.

One page per service line. The page names the specific service, shows that kind of work, and answers the questions that kind of buyer has. Somebody who clicked a retaining wall search should see retaining walls above the fold, not a slideshow of your whole portfolio.

Permits, retaining walls and drainage sort a buyer from a browser

A homeowner planning a patio, a deck or a wall has no idea that height thresholds and permit lines exist, and in the end it is the property owner who is responsible for complying with the building requirements.

Answer that on the page and two things happen at once. You have been useful to somebody at exactly the moment they need it, and you have filtered your inquiries hard, because a person worried about a retaining wall permit is not the person price-shopping a weekly cut.

Drainage is the same kind of content and an even better sales argument. The City of Toronto lists poor lot grading or drainage among the causes of basement flooding, and states that where flooding results from poor lot drainage on the property, the owner is responsible for repairs and any damage that follows. That is a liability point rather than a permit requirement, so do not present it as one. It is also how you explain why your quote is higher than the one from the crew who will pour on grade and drive away.

None of this is legal advice and all of it varies by municipality. But a page that answers it in plain language for the area you work in does two jobs at once. It ranks for questions nobody else bothered to answer, and it pre-qualifies everyone who reads it.

What to ask on the form, and what asking too much costs you

A short form gets more fills. A longer form gets fewer and better ones. Neither is correct in the abstract, and the useful move is to run them differently. Keep the maintenance form short, because volume and speed are the point there. Ask more on the build page, because you are trying to shift mix and a slower, better inquiry is the whole objective.

Ask what the job is, not what they want to spend. Which service, roughly what area, when they want it done, and a photo if you can get one. The budget question tends to get abandoned or answered dishonestly by people who genuinely do not know yet, and it reads as pushy on a first contact. You will learn the budget on the phone once you have earned the right to ask.

Send job value back, or the system optimizes for mowing

Inside the auction, a form fill is a form fill. The system counts them, notices which searches produce them cheaply, and pushes in that direction, because counting is the instruction it was given.

So the account has no idea that one inquiry was a small cleanup and the next was a full backyard build. It cannot know unless somebody tells it, and nothing tells it by default.

What value feedback asks of you

The loop is easy to describe and tedious to run.

When somebody arrives from an ad and submits a form or calls, the ad platform attaches an identifier to that visit, and your form or your CRM has to store it alongside the lead. Then you record what the lead turned into: quoted and lost, booked small, booked large, never answered. Then that outcome and its value go back to the platform against the original identifier, so the bidding can tell one search apart from another. Then you leave it alone long enough to mean something, because large, infrequent projects take a season to produce enough signal to steer with.

There is a lighter version that needs no upload at all. Track outcomes in a spreadsheet, look at which searches produced booked projects, and adjust budgets and keywords by hand. This does not teach the bidding system anything. It teaches you, which is the next best thing.

Why most landscaping accounts stop here

Because it is admin that never finishes. Capturing the identifier and sending the value back is configuration you do once. Recording what happened to every lead is a habit somebody has to hold, forever, in a business that is busiest exactly when the leads arrive.

That habit is the line between hoping for build work and steering toward it. It is the least glamorous part of landscaping lead generation and the part that changes what the system chases. Anyone who tells you a weekend of setup fixes your lead mix is describing the filters and skipping the only part that discriminates on value.

Measuring it without fooling yourself

Measurement in this trade breaks in specific ways, and each of them has talked somebody into switching off a campaign that was working.

Untracked calls make everything downstream fiction

Any lead that arrives as a phone call is invisible to a conversion count that only counts form fills. If your phone rings more than your form fires, every cost per lead you calculate is wrong in the same direction.

Run a different tracked number for each service line, too. If both lines ring the same phone, you will never know which half of the budget produced the calls, which is the one thing you need to know when you decide where the money goes.

A hardscape inquiry outlives a short attribution window

A mowing inquiry converts in days. Somebody calls, you quote, they say yes, it is done inside a week.

A patio inquiry does not behave that way. First click, browse, a quote, a design revision, a conversation with a spouse, a comparison against two other contractors, and a signature that might land two months after the click that started it. Judge the build line on a short window and you will systematically understate it.

The same logic applies to keywords showing no conversions yet. A high-value, low-frequency service does not produce enough clicks in a few weeks to prove anything either way, and killing those keywords early cuts exactly the work you were trying to grow. If you have run other project trades with long sales cycles, this will be familiar. Landscaping is harder, because the maintenance half of the account converts fast enough to make the build half look broken by comparison.

The season is a real season, not a performance problem

You know the season. The reason to be precise about it is that the trough gets misread as failure every year.

The Government of Canada's Job Bank describes landscape and horticulture technicians in Ontario as a seasonal occupation with better prospects in spring and summer. Note what that describes: employment in the occupation, not consumer demand and not advertising performance. It corroborates what you already live.

So compare like with like. This March against last March, not March against last August. And decide in advance what you are doing in the shoulder months, because a quiet January is not a signal to change anything.

Every channel, ranked against the goal of shifting your mix

Rank the channels by the question that matters here. Does this let you choose what kind of work arrives?

Search is the only channel where you choose the job type

You are buying the query. Somebody typed a specific thing, you decided in advance whether you wanted that thing, and you paid for the click. No other channel works that way.

For a business whose problem is mix rather than volume, that makes search the primary instrument and everything else secondary.

Local Services, Google Business Profile and reviews are volume and trust, not mix

Local Services is category-level, and in Canada that category is lawn care, for the reasons set out above. Nothing inside it points at design work.

Your Google Business Profile and your review base are a different matter. They are the main thing you control in local visibility, and landscaping SEO compounds on top of them. The catch is speed. Reviews move in months, not weeks. If you are trying to change what arrives this season, reviews are the right investment and the wrong tool, and it helps to be clear with yourself about which of those two problems you are solving.

Performance Max, Demand Gen and paid social are portfolio channels, weak on intent

Landscaping is visual, and a real portfolio of finished projects is genuine ammunition on these channels. They are demand creation. They put good work in front of people who were not searching, which is a legitimate way to sell design and build.

They are also the weakest on intent control. You are choosing audiences and letting a system decide who sees what, and neither of those is the same as choosing the search. On a modest budget, if your problem is mix rather than volume, they are a poor first move. Once search is structured properly and the value feedback loop is running, they are a reasonable second.

Referrals belong on that list for one reason. They control neither when work arrives nor what kind, which is usually what sits underneath a request for more leads without ever being said out loud. They are not free either. The cost is real and invisible, so compare them on cost per booked job like everything else.

Lead marketplaces are a different product again from Google's Local Services, with different economics and different rules about whether a lead is sold to you alone or to three of you at once. Ask which it is before you judge the price. Other trades share pieces of all this: anywhere the customer is buying a look rather than a repair, the same tension between browsing and hiring turns up in the keyword list.

The part nobody puts in the guide: this is upkeep, not setup

Most guides about landscaping advertising end by implying you now have a system. You have a starting position, and positions drift.

The work that never finishes

Reading the search terms report and adding new negatives as new phrasings appear. Keeping the exclusion list current across every variant somebody invents. Recording lead outcomes and sending values back for every closed job. Rebalancing budgets between the two service lines as the season turns. Chasing down what happened to the inquiry from five weeks ago so the loop closes.

None of that is one-time. The mix problem is not solved by a setup, it is held in place by maintenance, and when the maintenance stops the account drifts back toward the cheap conversion, because that is what it was always going to do.

Capacity is a lever, and sometimes the right move is to spend less

A design and build crew has a finite number of project slots in a season. Fill them and every extra lead is a cost with no revenue behind it. A maintenance route works differently. Density is the economics, and the extra stop three doors down is close to free margin.

Pay-per-lead channels do not know or care about your calendar. When the build calendar is full and a pay-per-lead channel is still running at full tilt, that spend is buying nothing, and on the dashboard it looks exactly like success. Turning a channel down when you are booked is a legitimate and badly underused move, and it is easier than turning one back up from zero later.

Doing it yourself, hiring in, or hiring out

None of the three is free.

Doing it yourself costs a few hours a week, permanently, in the evenings when the crews are done. That is viable for a lot of operators, particularly on a single service line with a modest budget, and plenty of contractors run their own landscaping marketing perfectly well. What usually goes wrong is that the recurring work quietly stops in June when you are busiest, which is when it matters most.

Hiring somebody in means a salary and the problem of managing a role you cannot yet evaluate. That tends to make sense at a spend level well beyond most contractors' first few years.

Hiring out costs a monthly fee and still requires you to hold up your end of the value feedback loop, because nobody outside your business knows what a lead turned into. Find out what this kind of management typically costs before you talk to anyone.

Where to start

You cannot buy job size. You can buy job type, and the levers are the same four every time. Split the money so the build side has its own budget. Buy service and role words instead of adjectives. Let the page and the form do the qualifying. Send closed job values back so the bidding knows the difference.

The first three are structure and you can finish them. The fourth is a habit, and it will either still be running next spring or will have quietly stopped in June. That is the difference between a landscaping account that drifts and one that goes where you point it.