TL;DR: Landscaping companies can buy ChatGPT Ads today in OpenAI’s self-serve Ads Manager, with ZIP-code targeting, a $25 daily minimum and a suggested starting bid of $3 to $5 per click. That bid is well under the $8.76 median a landscaping click costs on Google Search in LocaliQ’s benchmarks. The fit is narrow, though. ChatGPT is where homeowners think through a patio, a retaining wall or a new planting plan, not where they find someone to mow on Thursday. Our short verdict: worth a test for design/build work if you can track a lead to a signed contract and your crews have room; skip it for mowing and one-off cleanups. A sensible first test is one patio and wall campaign at the $25 daily minimum for 30 days, started only when your install crews can take new work within about two months.
This is the landscaping version of our ChatGPT Ads guide for home services, which covers sign-up and a general checklist. Every OpenAI setting below was re-checked in OpenAI’s Ads Manager help center on November 2, 2026. The product is still in beta, so expect changes.
Which landscaping jobs fit ChatGPT Ads
A ChatGPT ad shows below an answer and has no call button: OpenAI’s ads basics page lists name, logo, title, description, landing page and image. Every lead reads your page first. That suits a homeowner planning a new backyard more than one who wants the grass cut this week.
| Job | What the customer asks first | Fit for ChatGPT Ads |
|---|---|---|
| Mowing, one-off spring or fall cleanup | “Lawn mowing near me,” then picks whoever can start soonest | Weak. Low ticket, decided on price and availability |
| Season-long maintenance contract | What a full-service plan includes and what it should cost | Fair. Recurring value, worth one small ad group before spring |
| Landscape design and planting | “What can I plant in a shady, wet backyard?” | Good. A plan, not a reflex |
| Patios, retaining walls, outdoor kitchens | Paver vs concrete, cost per square foot, permits on a slope | Best. Big ticket, lots of questions, weeks of planning |
| Commercial maintenance (HOAs, property managers) | Bid lists and contract terms | Weak. See the note below |
This sorting is our reading of how homeowners buy, not survey data. One fact behind the last row: OpenAI does “not show ads to users on Plus, Pro, or any Business plan.” A property manager researching vendors on a company ChatGPT Business account will not see your ad at all. Keep commercial work on your existing bid and referral channels.
Plan around spring and fall, and around crew capacity
OpenAI’s campaign guide tells advertisers to “consider seasonality” when setting start and end dates, and to “use separate campaigns for meaningfully different objectives, regions, or product categories.” For a landscaping company that usually means two campaigns:
| Campaign | When we would run it | Ad groups | Leave on Google and referrals |
|---|---|---|---|
| Design/build | Late winter through early summer, then again in early fall for fall installs and next spring’s projects | Patios and hardscape, retaining walls, design and planting | Small repairs, one-off plantings |
| Maintenance sign-ups | The six to eight weeks before your first mow; end it when routes are full | Full-service maintenance plans | Mowing-only requests, cleanups |
These windows are our suggestion. In Phoenix or Tampa design/build can run most of the year; in Minneapolis the spring window is short and busy. Three settings shape the timing:
- You can’t copy last season’s campaign. Per OpenAI, “you can’t clone a Clicks campaign into another Clicks campaign.” Next spring’s campaign is entered by hand, ads included, so keep your ad text in a document.
- Set dates before you publish. OpenAI warns: “don’t publish the campaign just to access these settings. Your ads could start running before you’ve had a chance to set your schedule.”
- There is no weather trigger. The first warm weekend sends people into their yards and into ChatGPT, but the only lever is the daily budget. It is an average over a Sunday-to-Saturday week, and a single day “will not exceed twice your selected daily budget.”
Watch capacity, not just demand. If your install crews are booked eight weeks out, a lead you pay for in April may not get a start date until summer. Lower the budget or pause; OpenAI notes campaigns “can always be paused or adjusted later.” Turn it back on when the schedule opens.
The numbers to beat: what a landscaping lead costs on Google
Most landscapers who would test ChatGPT Ads already buy Google clicks, so start there. LocaliQ’s home services benchmarks, based on 3,211 US search campaigns that ran from April 2024 to March 2025 and updated July 15, 2026, put landscaping at these medians (LocaliQ home services search ad benchmarks):
| Google Search metric | Landscaping | All home services |
|---|---|---|
| Click-through rate | 4.69% | 6.37% |
| Cost per click | $8.76 | $7.85 |
| Conversion rate | 6.42% | 7.33% |
| Cost per lead | $117.92 | $90.92 |
Landscaping had the lowest click-through rate of any home services category LocaliQ tracked. Caveats: these are medians from one vendor’s client campaigns, and the landscaping row mixes mowing leads with design/build leads, which are worth very different amounts to you. Treat $117.92 as a blended reference, not the price of a patio lead. Google is also moving Local Services Ads into Google Ads, with lawn care in the first group starting August 2026 (Google help). Your own account’s cost per lead for each service beats any benchmark.
What a ChatGPT click has to do
ChatGPT Ads charge per click (or per 1,000 views), so your cost per lead depends on how many clicks call or fill a form. At OpenAI’s suggested bids:
| To match… | At $3 per click | At $5 per click |
|---|---|---|
| $117.92 per lead (landscaping, Google Search) | 1 lead every 39 clicks (about 2.5%) | 1 lead every 24 clicks (about 4.2%) |
That is below the 6.42% conversion rate landscaping Search ads get in the same study, so on paper the bar is reachable. Two cautions: ChatGPT users are earlier in their thinking than someone searching “patio contractor near me,” and the bid you need in your area may be higher than OpenAI’s starting suggestion. For design/build, the number that matters is cost per signed contract, not per lead. Our guide to Google Ads for landscapers covers budgets and structure on the Google side.
Ad groups: one per project type
OpenAI’s ad group guide says to “create separate ad groups when they require different messaging or landing pages,” and gives a plumbing example with separate groups for emergency repairs and preventive maintenance. For landscaping, that means one group per project type:
| What the homeowner asks ChatGPT | Ad group | Landing page |
|---|---|---|
| “How much does a 400 square foot paver patio cost?” | Patios and hardscape | Recent patios with what each cost, your minimum project size |
| “Do I need a retaining wall for a sloped backyard?” | Retaining walls | Wall types you build, when an engineer or permit is needed |
| “Our new-build yard is just builder sod. Where do we start?” | Design and planting | Your design process, design fee, plant choices for local soil |
| “What should a lawn maintenance contract include?” | Maintenance plans | What a visit includes, season pricing, open route days |
These prompts are our examples, not OpenAI data. Start with the one or two groups that carry your best margin. Our post on hardscaping marketing covers how patio and wall buyers compare contractors.
Use the multi-image format for finished work. Since October 26, 2026, OpenAI offers multi-image ads as an open beta: four to six square images that all link to one page (ad creation guide). Before and after photos of your own patios and plantings do more for someone picturing their yard than stock photos.
Context hints and ad copy for landscapers
Each ad group gets a few context hints: short sentences on what you offer, who it suits and when it helps. OpenAI’s context hints guide shows the level of detail with a plumber: “After-hours residential plumbing in [city], with an $89 service-call fee for urgent repairs.” Landscaping versions:
| Ad group | Type | Example context hint |
|---|---|---|
| Patios and hardscape | What | “Paver patios, fire pits and seat walls in Raleigh and Cary, with a 3D design before you sign and projects from $[your minimum]” |
| Design and planting | Who | “Owners of new-build homes with a bare yard, red clay soil and builder-grade sod” |
| Maintenance plans | When | “Homeowners choosing a weekly lawn and bed care service before the spring season starts” |
Use your real towns, services and minimum. ZIP codes go in the campaign’s location setting; context hints “cannot enforce geographic limits, schedules, or exclusions,” so a hint like “only show this in Raleigh” does nothing. When a promotion ends or your schedule fills, update the hint, as OpenAI asks.
How to keep mowing requests out
There is no hard switch for it. The campaign settings cover dates, locations, platforms and audiences, and context hints “cannot enforce geographic limits, schedules, or exclusions.” So a line like “don’t show for mowing” will not block anything. The filtering happens in four places you control: an offer that names the project type, ad copy that states your minimum, a landing page that says plainly you don’t take mowing-only work, and a form that asks for project type and budget before anyone calls back.
Ad titles and descriptions
OpenAI asks for “multiple title/copy variations per offering” and copy that “explains the practical value of what you offer, who it is for, and when it may be helpful.” Our examples:
| Title | Description |
|---|---|
| Paver Patio Prices in Raleigh, From Real Projects | See recent patios we built, what each one cost and how long it took. Patios and walls only, from $[your minimum]. |
| Bare New-Build Yard? Get a Planting Plan | Design and install for new homes in Raleigh and Cary, with plants that handle clay soil and full sun. |
| Weekly Lawn Care, Booked Before Spring | Mowing, edging and bed care on a set day each week, with season pricing in writing. |
Watch three claims. “Free design,” only if it really is free and not credited back later; say which. Financing: OpenAI’s ad policies restrict ads for financial products, so keep payment plans on the landing page, not in the ad. And “licensed, bonded and insured”: in California that line can break the rules, as the next section explains. Ads must be “truthful and not misleading, including unfounded claims about capabilities, pricing, outcomes.”
License numbers and the word "bonded"
OpenAI’s policies require licensed advertisers to “maintain those credentials and comply with applicable laws and regulations.” State rules then apply to the ad itself. California’s Contractors State License Board (CSLB) is the clearest example, and its advertising guidelines cover three points landscapers trip on:
- License number in every ad. Licensed contractors “must include their license number in all construction contracts, subcontracts, and calls for bid, and all forms of advertising,” and the guide counts “any website or post that solicits business on the Internet.” The civil penalty is “between $100 and $1,000 for a first offense.”
- No “bonded.” “The law prevents you from noting in your advertising or any presentation to the public that you are bonded.” Drop it from ad text and landing pages in California.
- Stay inside your classification. “You cannot advertise for construction work outside of areas for which you are licensed.” A landscaping license holder advertising outdoor electrical or gas work should check that it is covered.
Other states have their own rules. Ask your licensing board how they apply to a short ChatGPT ad, and show your license number on every landing page. This is not legal advice.
The landing page does the selling
The ad can’t place a call, so the page has to. On iPhone, the ChatGPT app opens the click in an in-app browser. What we would put on a patio or design page:
- A tap-to-call number at the top, a tracking number used only on ChatGPT pages.
- The project in the headline, matching the ad: “Paver patios in Raleigh, with real project prices.”
- Three to six finished projects with size, materials, price range and how long each took.
- Your minimum project size and current lead time, stated plainly. It turns away leads you can’t serve before you pay for the call.
- A short form (project type, rough budget, timeline, ZIP code) and your license number.
OpenAI asks for “product pages, collections, or content pages rather than defaulting to a generic homepage,” and landing pages “must not block OpenAI user agents (OAI-AdsBot and OAI-SearchBot).” Our guide to landscaping landing pages covers the layout.
Count signed contracts, not clicks
A patio lead can take weeks to become a signed contract, so set up tracking before you spend:
- UTM parameters on every ad link; OpenAI says they “persist on ad clicks.”
- A dedicated call tracking number on ChatGPT landing pages.
- The OpenAI Pixel on your forms, with the standard “lead_created” event (developer docs).
- Signed contracts sent back if you can, through the Conversions API with the “oppref” click reference (measurement guide). Our guide to offline conversion tracking for contractors covers the same idea for Google and Meta.
- Exclude current maintenance customers. Inclusion audiences need 25,000 matched users, but “exclusions can use smaller audiences.” Upload your customer list as an exclusion; not every customer will match, but each one who does is a click you don’t pay for.
OpenAI and your own records “may report different conversion totals,” so pick one scorecard: signed contracts and their value in your CRM or estimating software, with a lead source such as “ChatGPT Ads: patios” set on the first call. The weak spot is the handoff. The source gets logged on the call, then lost between the site visit, the design and the signed proposal. Check that it carries through, or you will never know what a ChatGPT patio cost you.
A first-season test, judged against your Google numbers
OpenAI has not published landscaping results, and 150 to 250 clicks give a direction, not a verdict. A simple rule: if the calls are real homeowners in your area with projects above your minimum and site visits are being booked, extend to a spend cap you set in advance. If most calls are mowing requests, out of your area or below your minimum, stop and fix the hints and page before spending more.
Should your landscaping company test ChatGPT Ads this season?
| Test now if… | Wait if… |
|---|---|
| You know what a design/build lead and a signed project cost you on Google | You can’t say what any Google lead costs |
| Hardscape, design or install work is a big share of revenue | Most revenue is mowing and cleanups |
| You have project pages with real photos and price ranges | Every ad would go to your homepage |
| Your crews have room in the months the ads would fill | You are already booked through the season |
On the “wait” side? Fix Google first; our landscaping PPC and marketing services page explains how Leads4Build runs Search and Local Services Ads for landscaping companies. Angi, the home services marketplace in OpenAI’s own case study, tested ChatGPT Ads across “roughly 20 home service categories, including roofing, handyman, and HVAC” (OpenAI customer story); the story does not say whether lawn care or landscaping was one of them.
The bottom line
For landscapers, ChatGPT Ads are a test for design/build work, not a source of mowing customers. Run a design/build campaign in your planning seasons, give each project type its own ad group and page, state your minimum, pause when crews are full and judge cost per signed contract against your Google numbers. For the general setup see our ChatGPT Ads guide for home services.
Plus one free tool from us, unrelated to ads: we built an AI social media post generator for landscaping companies that writes seasonal, tip and project posts for Facebook and Instagram. It is free and one of 21 trade generators on our tools page.
About the sources: OpenAI settings come from its Ads Manager help center (campaigns, context hints) and ad policies, re-checked November 2, 2026. Landscaping Google Search figures are LocaliQ’s published benchmarks (LocaliQ). Other sources are linked where each fact appears.