A landscaping firm can sell snow and ice clearing, dormant-season pruning, hedge and tree work, holiday lights, leaf and gutter clearance, and spring planning in the off-season. Test demand first: use AI to segment past customers, draft a short survey, then ask the best-fit group for a small deposit before you buy equipment.
The hard part is not the list of ideas. It is knowing which services your customers will pay for, at a price that covers the crew and the kit. Treat it as an owner's decision: build a menu, rank it, test the top two on people who already trust you, and spend on equipment only after paid deposits come in.
Why off-season planning starts with your own customer list
The customers you served last season are the cheapest audience you will ever have. They know your crew, they have seen your work, and they have a garden or a site you already know. A new service sold to them costs a message; the same service sold to strangers costs advertising.
The temptation is to start with the service you fancy: a plough on the front of a truck, or a van full of lights. The more reliable approach reverses it. Start with what your customers already ask you for in autumn, check whether you could do it with the people and kit you have, and test the answer before committing. AI is useful here for four narrow tasks: sorting your customer list into groups, drafting the survey, wording the offer, and doing the sums on break-even. It is not useful for telling you what the demand is. Only customers can do that.
A winter menu with the economics of each service
Below is a menu that fits most firms. The right-hand columns are what decide whether a service is worth testing: what it needs, how it earns, and how it can fail.
| Service | What you need | How it earns | Main risk |
|---|---|---|---|
| Snow and ice clearing | Shovels and spreaders for small sites; a plough or blower for larger ones; reliable call-out | Per push, seasonal contract or hourly | Feast and famine; storm weeks overload the crew; insurance terms |
| Holiday lights, install and take-down | Ladders, lights, timers, storage space | Fixed price plus take-down, often a repeat customer | Falls from height; a short window of weeks; stock cost |
| Dormant-season pruning | Your existing tools and skills | Per job or hourly | Timing varies by plant; wrong cuts cost you trust |
| Hedge, tree and shrub cut-backs | Existing kit; tree work may need a specialist | Per job | Tree work often needs separate qualifications and cover |
| Leaf, gutter and debris clearance | Blowers, ladders, bags, disposal route | Per visit, sometimes a bundle | Disposal cost; ladder risk |
| Spring planning and design consultations | A laptop, a tape, and a template | Fixed-fee plan; converts to spring work | Plans given away free; no deposit |
| Prepaid spring maintenance contracts | Your normal crew | Deposit now, work later; smooths cash | Customers forget; terms must be clear |
| Equipment servicing and rental for others | A workshop and time | Fixed-price service | Distracts from selling the work that earns more |
Notice what the table does not say: it has no demand figures. Demand depends on your area, your customers and how cold your winters are. That is what the test is for.
The Complete Excel Course with AI Integration
From formulas and pivot tables to Power Query, DAX, VBA and AI with Copilot, ChatGPT and Claude — hands-on from the first lesson.
By Sagnik Bhattacharya and Virginia Thorn, Primetime Emmy Award winner
Buy nowRank the menu on three questions before you survey anyone
Do not test eight services. Rank them first on three questions, and test only the top two or three. The questions: could my current crew do it without new qualifications or kit? Does it fit a slow week, rather than collide with a busy one? And does it lead naturally to spring work?
Illustrative scoring, one point per yes, for a firm with a four-person crew and no plough:
- Pruning: yes, yes, yes. Score 3.
- Leaf and gutter clearance: yes, yes, partly. Score 2.5.
- Spring planning: yes, yes, yes. Score 3.
- Holiday lights: no (needs stock), yes, partly. Score 1.5.
- Snow clearing: no (needs kit), no (storm weeks collide), no. Score 0.
The firm tests pruning and spring planning first. It leaves snow alone this year, but it may return to it with data. If you want to see what a snow service involves when you do want it, how to price snow removal per push, seasonally or hourly covers the pricing, and the snow removal contract template covers the terms.
Split your past customers into groups the assistant can reason about
Take your customer list from last season and tag each customer on a few simple fields in a spreadsheet. Do this in your own spreadsheet, not in an AI chat. Useful fields: type (home or commercial), garden size (small, medium, large), what you did for them (mowing, design, hard landscaping, maintenance), and whether they have hedges or trees. You are creating groups, not profiles.
Then give the assistant counts, never names. An illustrative prompt:
I run a landscaping firm with a four-person crew. These are counts of
last season's customers by group (no personal details):
- 90 home customers with regular lawn maintenance
- 45 home customers with large gardens, hedges or trees
- 60 home customers who only had a one-off design or build
- 25 commercial sites, mostly small offices and flats
I am considering these winter services: pruning, leaf and gutter
clearance, spring planning consultations, holiday lights.
For each group, which two services are most likely to fit, and why?
Give me the reasoning, not a prediction of demand.A realistic reply says that regular maintenance customers are the best fit for pruning and prepaid spring contracts because they already pay on a schedule, that large-garden customers fit hedge and tree cut-backs, that one-off design customers are the natural audience for a spring planning visit because their gardens are maturing, and that commercial sites fit leaf and gutter clearance and perhaps winter safety checks. What you fix: remove anything where it guesses a percentage. If it says "about 30% will say yes", delete the number; it has no basis for it.
The survey that takes thirty seconds
A demand survey fails when it is too long. Three questions is enough: a yes-or-no on a specific service at a specific price range, how soon they would want it, and a free-text box. Compare the two wordings below.
Before: "We are thinking about offering new winter services. Would you be interested? Let us know what you'd like."
After: "Hi [first name], we're planning winter slots for dormant-season pruning of hedges and fruit trees, from about [price range] depending on size. Would you want a visit between now and the end of [month]? Reply YES, NOT THIS YEAR, or tell us what else you'd like us to do."
The second gets answers because it names the service, the price and the deadline, and because one word is enough. The first collects polite non-answers. Ask the assistant to write three variants of the second, one per service, and one for a customer you have not contacted in a year. Check that none promises something you cannot deliver, such as a date before your crew is free.
Send to the best-fit group first, not everyone. If you send to all 220 people with the same message, you cannot tell whether a lack of response means no demand or a bad match.
A test run, worked through with numbers
As an illustration, the four-person firm above tests two services. It sends the pruning message to the 90 maintenance customers and 45 large-garden customers (135 people), and the spring planning message to the 60 one-off customers.
After a week it has 38 replies from the 135: 21 YES, 11 NOT THIS YEAR and 6 with questions. From the 60 planning customers it has 9 replies, 4 YES. The raw rates: pruning 21 of 135, which is 16%; planning 4 of 60, which is 7%.
The first instinct is to read 16% as demand. It is not, yet. A yes to a message is a soft yes. The firm now runs the harder test, and it is the one that matters: it offers the 21 YES customers a booking with a $50 refundable deposit, to be taken off the final bill. Say 13 pay. That is 13 of 21, which is 62% of the yeses and 9.6% of the original 135. That is the real signal.
Now do the economics. If the average pruning job is $240 and takes a two-person crew half a day, 13 jobs is $3,120, which is about 6.5 crew-days. At $180 a day per person, labour is about $2,340 for two people over 6.5 days, so the margin before van, fuel and insurance is around $780, or about $15 for each of the 52 crew-hours (13 half-days of four hours). That will not cover running costs, which is a useful finding before you add the service to a brochure: pruning at $240 for half a day of two people loses money once the van is counted, so the price needs to rise, or the job needs to be a one-person visit. A test that tells you that is worth the week.
The planning test tells a different story. Four yeses out of 60, offered at a $90 fixed-fee planning visit with the fee credited against any spring job, can turn into spring work worth many multiples of the fee. The sample is small, so repeat it before relying on it, but the economics are better because the visit costs one person two hours.
Using a deposit as the real test
A deposit turns an opinion into a decision. Keep it small enough that nobody feels trapped and large enough that people only pay if they mean it: somewhere between $25 and $100 for a job under $300 is usually enough. State clearly that it is refundable until a stated date and credited against the final bill. Read your own terms and make sure the offer does not conflict with them.
An assistant can draft the deposit message, which is the one you must check hardest, because it involves money and promises. Illustrative wording:
Thanks for replying, [first name]. To hold a pruning slot for you before [date], we ask for a $50 deposit, which comes off your final invoice. If you change your mind before [date], we refund it in full. Here is the link to pay: [link]. We'll message you with a day once we have your booking. [Your name]
What to fix: make sure it does not state a day, since you have not planned the crew; make sure the refund date is the same one in your terms; and make sure it does not call a deposit "non-refundable" unless your terms say so and you have checked the rules that apply to your customers. For the prepaid-contract idea, ask a qualified adviser about the terms you use.
A mistake that shows up in week one
A firm decides on holiday lights after a strong survey result. It buys stock, books 18 customers, and on the first day finds that eleven of the properties have no outside power point near the roofline, two have gutters that cannot take clips, and one has a roof the crew is not insured to work on. The survey measured interest; it did not check feasibility.
The fix is a second question for any service involving access or equipment: a short checklist the customer answers before you accept a booking. For lights it might be: is there an outdoor socket, how many storeys is the house, and can you send two photos of the front? The assistant can write the checklist and, if you send photos back to it, summarise them. But a person decides whether a site is feasible, especially at height.
Capacity planning so winter does not collide with something else
A winter service must fit around the work you already have. Before launching, list your crew's real availability in the three coldest months: holidays, training, equipment servicing and any staff you let go for the season. A spreadsheet with four rows (one per person) and thirteen columns (one per week) is enough. Mark each week in one of three colours: free, partly booked, unavailable.
An assistant can then match your tested demand to the free weeks. Illustrative prompt: "I have 13 pruning bookings of about half a day each for two people, 6 spring planning visits of two hours for one person, and these free weeks and days. Suggest a schedule that keeps each crew day full and puts the planning visits on the days with one person free." The output is a draft; check that it respects travel time between sites and that nobody works a day they have booked off. For routing visits by area once you have more than ten bookings, AI route planning for landscaping crews goes into time and fuel savings.
Pricing so that winter work earns its place
The most common failure in off-season work is pricing it to keep the crew busy rather than to make money. A crew-day costing $360 in wages for two people plus van, fuel, tools and insurance must be sold for more than that. A reasonable method: work out your cost per crew-hour, add the margin you want, and compare with the survey's price range. If the survey price is below your cost, the service does not work at that price, however popular it is.
An assistant is good at the arithmetic, if you give it your numbers. Prompt: "Two-person crew, $22 an hour each, van and fuel $18 an hour, tools and insurance $9 an hour. What is my cost per crew-hour, and what hourly price gives a 25% margin?" It should give $71 cost and about $94.67 at 25% margin on price (cost divided by 0.75). Check the maths yourself, because assistants sometimes apply the margin to cost instead of price, which gives $88.75 instead. The difference is $5.92 an hour, which over 100 hours is $592.
For a broader look at not underpricing, how landscapers use AI to stop underpricing jobs goes through the same sums on regular work. And once the winter jobs come in, working out job profit by crew from timesheets shows how to check whether each one paid.
The commercial sites need a different message
Commercial customers, such as small offices, flats and shops, rarely answer a text from a landscaper. The person who decides is often a property manager or a committee, and they buy on risk and paperwork rather than on enthusiasm. A message to them should be a short email with the service, a price range per visit, what the crew will wear and carry, and a line about insurance cover that you can back up with a certificate if asked.
An illustrative opening for a leaf and gutter clearance offer to a small managed block: "We look after the grounds at similar sites and can add a monthly leaf and gutter visit from November to February, from [price range] per visit depending on the number of downpipes. We can send our insurance certificate and a method statement for work at height." Ask the assistant to shorten it to five lines, then check that every claim in it is one you can prove. For this group, a phone call a week later beats a second email, and a one-month trial visit, priced normally, is a better test than a survey.
Setting a stop rule before you start
Decide in advance what result makes you stop. Without a rule, a service that loses money gets one more month for sentimental reasons. A simple stop rule has two numbers: the minimum number of paid deposits you need before the crew is committed, and the minimum earnings per crew-hour you will accept. For the firm above, that might be ten paid deposits and $40 per crew-hour after wages, enough to cover the $27 an hour of van, fuel, tools and insurance from the pricing sum and leave a profit. The pruning test cleared the first number with 13 deposits and missed the second badly at about $15, so the answer is to reprice and test again, not to launch.
Write the rule in two lines and keep it where you can see it. Ask the assistant to play devil's advocate against your plan once the results are in, listing three reasons the numbers might be misleading, such as a small sample, a mild week, or customers who paid the deposit as a favour. Read the list, discard the ones that do not apply, and act on the rest.
Checking after the winter: did the test predict the result?
When the season ends, compare. For each service you tested, record: the number surveyed, the replies, the yeses, the deposits, the jobs actually done, the revenue and the labour hours. Divide revenue by labour hours to get a figure you can compare with your normal work. If the winter service earns less per crew-hour than a quiet week of regular work would, it is a service for filling gaps, not growing.
Also note the number of repeat bookings and any complaints. Ask the assistant to compare your test predictions with the outcome and draft three questions for next year. The most valuable column is the gap between the number of yeses and the number of paid deposits, because that ratio predicts what a marketing message will turn into next time. If it was 62% this year, plan on that, not on 100%.
After the season, a short review request to winter customers is worth sending. Review requests built from job photos covers how.
If you would rather work through your own customer list with someone, that is what my AI implementation consultation covers; the contact page has the details.
Questions landscapers ask before adding winter services
How many past customers do I need for a demand test?
Around 100 gives a rough signal, and 200 or more is better. With fewer than 100, treat the result as a conversation rather than a measurement. What matters more than the total is the response rate: if fewer than 10 people answer, ask again by phone or text instead of drawing a conclusion.
Should I put customer names into ChatGPT or Claude?
No. Use the assistant for wording, segments and arithmetic, and keep names, addresses and phone numbers in your own system. Give it anonymised counts such as '40 customers with large gardens' rather than the list. Check your assistant plan's data settings before pasting anything personal.
Is a deposit a better test than a survey?
Yes, because people say yes more readily than they pay. A refundable booking deposit, or a signed seasonal contract with a start date, shows real demand. Use the survey to find which two services to test, then use a small deposit offer to test those two.
What if my insurance does not cover a new service?
Ask your insurer before you sell it. Snow clearing, lights installation on ladders and tree work often sit under different terms from ordinary garden maintenance. Get the answer in writing, because a claim in the first week of a new service is the worst time to find a gap.
Can I do this if my winters are mild?
Yes, but the menu changes. Without snow, the demand sits in pruning, leaf and gutter clearance, hedge and tree work, winter planting where the ground allows, and planning and quoting for spring. The demand test works the same way whatever the weather.
Further reads
- Snow Plow Route Optimisation: Plan Pushes and Priority Sites With AI — Plan snow pushes and priority sites once you decide to offer it.
- Snow Removal Software: Per-Push Invoicing and Storm Texts With AI — Per-push invoicing and storm texts for a snow service.
- Best AI for a Landscaping Business: Eight Tools for Quotes, Calls and Crews — Eight tools for quotes, calls and crews in a landscaping firm.
- Get Landscaping Clients: Project Photos, Local Search and AI Follow-Up — Fill the spring diary using project photos and search.
- AI Marketing Video Generator: Before-and-After Clips for Landscapers — Turn your best before-and-after jobs into short clips.
- AI in Landscape Maintenance: Contract Renewals, Visit Notes, Upsells — One contract register feeds three jobs: visit notes become client summaries, renewal letters go out early, and aeration or mulch offers fit grass and season.
- Gritting Contracts: Price Winter Call-Outs and Plan Night Runs With AI — Work out what a night run costs, compare three ways to charge, write the trigger clause, order the route and keep a log that protects you.
- How to Price Garden Maintenance: Visit Rates and Seasonal Extras With AI — A six-garden route shows what travel really costs, then builds visit rates, hedge and leaf-clearance extras and a monthly plan from your own times.
- Landscape Business AI Systems: Link Enquiries, Quotes, Crews and Invoices — One workflow map for a five-person landscape firm: seven handoffs between tools, what AI does at each, who checks it, and a four-week start.
- Lawn Care Promotion Ideas: Aeration and Overseeding Offers With AI — Aeration and overseeding offer scripts for existing lawn care customers, segmented by crew notes, with an AI tone check so nothing sounds pushy.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: no vendor prices are used; figures are illustrative sums. Check your own insurer and local rules before adding services.