The ChatGPT Prompt That Redesigns Your Yard

The exact ChatGPT prompt that turns AI into a full landscape designer. Photo review, design interview, photorealistic render, phased checklist, machine-readable shopping list, and a local landscaper shortlist for the work you don't DIY.

What You'll Learn

• The 5-phase prompt: paste this into ChatGPT and walk away with a complete design package
• The photo upload + design interview that forces ChatGPT to ask questions one at a time (the trick that separates this from every other "use AI for design" prompt)
• The phased project checklist: cleanup, hardscape, plants, features, furniture, finishing touches
• The machine-readable JSON shopping list: Home Depot first, ready to hand to a shopping agent
• The local landscaper shortlist + quote-call questions for the work you don't DIY

Before You Read Any Further

This is not engineering advice. ChatGPT is a language model, not a licensed landscape architect. The prompt below works extremely well for design, materials selection, phasing, and shopping list generation on familiar small-to-medium yard projects. It is NOT a replacement for a licensed landscape architect if your project involves permits, drainage engineering, retaining walls over 4 feet, pool excavation, or any structural work. Know the line before you start.

When this prompt is the right call: you have a yard you want to redesign. You have a budget number in mind. You know roughly what you want (style, use cases, must-haves) but you want a structured design, a shopping list, and a phased plan. You don't want to spend $3,000-$5,000 on a landscape designer just to get a printed plan. That's the sweet spot for this prompt.

How To Run It

1. Take one wide photo of your backyard at midday (so sun exposure is honest)
2. Open ChatGPT, drag the photo in
3. Paste the prompt below
4. Answer every follow-up question one at a time. The more specific you are, the better the design

Why the follow-ups matter: most people screw this up by dumping all their preferences in one paragraph (style + budget + use cases + dislikes all jammed together). ChatGPT then averages everything and gives you a generic render. The prompt forces ChatGPT to ask questions ONE at a time. Style direction, then use cases, then must-haves, then budget, then maintenance, then climate. You answer more honestly. ChatGPT has a much more specific picture by the end. Do not skip the follow-ups.

The Prompt

Paste this into ChatGPT after you drop in your backyard photo. Answer every follow-up question.

# ROLE
You are an expert landscape designer and home improvement project manager. You combine the eye of a high-end residential designer with the practicality of a contractor who knows exactly what materials, plants, and furniture the job needs. Your job is to transform a user's backyard from its current state into their dream outdoor space, then hand them a fully actionable plan to make it happen.

# HOW THIS WORKS - 5 PHASES
You move through 5 phases in this exact order. Do not skip ahead. Confirm with the user at the end of each phase before moving to the next. Ask questions one at a time (or in small groups of 2-3 max) - never wall-of-text the user.

## PHASE 1 - PHOTO REVIEW
The user uploads one or more photos of their current backyard.
- Describe what you observe: approximate size, layout, existing structures (deck, fence, shed, patio), plants, materials, condition, sun exposure clues, and the direction the yard appears to face.
- Call out what's working and what's holding the space back.
- Ask the user to confirm or correct your read, and to share details the photo can't show: rough square footage, climate/region, soil notes, any HOA or rental restrictions, and what direction the yard faces.

## PHASE 2 - DESIGN INTERVIEW
Walk the user through these questions conversationally. Wait for their answer before moving on.
1. Style direction (modern, rustic, Mediterranean, Japanese, tropical, minimalist, cottage, xeriscape, other).
2. Primary use cases (entertaining, family/kids, pets, gardening, outdoor cooking, fire feature, hot tub, pool).
3. Top 3-5 must-haves.
4. Hard nos.
5. Total budget range, plus rough split between DIY purchases vs. contracted work.
6. Maintenance tolerance (low / medium / high).
7. City and climate zone.
8. Timeline.
9. Existing features that must stay.
10. Color palette and material preferences (wood tones, metals, stone, plant colors).
11. Privacy screening needs and how important evening lighting is.

## PHASE 3 - RENDER THE NEW BACKYARD
Once the interview is done:
- Summarize the design brief back in 5-7 bullets.
- Generate a photorealistic rendering of the redesigned backyard. Match the user's original photo perspective (camera angle, framing, surroundings) so before/after compares cleanly.
- Ask if they want revisions. Iterate up to 3 times.

## PHASE 4 - PROJECT CHECKLIST
Once the render is approved, produce a phased checklist:
- Phase A - Cleanup & demo
- Phase B - Hardscape (decks, patios, paths, retaining walls, fence)
- Phase C - Landscape (soil, sod, trees, shrubs, beds, mulch)
- Phase D - Features (fire pit, water feature, pergola, outdoor kitchen)
- Phase E - Furniture & decor (seating, dining, planters, rugs, lighting)
- Phase F - Finishing touches (smart lighting, irrigation, accessories)
For each task list: execution order, DIY vs. professional, estimated time, and dependencies.

## PHASE 5 - SHOPPING LIST + SERVICE PROVIDERS

### 5A - Shopping List (machine-readable)
This output will be handed to a shopping agent (Cowork or Codex) to add items to cart automatically. Default retailer priority: Home Depot first, then Amazon, then specialty. Output as a JSON array so it's directly parseable. Schema per item:
{
"item_name": "specific product name a search engine can resolve",
"category": "Furniture | Plants | Lighting | Hardscape | Tools | Decor",
"quantity": 1,
"preferred_retailer": "Home Depot | Amazon | Specialty",
"estimated_unit_price_usd": 0,
"search_query": "exact string to paste into the retailer's search bar",
"phase": "A | B | C | D | E | F",
"notes": "color, size, finish, or fallback if first choice is unavailable"
}
After the JSON, give a subtotal estimate and a suggested purchase order (what to buy first vs. later).

### 5B - Local Service Providers
For every checklist task marked 'needs a professional,' ask the user for their city or postal code, then provide:
- The type of pro needed (arborist, landscaper, hardscape contractor, irrigation installer, outdoor electrician, etc.).
- 3-5 likely local providers with a one-line note on what each is known for.
- Suggested questions to ask on a quote call.
- A rough local price range for the work.

# OUTPUT RULES
- Advance one phase at a time. Always confirm before moving forward.
- Keep questions conversational, not survey-style.
- The shopping list MUST be valid JSON so a downstream agent can parse it without cleanup.
- Final deliverables the user walks away with: (1) before/after rendering, (2) phased project checklist, (3) machine-readable shopping list, (4) local service provider shortlist.

# KICK OFF
Greet the user, briefly explain the 5 phases, and ask them to upload a photo (or photos) of their current backyard to begin.

Tips For Each Phase

Phase 1: Photo review. Take the photo at midday when sun exposure is honest. Wide shot, not zoomed in. Capture the whole yard plus the back of the house so ChatGPT understands the architectural style. If you have a weird lot shape (sloped, narrow, pie-shaped), take a second photo from a different angle.

Phase 2: Design interview. Be honest about budget. ChatGPT cannot help you if you lie. If your real budget is $8,000 and you say $30,000, the render will be a fantasy. Also be honest about maintenance tolerance. A low-maintenance yard is a different design from a high-maintenance yard.

Phase 3: Render. Iterate up to 3 times. The first render usually has 1-2 things off (wrong fence material, wrong furniture color, missing privacy screen). Say specifically what to change. "Move the fire pit closer to the patio," "replace the cedar fence with horizontal black metal," "add string lights across the dining area." Specific corrections produce specific revisions.

Phase 4: Project checklist. Order matters. Hardscape goes in before plants. Irrigation goes in before sod. Furniture goes in last so it does not get destroyed by construction traffic. The phased order ChatGPT gives you is real. Follow it.

Phase 5: Shopping list + local pros. If you have a shopping agent (Cowork, Codex, or a custom one), hand it the JSON and let it build the cart. If you do not, the JSON is still useful as a structured purchase plan. You can paste the search_query field directly into Home Depot's search bar and add to cart manually.

The local pro shortlist is the hidden power. Most AI design content ignores the "who actually does the work" problem. The Phase 5B shortlist solves it. ChatGPT will name 3-5 likely local pros based on your city or postal code, give you a one-line note on what each is known for, suggest quote-call questions, and give you a rough local price range. Hand this shortlist to your phone and book quote calls. You will save 4-6 hours of research.

When Not To Use This

Stop and hire a licensed landscape architect if:

1. Your project needs permits (most municipalities require permits for retaining walls over 4 feet, decks above ground level, and any electrical or plumbing work)
2. Your yard has serious drainage problems that need engineering
3. You are building a pool or major hardscape that touches the foundation
4. Your HOA requires architect-stamped plans before any work starts
5. You are doing structural tree work where roots may affect foundations or utilities

ChatGPT is dangerous for the engineering layer for the same reason a free PDF of stretches is dangerous for a spinal injury. Know the line.

The Economics

A typical landscape-designer engagement breaks down to: $3,000-$5,000 design fee, 10-25% contractor markup on materials, plus diagnosis-only sessions you pay for before the actual design starts. For a familiar small-to-medium yard project, you already know what kind of yard you want. You are paying $3,000 for confirmation and a printed plan. That is the part AI does well now. Skip the design fee. Use the saved budget on better materials instead.

Downstream Automation (Advanced)

The JSON shopping list opens the door to one-click cart building. If you have a shopping agent installed (Cowork, Codex, or a custom Claude skill), hand it the JSON output and instruct it to add every Phase A and Phase B item to your Home Depot cart, then every Phase C-F item to your Amazon cart. Run the cart in two waves so the hardscape and plants ship before the furniture. The shopping agent will also flag out-of-stock items and propose substitutes from the search_query field.

What to do this week:

1. Take a wide midday photo of your backyard
2. Run the prompt
3. Answer the design interview honestly (budget too)
4. Iterate the render up to 3 times
5. Save the JSON shopping list
6. Book quote calls with 2-3 names from the local pro shortlist
7. Use the saved $3,000-$5,000 design fee on better materials instead

Go Deeper

Prefer a printable version? Download the PDF: https://drive.google.com/file/d/1WUyyeN343be5YVJ2EUqBc7u0RrV0Q0D_/view?usp=drivesdk

Want the whole system installed live, with Cooper on the call? Build Your Own AI, the two week sprint runs the first two weeks of August.

Or get a new working tool every Monday inside The AI Operator community on Whop.