Automate real estate listing creation from inspection photos to virtual tour to social posts

A real estate agent sits down on Monday morning. She has five property inspections completed over the weekend, and each one needs a full listing page, a virtual walkthrough video, carousel posts for Instagram and TikTok, and professionally edited photos. By hand, this takes her roughly 12 hours. By Thursday, she's still not done, and she's already got three more properties waiting. The work is identical each time: extract data, write copy, format images, create video, post everywhere. Nothing about this requires her judgment or expertise. Everything about this could be automated. Most agents accept this as the cost of doing business. They've built their workflow around tools designed for manual work, and they've never seen how much time actually disappears when you remove the handoffs. If you're willing to spend an afternoon setting up automation, you can cut this entire process down to minutes. Every new property walks through the same pipeline: photos in, professional listing and promotional assets out. This is where combining the right tools matters. Pixelcut AI handles photo enhancement and background removal, Dora AI generates custom listing websites, Hour One creates video presenters from property descriptions, and Mirra turns those same descriptions into social media content. Wire them together with an orchestration platform, and you've eliminated the manual steps entirely.

The Automated Workflow

For this workflow, we'll use n8n as the orchestration layer. n8n gives you fine-grained control over data transformation and can handle the conditional logic needed when managing multiple output formats. You could use Make or Zapier if you prefer a simpler interface, but n8n scales better as you add more properties and outputs. The workflow starts with a webhook that receives inspection data. Your mobile app or form submission sends property details: address, property type, bedroom count, square footage, and a folder link containing 15-20 inspection photos.

Step 1: Photo Processing with Pixelcut AI

Once the webhook fires, the first node calls Pixelcut's API to process your inspection photos. Pixelcut removes distracting backgrounds, enhances lighting, and ensures consistent quality across all images.

CopyPOST https://api.pixelcut.ai/v1/batch-process
{ "input_source": { "type": "folder_url", "url": "s3://your-bucket/inspection-123" }, "operations": [ { "type": "background_removal", "fallback_color": "#ffffff" }, { "type": "auto_enhance", "brightness": 0.15, "contrast": 0.1 } ], "output_format": "jpg", "output_destination": "s3://your-bucket/processed-123"
}

n8n stores the output folder URL in a variable for use in later steps.

Step 2: Generate Property Description with Claude

Before creating any assets, you need coherent copy. Pass the inspection data (address, property specs, and a brief agent note) to Claude Opus 4.6. Claude generates a professional property description suitable for both the website and social posts.

CopyPOST https://api.anthropic.com/v1/messages
{ "model": "claude-opus-4.6", "max_tokens": 1024, "messages": [ { "role": "user", "content": "Generate a professional property listing description for a {{bedrooms}}-bed {{bathrooms}}-bath {{property_type}} at {{address}}. Key features: {{features}}. Keep to 300 words. Include a compelling hook and emphasise unique selling points." } ]
}

n8n extracts the description text and stores it for the next three steps.

Step 3: Create Listing Website with Dora AI

Dora AI accepts the property description and a handful of processed photos, then generates a custom listing website with no template boundaries. The generated site is mobile-responsive and includes integrated contact forms.

CopyPOST https://api.dora.ai/v1/websites
{ "project_name": "Property-{{property_id}}", "content": { "headline": "{{property_address}}", "description": "{{property_description}}", "price": "{{asking_price}}", "bedrooms": {{bedrooms}}, "bathrooms": {{bathrooms}}, "sqft": {{square_feet}} }, "images": [ "{{processed_photo_urls}}" ], "style_prompt": "Luxury real estate, modern and minimal, high-end photography feel", "cta_button": "Schedule a Viewing"
}

The API returns a live URL. Store this for your listing database and email confirmations.

Step 4: Generate Video with Hour One

Hour One turns the property description into a short video featuring a virtual presenter. The AI constructs a natural script from the description, then renders it with customisable presenter avatars and background overlays of your listing photos.

CopyPOST https://api.hourone.com/v1/videos
{ "script": "{{property_description}}", "presenter": { "name": "Sarah", "gender": "female", "appearance": "professional" }, "background": { "type": "slideshow", "images": "{{processed_photo_urls}}", "transition": "fade", "timing_per_image": 4 }, "output_format": "mp4", "duration_seconds": 90, "voice": { "language": "en-GB", "speed": 1.0 }
}

The video URL gets stored and can be uploaded directly to YouTube or your property website.

Step 5: Create Social Media Content with Mirra

Mirra generates carousel posts and short-form videos optimised for Instagram, TikTok, and Facebook. Feed it the property description and selected photos.

CopyPOST https://api.mirra.ai/v1/campaigns
{ "content_type": "real_estate", "title": "{{property_address}}", "description": "{{property_description}}", "images": "{{processed_photo_urls[0:5]}}", "formats": [ { "platform": "instagram", "type": "carousel", "size": "1080x1080", "slides": 5 }, { "platform": "tiktok", "type": "short_video", "duration": 15 }, { "platform": "facebook", "type": "carousel", "size": "1200x628" } ], "brand_tone": "professional_approachable", "hashtags": true
}

Mirra returns files ready to upload directly to each platform.

Step 6: Store Metadata and Notify

The final n8n nodes log all URLs (listing website, video, social assets) to your database or CRM, then send you a summary email. You can also use this data to post automatically to your social channels using native integrations for each platform.

Copy{ "property_id": "{{property_id}}", "listing_url": "{{dora_website_url}}", "video_url": "{{hour_one_video_url}}", "social_assets": { "instagram_carousel": "{{mirra_instagram_url}}", "tiktok": "{{mirra_tiktok_url}}", "facebook": "{{mirra_facebook_url}}" }, "processed_photos": "{{pixelcut_folder_url}}", "created_at": "{{timestamp}}", "status": "ready_for_posting"
}

The entire workflow runs asynchronously. From the moment you submit inspection photos until all assets are ready typically takes 8 to 12 minutes, with no manual intervention.

The Manual Alternative

If you want to keep tighter control over individual assets, you can run n8n in semi-automatic mode. Set it to process photos and generate descriptions automatically, but pause before video and social content creation. Review the website and copy manually, then use one-click buttons in n8n to trigger downstream steps. This hybrid approach still saves 70% of your time while preserving creative oversight. Some agents prefer this, especially for premium properties where custom messaging matters.

Pro Tips

1. Rate Limiting and Batching

Pixelcut and Hour One can process multiple files, but don't flood them with requests in parallel. Set n8n to process properties sequentially or limit concurrent calls to 2 or 3. Check each API's documentation for rate limits; most free or starter tiers allow 10-30 requests per minute.

2. Image Quality Verification

Pixelcut sometimes struggles with very dark or extremely overexposed photos. Add a validation step in n8n that checks output file sizes; if a processed image is suspiciously small, it likely failed, and you should flag it for manual review rather than pushing a bad photo into the listing website.

3. Content Customisation Layers

Create a lookup table in n8n with property type (apartment, house, commercial) mapped to different Claude prompts and Dora style instructions. A luxury apartment listing should sound different from a three-bed suburban home. This costs nothing extra but significantly improves output relevance.

4. Archive Everything

Store all generated assets (photos, videos, descriptions, website URLs) in cloud storage with a consistent folder structure. You'll want this archive for legal compliance, historical comparisons, and resale references. Use n8n's storage integrations to organise everything automatically.

5. Cost Efficiency with Model Selection

Claude Opus 4.6 is excellent for property descriptions but slower and more expensive than Claude Sonnet 4.6. Consider using Sonnet 4.6 for the initial description generation, then reserve Opus 4.6 for complex rewrites or multi-property batch analyses. This can cut your LLM costs by 40%.

Cost Breakdown