Automate real estate listings from inspection photos to virtual tours to social posts

A property listing that goes live without professional photos, a compelling video walkthrough, and social media posts ready to share is money left on the table. Yet many real estate agents still handle this manually, spending three to four hours per property on photo editing, writing descriptions, filming video tours, and crafting social posts. By the time they've finished one property, they've lost the momentum to market it effectively across multiple channels. The workflow I'm going to show you collapses those hours into minutes. You'll feed inspection photos into an automated pipeline that edits them, generates a virtual tour video with AI presenters, writes marketing copy, and publishes ready-made social posts, all with no human intervention between steps. This isn't about replacing your judgment; it's about removing the tedious production work so you can focus on selling. We're combining four tools here: Pixelcut AI to enhance and clean up your property photos, Hour One to turn those photos plus descriptions into a professional virtual tour video, and Postwise to generate and schedule social media content. The orchestration happens in either Zapier or n8n, depending on your preference for simplicity versus flexibility.

The Automated Workflow

Let's map this out in detail. The trigger is simple: you upload a batch of property photos to a specific folder in Google Drive or Dropbox. From there, the workflow processes each photo, creates a video tour, and publishes marketing posts.

Why n8n instead of Zapier?

For this workflow, n8n is better. Zapier charges per task, and with multiple image processing steps plus video generation, costs climb quickly. n8n runs on your own infrastructure or a self-hosted cloud instance, so you pay a flat monthly fee regardless of how many properties you process. The trade-off is that n8n requires more technical setup; Zapier gets you running in minutes. If you're only doing this for a handful of properties monthly, Zapier works fine. If you're a team processing dozens of listings, n8n saves money.

Step 1: Trigger and file monitoring

Set up n8n to monitor a Google Drive folder (or Dropbox, or even an SFTP server) for new image files. When a new property folder appears with photos inside, the workflow springs to life.

CopyGoogle Drive Trigger Node Configuration:
- Folder: /Real Estate Listings/Pending Processing
- Watch for: File created
- File type: Image (jpg, png, webp)

Step 2: Photo enhancement with Pixelcut AI

Each photo passes through Pixelcut's API. Pixelcut handles standard real estate photo tasks: removing clutter from backgrounds, adjusting exposure, enhancing colours, and straightening horizons. You send each image as a base64-encoded string in the request body.

CopyPOST https://api.pixelcut.ai/v1/images/enhance
Content-Type: application/json
Authorization: Bearer YOUR_PIXELCUT_API_KEY { "image": "base64_encoded_image_data", "enhance_type": "real_estate", "remove_background": false, "auto_correct": true, "colour_enhance": true, "resolution": "hd"
}

Pixelcut returns the enhanced image as a file URL. Store this URL in a temporary variable so we can use it in the next step.

Step 3: Compile listing details

You'll need text input for the property details. This can come from several sources: a Google Sheet where you manually enter basic info (address, price, bedrooms, square footage), a template that populates from your CRM, or even Claude Opus 4.6 reading handwritten notes from a voice memo. For this example, assume the data lives in a Google Sheet. The workflow reads the row corresponding to this property batch and extracts: property address, price, number of bedrooms and bathrooms, square footage, key features, and the selling points you want emphasised.

Step 4: Generate the virtual tour video with Hour One

Hour One's API takes a script, a selection of virtual presenters, and your enhanced property photos, then generates a video. First, the workflow constructs a property description using Claude Haiku 4.5 (fast and cheap for straightforward copywriting tasks).

CopyPOST https://api.hourone.com/api/videos
Authorization: Bearer YOUR_HOUR_ONE_API_KEY
Content-Type: application/json { "script": "Welcome to 45 Elm Street. This stunning 4-bedroom Victorian sits on a quiet, tree-lined avenue in the heart of the conservation area. Built in 1952 and recently refurbished, the property features original wooden floors throughout the ground floor, a modern fitted kitchen, and a spacious master bedroom with an ensuite. The south-facing garden is perfect for entertaining, with mature apple and pear trees providing natural screening from neighbouring properties. The property benefits from off-street parking and is within walking distance of excellent schools and local amenities.", "presenter": { "name": "Emma", "style": "professional", "language": "en-GB" }, "images": [ "https://pixelcut-output.s3.amazonaws.com/enhanced_1.jpg", "https://pixelcut-output.s3.amazonaws.com/enhanced_2.jpg", "https://pixelcut-output.s3.amazonaws.com/enhanced_3.jpg" ], "duration_per_image": 3, "video_format": "mp4", "resolution": "1080p"
}

Hour One returns a video ID. Poll the API every 30 seconds until the video finishes rendering, then retrieve the download link.

CopyGET https://api.hourone.com/api/videos/{video_id}/status
Authorization: Bearer YOUR_HOUR_ONE_API_KEY

Once complete, upload the video to YouTube (unlisted initially so you can review it) or directly to your listing website.

Step 5: Generate social media content with Postwise

Postwise takes the property description and your instructions (tone, platform focus, hashtag preference) and generates multiple tweet variations, LinkedIn posts, and Instagram captions. Use the Postwise API or Zapier integration.

CopyPOST https://api.postwise.com/v1/generate-posts
Authorization: Bearer YOUR_POSTWISE_API_KEY
Content-Type: application/json { "content": "Stunning 4-bed Victorian property. Original features, modern kitchen, south-facing garden. £495,000. Conservation area location.", "platforms": ["twitter", "instagram", "linkedin"], "tone": "professional_approachable", "style": "real_estate", "include_hashtags": true, "num_variations": 3
}

Postwise returns several post options. The workflow picks the best-performing variant (you can set rules like "use the one with 280 characters or fewer for Twitter" or "pick the one with the most engagement-driving language"), then schedules posts across your chosen social channels.

Step 6: Store metadata and send confirmation

Log everything: property address, date processed, video link, social post links, and any errors. Store this in a Google Sheet or database so you can track what's been published and follow up on performance. Send yourself (or your team) a Slack notification when the workflow completes, with links to review the video and posts before they go live.

The Manual Alternative

If you want final approval before anything publishes, modify the workflow to pause at steps 4 and 5. Instead of automatically uploading the video and scheduling posts, the workflow generates them and sends you links via email or Slack. You review, make edits if needed (Postwise and Hour One both have web editors), and then manually publish. This adds 15 minutes per property instead of zero, but ensures quality control. For smaller teams or agents handling fewer than five listings monthly, this hybrid approach is sensible. You get automation's speed without the risk of publishing something that misses your brand voice.

Pro Tips

Batch processing and rate limits.

Pixelcut and Hour One both have rate limits.

Don't try to enhance 50 photos and generate 50 videos simultaneously. Use a workflow delay node to space requests two seconds apart. This keeps you well within API quotas and prevents timeouts.

Video rendering is slow.

Hour One typically takes two to five minutes per video depending on length and resolution. Don't expect real-time results. Build a queue system in n8n so multiple properties can be queued up without blocking each other.

Fallback descriptions.

If Claude fails to write a compelling property description (rare, but it happens), have a template description ready to use instead. This prevents workflow failure when you're just missing one phrase. Better to have a generic description than no video at all.

Cost optimisation.

Use Claude Haiku 4.5 for property descriptions instead of Claude Opus 4.6. Haiku is five times cheaper and perfectly capable of writing real estate copy. Reserve Opus for more complex analysis if you need it elsewhere.

Storage and cleanup.

Enhanced images, video files, and metadata will accumulate fast. Set a workflow rule to archive processed files after 30 days and delete temporary base64 image data immediately after processing. This saves storage costs on Google Drive or your cloud provider.

Cost Breakdown