🏆 Community Challenge Winner: Website Development System

Community Challenges
,
1

Hey Makers :waving_hand:

The wait is finally over! We are presenting to you the winner of our Make and MCP Community Challenge! Meet @NeillHuman, who impressed us with his solution that develops custom websites. Without further ado, let’s get into it! :trophy:

Neil Human
Neil Human1920Ă—1080 257 KB

:light_bulb: The Story: Fixing the Static Web & Lead Loss

Neill identified a pain point that keeps small business owners up at night: The Speed Gap.Traditional web development takes weeks and costs thousands. Even when a site finally goes live, it’s often a ghost town. Leads come in through contact forms and sit in an inbox for days, resulting in a 50% drop in conversion for every hour of silence.

Neill’s mission was clear: He wanted a way for AI to not just talk about building a website, but to actually deploy a complete, functional web presence through natural conversation, and ensure that once live, the site handles leads instantly, 24/7.

Screenshot_6_UTF-8''make_mcp_tool_deploy (1)
Screenshot_6_UTF-8''make_mcp_tool_deploy (1)1280Ă—686 143 KB

:white_check_mark: The Solution: The AI-Driven “Business in a Box”

This solution utilizes a cutting-edge architecture that bridges the gap between AI reasoning and Make’s execution power:

  • MCP Integration (The Hands): Unlike simple webhooks, this solution exposes Make scenarios as Model Context Protocol (MCP) tools. This allows Claude or ChatGPT to “grab” Make modules and build the site in real-time.
  • The 2-Minute Deployment: By telling the AI, “Build a website for a Dublin bakery,” the system fetches brand configs, generates HTML, and deploys it to WordPress instantly.
  • Multi-Tenant AI-CRM: The real magic happens post-launch. Every site includes an AI-powered contact form. When a lead hits the form, Make looks up the specific site_id, fetches that client’s unique config, and sends a personalized AI response immediately.

The biggest technical hurdle? JSON escaping. Handling complex HTML through MCP pipelines meant nested replace() functions were needed to keep the data flow clean—a masterclass in technical troubleshooting!

:man_judge: What our Judges Loved

The judges were our well-respected and knowledgeable Community Champions. Here’s what they loved about the solution:

:bust_in_silhouette: “Very exited to see that this works as a full system, based on a single prompt in Claude Code. Maybe not directly enterprise level, but very impressed!”

:bust_in_silhouette: “Best demonstration of Make MCP compared to all the other submissions in my opinion. The end result was a great starting point without touching WordPress.”

Screenshot_3_UTF-8''make_mcp_comp_AI_CRM (1)
Screenshot_3_UTF-8''make_mcp_comp_AI_CRM (1)1280Ă—686 136 KB

:rocket: The Impact & What’s Next

The numbers speak for themselves:

  • Deployment time: Slashed from 4 weeks to under 2 minutes.
  • Cost: Reduced from ZAR 15,000+ to just ZAR 2,500.
  • Lead Response: Instant, 24/7 automation.

Neill isn’t stopping here. The next phase involves adding WhatsApp integration (crucial for the South African market), automated appointment booking, and expanding the AI’s “toolbox” to include WooCommerce and payment gateways via WP_CLI.

This is the future of the web, where an idea becomes a live, lead-generating business before you’ve even finished your coffee.

:movie_camera: Watch the solution in action here

13 Likes