How Many Dental Clinics in the UK Can Get a Chatbot Set Up for £100?

I woke up at 3 am with an idea: what if I could help small businesses in the UK set up AI chatbots that pay for themselves, without requiring a big upfront investment. I knew it was possible because I'd already done it for 3 dental clinics in Karachi, Pakistan, saving them over £5,000 in just 12 weeks. In this post, I'll walk you through exactly how I did it, including the specific tools and setup I used, and what I wish I'd known before I started.

I knew that getting a small business to invest in an AI chatbot required more than just a PowerPoint presentation and a sales pitch. I had to show tangible results, and I had to do it on a tight budget. So, I set out to create a chatbot that would not only save small businesses money but also generate real revenue. The challenge was to do it all for £100 or less.

When I first started, I tried using a well-known but expensive AI tool, LLaMA, via the Meta AI API at $1.75 per thousand tokens. I spent 2 weeks trying to get it set up, but it was a disaster. The setup time was 4 times longer than I'd anticipated, and the cost was prohibitive. I lost 2 weeks and £200 in the process.

Fortunately, I didn't give up. I discovered a game-changing AI assistant called Claude, which allowed me to generate text at a fraction of the cost. Here's how I set up my AI chatbot in 5 easy steps:

1. Choose the right platform: I selected Baileys, a Node.js library specifically designed for WhatsApp bots. This allowed me to integrate my chatbot with the popular messaging platform, which many UK small businesses already use. 2. Select the right AI tool: I chose Claude, version 4.5, via the Groq API at $0.27 per million tokens. This provided fast and affordable AI inference, critical for handling a high volume of user queries. 3. Design the chatbot experience: I worked with Claude to create a conversational flow that would engage users and provide valuable information. I included features like appointment booking, payment processing, and FAQs. 4. Integrate with WhatsApp: I used Baileys to integrate my chatbot with WhatsApp, allowing users to interact with the chatbot directly from their messaging app. 5. Deploy and test: I deployed my chatbot on a Hostinger VPS, which provided a reliable and scalable hosting environment. I tested the chatbot with a small group of users to identify any issues and make adjustments as needed.