SHOULD I GO FOR THIS BUSINESS IDEA.
retained text
**I researched an AI robot keychain business for a competition in Pakistan. Should I try building it myself? Need advice on costs, marketing and funding.** **A couple of weeks ago, I was researching business ideas for an entrepreneurship summit when I came up with a product I genuinely fell in love with: a small, AI powered robot that you could carry as a keychain.** **During my research, I couldn't find anyone in Pakistan making quite the product we had in mind, so my team and I started working on its design, features, hardware requirements, production costs and selling price.** **The concept:** **- Responds when you call it by its name.** **- Answers questions using AI.** **- Sets timers and handles simple tasks.** **- Displays different animated facial expressions.** **- Has a good microphone and speaker so you can talk to it.** **- Provides entertainment for students without requiring them to open social media and get distracted.** **We wanted to make it a fun, interactive gadget rather than just another device that encourages screen time.** **Our initial financial estimates:** **- Production cost: PKR 10,200(37$) per unit** **- Target retail price: PKR 14,999(54$)** **- Gross profit: PKR 4,799(17$) per unit (\~32% gross margin)** **These were preliminary estimates, not costs verified through actual manufacturing. We also hadn't prioritised affordability because our main goal was to build an impressive product and win the competition.** **Unfortunately, our team eventually split up, and we decided not to participate. However, I loved the idea so much that I genuinely wanted to turn it into a business and sell it nationwide.** **Recently, I came across @rekko\_co in Pakistan working on a similar concept. I also noticed someone in India starting something similar around October. That made me reconsider whether I should try building this myself.** **The technical side** **My proposed hardware setup includes:** **- ESP32-S3 N16R8 (16MB flash, 8MB PSRAM)** **- 1.28-inch round IPS LCD (240 × 240)** **- INMP441 microphone and MAX98357A audio amplifier** **- Miniature speaker** **- 500mAh rechargeable LiPo battery and USB-C charging module** **- Vibration motor** **- Custom 3D-printed enclosure** **For software, I was considering the XiaoZhi open-source ESP32 firmware, LVGL for animated expressions, ESP-SR for offline wake-word detection, and DeepSeek's API for AI responses.** **This is still a proposed setup, and I haven't validated the final design or its actual production cost.** **Where I'm stuck** **I'm a student, so I need to balance this with my studies. I could potentially build a working prototype to showcase the product, but I don't have enough investment to manufacture inventory.** **My biggest concerns are:** **1. Production costs: My target customers would mainly be school and university students, who might not be able to afford a PKR 15,000(54$) gadget. Which components could I replace or simplify to bring the cost down significantly without losing the core experience?** **2. Manufacturing and funding: Could I start by building one prototype, testing demand and taking pre-orders before manufacturing a small batch?** **3. Marketing: How would you market a product like this in Pakistan? Would Instagram, TikTok, student communities, influencers or school/university partnerships work best?** **4. Product viability: Is there enough potential for this to become a real business, or would the hardware costs, cloud AI expenses and competition make it difficult? What should I validate before investing anything?** **5. Balancing studies and business: Is it realistic to develop and launch this gradually while studying, or would it be better to wait?** **I don't want to build a prototype, get excited, and then discover that nobody is willing to buy it. I'd rather understand the market and economics before committing money.** **If you've built a hardware startup, manufactured electronics, launched a physical product, or started a business with limited capital, I'd really appreciate your advice.** **What would you do first if you were in my position? And if you have experience with ESP32-based AI gadgets, I'd especially love to hear how you'd approach reducing the hardware cost.** **Thanks!**
document fetched 2026-10-25 00:29 · canonical https://reddit.com/r/smallbusiness/comments/1xefvjd/should_i_go_for_this_business_idea