VendorIQ

Business Intelligence in your pocket

Comment

Inspiration

Small and medium business owners across Africa face a daily operational blind spot because traditional software requires tedious manual data entry that busy traders do not have time to maintain. The inspiration behind VendorIQ came from realizing that millions of merchants already manage their lives on WhatsApp, making it the ideal place to bring software directly to them. By acting as an AI operating system, VendorIQ turns everyday natural text messages into structured sales records, inventory updates, and proactive business intelligence without forcing users to learn complex applications.

What it does

VendorIQ functions as an autonomous conversational business assistant and intelligence engine built natively for WhatsApp. It allows users to log sales, record expenses, update inventory, and track customer debts simply by texting in natural language like English, Pidgin, or local phrasing. Beyond simple transaction logging, the platform automatically generates daily summaries, weekly performance reports, and deep financial insights to give merchants total business clarity.

How we built it

VendorIQ relies on a robust backend architecture paired with an intelligent AI processing layer. The communication layer integrates with WhatsApp through the Green API bridge to capture incoming messages via secure webhooks, while a high-performance FastAPI service handles asynchronous request routing. Incoming texts are processed by Google Gemini models configured with specialized system prompts to parse user intent and extract structured entities. Finally, persistent data storage is managed securely via Supabase, and the entire system is containerized and hosted on a dedicated cloud VPS using Docker.

Challenges we ran into

Building an AI business intelligence assistant for informal commerce presented unique engineering hurdles, particularly handling unstructured human conversation that mixes Nigerian English, Pidgin, slang, and varying currency formats. We successfully overcame this by implementing strict JSON response schemas and fallback layers within the Gemini parsing logic to guarantee clean entity extraction. Additionally, maintaining continuous webhook uptime required migrating away from ephemeral serverless environments to a stable, containerized VPS infrastructure.

Accomplishments that we're proud of

We successfully built a zero-friction conversational bookkeeper that requires absolutely no learning curve for local traders. Achieving production-grade reliability by migrating our backend to a containerized Docker setup paired with Green API ensured uninterrupted WhatsApp webhooks. We also crafted a high-end, venture-backed web platform featuring dark mode integration and authentic WhatsApp chat simulations that position VendorIQ as a premier business intelligence tool.

What we learned

Throughout this journey, we learned that user experience is paramount when building software for non-technical users, because if an application requires any learning curve, traders will abandon it. By keeping the interface entirely inside WhatsApp, we realized that the best software completely hides itself. We also gained deep technical expertise in asynchronous webhook management, prompt engineering for structured intent classification, and building scalable backend architectures.

What's next for VendorIQ

Pushing the MVP and unboarding of customers

Built With

Share this project:

Updates

Submission history