Skip to content
← All work

Case study 03 — SellWise

Forecasts and customer intelligence for small online shops.

A business-intelligence dashboard with three ML modules — sales forecasting, customer segmentation and churn prediction — served from an independent Python microservice.

Status
Live
Year
2026
Context
Independent product
Links
Live app ↗Source on GitHub ↗

Anatomy

  • Deterministic code
  • Model output
  • Guardrail
  • Data
  1. 01 · Data
    Orders, customers, stockPostgreSQL · CSV import
  2. 02 · Code
    Node.js REST APIOrders, inventory, invoices, reports
  3. 03 · Model
    Demand forecastProphet · 7 / 14 / 30 days
  4. 04 · Model
    Segments & churnRFM · logistic regression
  5. 05 · In / out
    DashboardForecasts, customer risk, stock alerts

The problem

Small online sellers keep their sales history in a spreadsheet or a basic shop backend, but have no way to answer the questions that decide cash flow: what will sell next month, which customers are slipping away, and what to reorder now.

What was built

  • 01An operations app for the day-to-day: products, orders, quick sell, invoices, expenses, CSV import and reports.
  • 02Demand forecasting with Facebook Prophet over 7, 14 and 30-day windows, alongside reorder suggestions and inventory alerts.
  • 03RFM-based customer segmentation and churn prediction with logistic regression.
  • 04The ML layer as an independent FastAPI microservice, called from the Node.js API and backed by the same PostgreSQL database.

Key decisions

  1. ML as its own service.

    Python where the ML ecosystem is, Node.js where the product logic is. Models can be retrained or replaced without touching the app's API.

  2. Models a shop owner can trust.

    Prophet, RFM and logistic regression are explainable and work on the small datasets small shops actually have. No deep model where a simple one is enough.

Stack

  • React
  • Node.js
  • Express
  • PostgreSQL
  • FastAPI
  • Prophet
  • Logistic regression

Need something like this?

Forecasts & customer intelligence →

Demand forecasts, customer segments and churn risk from the sales data you already have.

Next case study

BUBT AI Assistant →

A university chatbot that answers from official documents — and cites them.

Have a workflow that should run itself?

Tell me what it is, what data it touches and what “done” looks like. You'll get a straight answer on whether AI is the right tool — and what it would take.

Start a project

Free 30-minute intro call. Replies within 24h.