✈ Discuss your project
All case studies / Custom Products & SaaS

Excel profitability model became a server calculator: one link for managers, PIN-protected admin

Industry: Coffee and tea supplier for businessesScale: sales department, managers using phones and PCsService: End-to-end products and SaaS
4 of 4test cases matched Excel within 0.1%
1 linkphone or PC, nothing to install
3 signalstraffic lights for profitability, profit, and payback
I need something similar

Before

  • The calculation had to match the client's Excel exactly, which was the main reference
  • A manager opens one link on a phone or PC and sees traffic lights, profit, and payback
  • An admin uses a PIN to change cost, leasing, and thresholds, while managers cannot see the cost

What we built

  • Excel formulas were moved to a Python server engine, each calculation reads the current settings
  • Matched against Excel with an automated test on four test vectors
  • PIN-protected admin: cost, leasing, traffic-light thresholds; admin token uses HMAC
  • The API does not expose cost, equipment prices, or leasing without a token
  • Mobile-friendly interface in the client's brand colors with their logo, calculation history stored in SQLite

How it works

Managervolumes and sale price
→
Calculation APIFastAPI
→
Engineformulas from Excel
→
Settingscost, leasing, thresholds
→
Resulttraffic lights, profit, payback

Result

  • Work accepted July 1, 2026
  • The client approved the look: "visually, I like everything"
  • Admin changes appear in the next calculation on any device, verified in the browser
  • The client returned with a new order

Technology

PythonFastAPIuvicornJavaScriptSQLiteCaddysystemd
What matters in a project like this. While settings lived in each device's browser, admin changes did not reach managers. Team-wide values are now stored on the server and read for every calculation.

Have a similar project?

Describe it in two lines in Telegram. We'll explain how we'd approach it, how long it would take, and what it would cost.

Message me on Telegram

Similar case studies