Skip to content
Code by Sergio
PTENES

2026

RouterPlanner

From a manual, spreadsheet-based processto an operation connected to the CRM.

Internal projectProprietary code

Back to category
Router Planner

The previous process

Before, the spreadsheet was the process.

Every week, each client in Ploomes was copied by hand into its route’s tab.

  • ~40

    Clients in a typical week

  • 8+

    Routes usually organized

  • 1+ day

    In heavier weeks, from Friday into Saturday

Approximate figures.

Before

  1. 01Ploomes
  2. 02Spreadsheet
  3. 03Copy clientsManual
  4. 04Sort into tabsManual
  5. 05Check
  6. 06Print
  7. 07Operation

After

  1. 01Ploomes API
  2. 02Router Planner
  3. 03Select routeHuman decision
  4. 04Validate data
  5. 05Generate file
  6. 06Operation

What’s left for a person to decide:

“Which route should this client follow?”

Selecting routes inside Router Planner

The spreadsheet stayed.The manual process didn’t.

Now Router Planner is the system. Excel is just the output.

  • One general tab, with every client.
  • One tab per route — around eight or more a week.

The generated file

  1. General
  2. Route 01
  3. Route 02
  4. Route 03
  5. Route 04
  6. Route 05
  7. Route 06
  8. Route 07
  9. Route 08

Fast wasn’t enough.It had to be right.

Before it goes out, every route is checked against Ploomes.

  1. Record changes

    The address that counts is the one in Ploomes now, not the spreadsheet’s.

  2. Cancellations

    Clients in the cancellation flow are checked before the operation.

  3. Up-to-date data

    The query goes straight to the CRM, through the API, at planning time.

Checks before the operation

System flow

  1. 01Ploomes CRMWhere the clients live
  2. 02APIQueried at planning time
  3. 03Router PlannerFastAPI, validations & rules
  4. 04Excel exportGeneral tab plus one per route
  5. 05LogisticsThe operation

Result

~1–2h

To organize a typical operation today, depending on the number of clients. Before, it could run from Friday into Saturday.

Operational estimate, not a benchmark.

  • Less manual work

  • Data connected to the source

  • Checks before the operation

  • Automatic export

My role

End‑to‑end conception and development.

  1. Understanding the flow
  2. Solution design
  3. Development
  4. Ploomes integration
  5. Validation rules
  6. Deployment

Tech specs

Backend
PythonFastAPIPydanticUvicorn
Integration
Ploomes APIRequests
Data & export
OpenPyXL
Frontend
HTMLCSSJavaScript
Infrastructure
VPS Linux

Internal projectProprietary codeEnvironment with managed access control