Exlaining the Decisions That Led to ERPNext

History

Syspro

When STARKART was founded, an independent contractor set up our equipment and systems, including an ERP system called Syspro. As implemented, the system was incapable of forcasting inventory and generally was incapable of operating the way we wanted to do business. When the Syspro server crashed, Syspro itself followed soon after.

Dashboard

There were a few initial attempts at creating an internal system for tracking contracts. The development and maintenance ultimately was appropriated by Dave Thomas, who maintained the system until his departure in Dec 2020.

The first system designed by Dave was called "Carts" and dealt exclusively with STARKART. This system was designed exclusively in PHP. Encouraged by his successes, he designed an implemented the much more ambitious "Dashboard" project, which evolved into a sort of accounting-less ERP system for the company. This system was supplemented and integrated with Fogbugz, which was ultimately replaced by Trello + Front that we called "Truffle".

Dashboard was an opinionated and inflexible system, which made changes to the way we sold reservations and services was a grueling process. It required significant lead time, or even prevented us from experimenting at all due to other priorities.

Why ERPNext?

Dashboard was a high-maintenance project. When Dave left the company, our immediate concern was that one person would be unable to manage servers, services, Dashboard, and the demands of sales. We needed a system that requires less effort to maintain, and can handle changes through configuration changes rather than programming projects.

Options Vetted

We looked at a number of options:

Odoo: Odoo is a popular ERP system with modular components written in Python. A lot of functionality can be tacked onto the core system by installing apps. However, many modules cost money to use, and the modular system lends itself to sprawl rather than deep integration.
Zoho: A paid option, Zoho would be complicated to move over to. Zoho offers a lot of functionality, but when things don't work there isn't much recourse.
Netsuite: We've looked at Netsuite a number of times in the past. The options offered by the platform suggest that it would make a decent fit; however, it is also very expensive.
Salesforce: Like Netsuite, Salesforce would be very expensive to implement, without guarantees that it would meet our demands.
Quickbooks: We also considered handling various parts of our business in Quickbooks, but I strongly believe QuickBooks would not be able handle it well. It also is unreliable, which could criple our business at critical times.
ERPNext: ERPNext is an open source ERP system that is used the world over. It is written in Python, a common language, so it should be relatively easy to find developers willing to help if needed. It is also free with no strings attached (provided you are willing to implement and support it yourself). Additionally, the inteface is consistent and relatively simple to grasp. All of these reasons made ERPNext an easy choice.

Cost

When we started looking at options, the business was dealing with poor sales due to Covid 19. We had an unusually small budget, so most options that involved money were out of the question. ERPNext was the only solution that did not come with any strings.

Maturity

ERPNext is used the world over, and competes with heavy-hitters like Netsuite.

Customizability

ERPNext has a robust interface for customizing and extending existing behavior.

Design Goals

Design a system that offers us flexability in the way we group and sell items and services.
Give users access to the information they need, when they need it.
House our disparate systems under one roof:
- Dashboard
  - Contract entry
  - Store inventory
  - Print order
- Trello
  - Project management
- Front
  - Team email
  - SMS messaging
  - Comments
- Sales Website (starkartthenala.com)
  - Sales reports
  - Sales material
  - Store inventory
- Reporting