The choice between custom ERP and SaaS is often oversimplified. On one side, SaaS is treated as always faster and cheaper. On the other, custom ERP is treated as automatically better for control. Neither assumption is reliable on its own.
The right decision depends on how standard or specific your operations are, how much edge-case logic matters, how many integrations are required, and how much adoption friction the business can realistically absorb.
When SaaS can be enough
- The core processes are relatively standard and not a unique source of competitive value.
- The team can adapt its way of working without introducing too much operational friction.
- Required integrations are limited or already well supported by the platform.
- The goal is to move quickly with a simple and controlled scope.
When custom ERP makes more sense
- The process is specific and central to the business.
- There are many exceptions, roles, and workflows that a SaaS tool would push outside the system.
- Data needs to become a decision surface, not just a repository.
- The company is already carrying workarounds, side spreadsheets, and fragile integrations.
The hidden cost of workarounds
Many companies view SaaS as cheaper because they compare the subscription fee with the initial cost of custom delivery. That comparison is incomplete. You also have to include the hidden cost of adaptation: duplicate work, side spreadsheets, slow decisions, fragmented data, and weak adoption.
When a system forces people to work around the platform instead of inside the platform, the total cost moves into daily operations and gets paid every week by the team.
The right question to ask
The real question is not which option feels more advanced. The right question is this: is it more sustainable to adapt the process to the software, or to build the software around the process that creates value?
If the process is secondary, a SaaS platform can be excellent. If the process is critical, differentiating, or full of non-negotiable operational logic, forcing it into a standard platform can cost more than the custom project itself.
“An ERP should not be chosen by category. It should be chosen based on the operational cost of the rigidity it introduces.”
Davide Gentile
How to decide without getting stuck
The best way is to assess four elements clearly: processes, roles, integrations, and adoption. If that analysis shows the team would have to change too much just to fit the software, that is a signal worth taking seriously.
On the other hand, if the scope is simple and the SaaS platform truly supports operations without generating workarounds, starting with a standard platform may be the most pragmatic path.
The final criterion
Choosing well does not mean always choosing custom. It means avoiding a solution that feels comfortable at the start but slows the business down every day after launch.
If you want to make this decision seriously, do not start from modules. Start from the critical flows and the points where you currently lose time, control, or trust in the data.