It is not a war. It is a business decision.
The tech industry loves holy wars. But the reality is that both custom software and SaaS have their place. The key is knowing when to use each one, without dogma.
When to choose SaaS
SaaS is the right choice when the process you need to solve is standard (accounting, email, basic CRM), you have no competitive advantage in how you execute that process, you need to be operational in days not months, the initial budget is limited, and you do not have an internal technical team to maintain custom software. Examples where SaaS wins hands down: email, basic accounting, standard project management, video conferencing.
When to choose custom software
Custom software makes sense when your process IS your competitive advantage, no SaaS covers more than 60% of your needs, you are adapting your operation to the software instead of the other way around, the accumulated cost of SaaS plus integrations exceeds custom development, and you need total control over data and product evolution.
If your team spends more time adapting to the software than the software adapting to your team, you need something custom.
The gray zone: the 60/40 rule
Most decisions fall in a gray zone. For these cases, we use the 60/40 rule: if a SaaS covers 60% of your needs, it probably makes sense to use it and adapt the remaining 40%. If it covers less than 60%, the cost of adapting your operation will exceed the cost of building.
The real cost of custom software
Many companies underestimate total cost by looking only at initial development: initial development accounts for 30% of total 3-year cost, maintenance and evolution for 50%, and infrastructure and support for 20%. If you only budget for initial development, you will fall short. Always. Custom software is an ongoing investment, not a one-time expense.
Final recommendation
Before deciding, do this exercise: list the 10 most critical processes in your operation. For each one, answer: "Is there a SaaS that covers 60% or more of what I need?" If the majority says yes, combine SaaS solutions. If the majority says no, consider custom software for the core and SaaS for the rest.
The best technology decision is not the most sophisticated one. It is the one that solves your problem with the least overhead possible.