The default answer to 'should we build custom software?' should usually be no — until a specific set of conditions makes off-the-shelf software actively work against you.
Off-the-shelf makes sense when your workflow is close enough to the standard case that configuration, not code, closes the gap. Custom software earns its cost when your process is genuinely different, when integration between multiple existing systems is the real problem, or when the workflow itself is your competitive advantage.
A useful test: if your team has built more than two spreadsheet workarounds to compensate for what the off-the-shelf tool can't do, that's usually the signal it's time to scope a custom build.
The real risk in custom software isn't cost — it's scope creep and vague requirements. We spend the first phase of every custom engagement mapping the actual operating process before writing a line of code, specifically to avoid that risk.