MVP does not mean a bad product
A first version is minimal relative to learning, not essential quality. It may have one flow, but that flow must be understandable and reliable enough for its intended test.
Cutting features reduces cost. Ignoring security, data or the ability to fix failures only moves cost into the future.
The variables that carry real weight
Authentication, roles, payments, notifications, data imports and external integrations add work because they introduce states, errors and edge cases.
The quality of initial material, decision speed and access to domain owners matter too. A prepared project cuts time lost to assumptions and revision.
Ask for comparable estimates
Send the same brief to a few professionals and ask for scope, exclusions, milestones, assumptions and maintenance cost. A total without that context is hard to assess.
Ask what they would build first and which risk they would test. A strong proposal makes decisions visible, not only rates.
Reduce budget without losing the product
Limit platforms, roles, automations and integrations. Some work can stay manual behind the scenes if that accelerates learning.
The right freelancer separates the core that must be software from operations that can wait. A budget becomes a priority tool rather than an obstacle.