How do I recover a technology project before spending more money on it?

Pause nonessential work, verify the current state, and define the minimum outcome still worth completing. Estimate the cost and risk of finishing, changing direction, or stopping. Spend again only after there is a credible plan tied to business value.