I've done some large refactor work previously and have attempted the rewrite. The rewrite didnt go well - the only thing harder than the engineering tasks were the stakeholder updates. The hardest but best course of action I took was to go back to the roots of the plan and based on the errors made recalculate a new course. The improved course of action eventually proved to be incredibly effective.

The problem with the parallel system is its rotten almost by existing. Code that isnt in the hands of users is a waste. The real solution is to find a way to integrate the 'new' concept into the old one, and progressively refactor old into new. This takes the stakeholder conversation from an incredibly stressful 'whens it going to be ready' to 'how far along are we'.

Ive just got dropped onto one of these. Actually, I was first informed of it 6 months ago and my initial reaction was 'oof'. Anyone with rewrite experience would immediately recognise it as disaster territory.

I have a plan to try to get this project on track based on my experience but im going to need to work hard with the senior management to get it back on the rails.