If you choose wrong you are going to have a team following someone that nobody else thinks it's right.
The solution is to have an arbiter,some one that makes sure that compromises are reached and the design moves forward.
Document all decisions and stuck to them unless a major problem doesn't allows it. Some people will try to reopen closed discussions if the result was not the one that they wanted.
To find a great solution requires a lot of time, and trial and error. Don't get to change all the system at the same time. Do small improvements in the agreed direction. Get feedback often about the quality of the changes. If there is any major flaws you will need to go back to design.
In the future is better that redesign is part of the normal process. Don't let the project get old. Technical debt accumulates easily, remove it as soon as possible.
But it's going to be difficult depending on the state and size of the project. Follow up all decisions,and support the team if they break something while doing the improvements.
Depending on the size you can read The Phoenix project. It's a book on how to work with improvements to big projects.