A redesign that loses your existing Google ranking is not a redesign, it is a reset. Most of them do exactly that.
A site that has been running for years has something genuinely valuable attached to it: history. Google knows the URLs, other sites link to some of them, and a few pages quietly bring in traffic that nobody has ever looked at. The standard redesign throws all of that away — new URLs, no redirects, and three months later the enquiries have not come back and nobody can work out why.
The rebuild is the easy half. The half that decides whether it works is the migration: every existing URL mapped to its replacement, 301 redirects in place before launch, the pages that already earn traffic identified and kept rather than quietly dropped, and the whole thing checked in Search Console afterwards. Done properly, a redesign holds its rankings through the change and improves from there because the new site is faster.
A crawl of the current site plus whatever Analytics and Search Console data exists, to find which pages are actually earning something. Usually at least one surprise.
Every existing URL gets a destination on the new site before a single page is built. This is the step that gets skipped, and it is the one that costs rankings.
New site on a modern stack, content carried across and tightened, speed handled during the build rather than bolted on afterwards.
Redirects go live with the site, Search Console gets checked for crawl errors in the following days, and anything that moved wrong gets fixed while it still matters.
It will if the redirects are not done, which is why most redesigns cost traffic. Every old URL gets a 301 to its replacement before launch, and Search Console gets checked afterwards. Done that way, rankings carry across and usually improve, because the new site is faster.
Most small business redesigns land between $900 and $2,500 depending on how many pages carry across and how much of the content needs rewriting. The migration work is included rather than being an add-on you find out about later.
If it still works, yes — sometimes the problem is speed and mobile behaviour rather than looks, and rebuilding the same design on a better foundation is the cheaper and less disruptive answer. I will tell you which situation you are in honestly.
Then we set both up as step one and work from a crawl of the existing site instead. It is less information than I would like, but it is still enough to map URLs properly and avoid the damage.
It is not. The new site is built and tested on a separate address, and the switch happens once everything is ready — redirects included. The changeover is a few minutes, not a few days.