There are many, many ways to do this with branching and deployment strageties but the key is to identify exactly what you want to achieve and then look at the different routes.

How many different clients and/or environments are you looking at, should you be trying to run them all from the same master or would it be better to run them all from their own branches,what servers are you using, what's their distribution, etc.?

In our case we have 5 different environments each with its own Db. We have 4 branches from master (master being production environment). 

Each environment has its own build and release pipeline, pipelines contain variables which allow us to change config settings and turn features on and off which is down to how we write the code to enable that.

Can use something like Redgate toolkit to clone the Db between environments, very handy for masking and trim downs when you need a copy of production Db to for your dev/test environment.