Well as it has been proved to be profitable and you already have clients, it's up to you do decide it(well... it has always been up to you) as you are probably

Well as it has been proved to be profitable and you already have clients, it's up to you do decide it(well... it has always been up to you) as you are probably the only who know how big your public will be. If you are afraid of stuff breaking up/it not scaling so well and so on, work on it until you feel satisfied. As you already said, it's B2B and service standards may vary from country to country. A good experience would be try to get some clients from another country with "high service standards" and try to give them some benefits to try it out the services(for free maybe?) and if the feedback is good enough you launch it. If you would like to wait for the API to be ready, just do it. I know that there's a lot of people that would say that "launch it with 50% done" or something, but the person that know your product best is YOU. I hope you have much success with it!