I am a regular high-volume user of treeshare, replicating information from Ancestry where I enter it to RootsMagic. I entirely agree with @Sami that I don’t need the treeshare process to force me to check everything I have entered, and I certainly find the amount of manual work it currently involves utterly unacceptable. However, I used to work in process automation and I know that what @rzamor1 writes is also true, and that automation done badly can result in screwing things up. It is often not possible to automate something 100%, especially when there are differences in database design between the two databases involved.
That’s why I spent a long time thinking hard about what could be done, setting myself the target of eliminating at least 90% of the current keystrokes/mouse clicks, and working out how this target could be achieved without introducing a risk of corrupting either database and implemented in stages each of which would be relatively easy to code and test.
I posted it in a thread called Proposal for Treeshare improvement. Unfortunately, apart from the section about Treeshare corrupting your RM database if you import two children with the same name in a single operation (which still hasn’t been fixed), none of the other suggestions gained much traction here.
Even so, I remain convinced that an approach like this is the only way to make serious improvement and I recommend my suggestions again to the RM community and the development team.