Seems like a cool little widget, but I doubt the market for it would be big enough to be worth your time. It's a small enough of a problem that most people won't notice it, and if they do they could easily solve it themselves. People won't be looking for a solution, and it's not cool enough that people would share it just because.
As for your questions, your responses are way too long.
I think this seems like an "easy" thing to build in-house, but we believe getting this right will take a ton of time, which companies could put towards doing what they do best.
Here is what we have: 1. The rules engine, to track visitors and fine control who see's feature experiments and how often. 2. An analytics suite, to capture all data and show it in a form which is digestible and allows a PM to take action 3. Making overlays work cross browser and device. 4. Solution which requires absolutely no-code to get feedback on existing features. A gui where PM's can visually find features on their website, tap on them and configure them to ask for feedback.