I phrased it a little oddly. What I mean is that if not even my employees (who hopefully share my ideals and values) volunteer to use my software, I would take

I phrased it a little oddly. What I mean is that if not even my employees (who hopefully share my ideals and values) volunteer to use my software, I would take that as a strong signal that I've failed to find a sustainable niche in the market, and maybe the right approach is not to force it down people's throats and hope that they can dig me out of my hole, but rather to pivot and ask myself, "what should I have done differently to get a product people want to use, despite technical flaws?" So yes, dogfooding is there to get immediate feedback on technical flaws of an otherwise desirable product. Dogfooding is not there to fix a product--market mismatch. That needs to happen at a different level of development.