Can I suggest that the whole "Lean startup" process is an example of dangerous hyper-reason? You're A/Bing your test subjects, you're responding purely to dat

Can I suggest that the whole "Lean startup" process is an example of dangerous hyper-reason? You're A/Bing your test subjects, you're responding purely to data and a dogmatic following of the technique results in you never asking how your users/test subjects _feel_ but only how they behave. Thus we end up with software like Facebook that arguably makes a lot of people sad while being unable to stop using it.