It's an interesting experiment, thank you for running it.
I'd be willing to stake good money on the fact that your respondents' awareness that this was a generic reading has influenced the apparent low scores. I remember we have discussed in the past the possibility of running a blind trial, whereby some respondents get a bespoke reading while a control group is presented with a generic reading, but the respondents don't know which group they fall into. I wonder whether scores would be higher across the board in such a scenario.
Of course, it would be the devil of a job to organise, and I doubt we would get anywhere near enough volunteers to lend any kind of statistical credence to the results, but it's an intriguing concept to idly contemplate. For me, at least...