Experiment: E-newsletter content chosen by the reader outperformed e-newsletter content selected by an editor