The first aim of this contribution is to study the role of cheap talk in the determination of a focal point in a bargaining game. A general discussion was therefore conducted before the bargaining, followed by a poll in which players must find a common solution to an abstract bargaining problem. On this issue, our conclusions are negative: no signifcant effect seems to emerge. This is not entirely unexpected; since the common discussion and the voting procedure staged the confrontation of different viewpoints, there has been no unanimous result. On the other hand, we can conclude that knowing individual preferences may help predicting towards which solutions players will converge, at least in a social setting, and understanding their satisfaction. These considerations refer to variables traditionally left out of economic analysis, which focuses instead on the final payoff and not on its relation to preferences or satisfaction, which are deemed non-measurable. On the contrary, this work has shown that consistent data emerge by simply asking players to express preferences and satisfaction. This makes us confident in formulating new hypotheses aiming at uncovering and understanding individual behaviour. Even if our results do not allow drawing strong final conclusions, they point out new possible answers, new ideas and topics for discussion.