Bootstrapping is used to get results about the variance, not the mean. You can have a high mean and still the data has high variance. You can see that also in your second image. the variances generated from bootstrapping are rather high. either the variance in your data is high or your sample size is very small. If the sample size is large this could be a hint for heterogeneous data (i. e. there are multiple segments in your data that create the high variance).
as a starter to understand bootstrapping see:
https://en.wikipedia.org/wiki/Bootstrap ... tistics%29
and when using smartPLS you definitely want to read through:
Hair, J.F., Hult, G.T.M., Ringle, C.M., Sarstedt, M., 2014. A Primer on Partial Least Squares Structural Equation Modeling. Sage, Los Angeles.
after that you should have an idea of where the problems come from. The above mentioned reasons are only some of the reasons that could exist. the search function for this forum is usually also of great help.