Skewness is just the third moment about the mean divided by the standard deviation raised to the power of 1.5.  In order to completely characterize a distribution, you  need all of the moments. Skewness is somewhat useful in characterizing the asymmetry of a distribution. Symmetric distributions have a skewness of zero, a distribution with a tail on the left will have a negative skewness, and a distribution with a tail on the right will have a positive skewness. For example, a normal distribution has a skewness of zero, but a skewness of zero does not mean that the distribution is normal. Depending on what you're doing the mean, standard deviation, and skewness might be sufficient. I've used skewness in some of my research. One of the ways I've use it was to show that particular data sets were non-normal. I've never seem a  normal data set in my field. Instead, I've usually used non-parametric statistics to characterize data sets.

You may find useful this article:

@Article{VisualizationSkewedDataR,

Title = {Visualization of Skewed Data: A Tool in {R}},

Author = {R. Ospina and A. M. Larangeiras and A. C. Frery},

Journal = {Revista Colombiana de Estad\'istica},

Pages = {399--417},

Volume = {37},

Year = {2014},Arxiv = {1403.0532},

Month = {Dec.},

Number = {2Spe},Abstract = {In this work we present a visualization tool specifically tailored to deal with skewed data. The technique is based upon the use of two types of notched boxplots (the usual one, and one which is tuned for the skewness of the data), the violin plot, the histogram and a nonparametric estimate of the density. The data is assumed to lie on the same line, so the plots are compatible. We show that a good deal of information can be extracted from the inspection of this tool; in particular, we apply the technique to analyze data from synthetic aperture radar images. We provide the implementation in R.},

Doi = {10.15446/rce.v37n2spe.47945},

Keywords = {Exploratory Data Analysis, Skewed Data, Visualization}

}

@ Michael Arndt: You are right sir.

 thanks all for your efforts answering my question.

Skewness is primarily the second order effect . When studying waves beyond the second order use of skewness alone is not sufficient, kurtosis needs also to be used. Further, the estimator of skewness is affected by sampling variability, the uncertainty due to limited number of data so it is important that a data set considered is sufficiently long not to get a bias results. 

Though some answers were given, but let me add that the skewness value if the third moment is used, there is possibility of identifying only the tail end of the distribution. As for the shape of the distribution, the kurtosis tends to be more efficient. And as Michael said, you hardly get a data that is normally distributed, but the run a normality test to have a picture of your data.