In short, is not so easy. A cps measure is a count per second (the abbreviation means that), those are the number of ions that hit the detector per unit of time. Now there is a correlation between that number and the quantity of ion but for a number of reasons, error in counting and amplification of signals, various source of noise, the efficiency and variability of the ionizing sources in respect to the ion production make the direct correlation between quantity present and signal unreliable. This is so true that in a number of MS software on the y-axis you will found "arbitrary units" as measure. The measurements about quantities on a single spectra is relative to the rest of the peaks, at best.

Said that to quantify something you should construct a calibration curve for the substance you want to quantify, and construct experiments in a way to measure that substance. Or use an internal standard with similar characteristics added to your sample.

Dear Sangita,

As Manlio mentioned the best way to infer on the significance of signal intensity is to conduct a calibration curve as signal intensity (minimum of cps) will not be the same for all mass spectrometers. Also if source or detector are not properly maintained and cleaned regularly then the signal intensity will decrease. By performing a cal curve these and other issues will be accounted for.

Hope this helps,

Ana

In addition to the previous comments it should be said also that in a mixture you may have components easily ionizable than others and in many cases minor components may give the more intense signals. 

To add to the previous comments, the minimum intensity you can consider relevant for your application is related to the signal-to-noise ratio, rather than the absolute number of counts hitting the detector. The signal-to-noise ratio can be used to determine your limit of detection (also the lowest point on your calibration curve) and your limit of quantitation. 

Amy L Heffernan What would be the minimum S/N to be considered. And since the S/N is calculated through the software by measuring the Signal to noise ratio, where the noise region can be chosen from any part of the chromatogram. Are there any chances of manipulating the S/N because there are options of Smoothing too.