In automotive manufacturing, MQL (Minimum Quantity Lubricant) machining has been implemented on multi-axis CNC Machining Centers. In MQL machining, since there is no flood coolant, small quantities of coolant through spindle-tool is used at targeted cutting zones. However, a challenge that MQL machining faces is significant thermal growth of machine components (spindle, column, fixture etc) that must be appropriately addressed in order to achieve quality with tight tolerances. A strategy adopted by CNC OEMs is to measure the thermal growth and offset it during machining (machine thermal compensation). A gage bore is mounted in the machine fixture and its position measured using a probe. Required offsets are then calculated and applied for subsequent machining. Different OEMs use different methods for this gage bore-probe based referencing. It is imperative that this compensation be done correctly for it directly impacts quality. There had not been any effective ways of assessing thermal compensation in CNCs. In this paper, we discuss a newly developed artifact based assessment methodology of CNC thermal growth and its compensation. A granite artifact with precision gage bores in multiple planes is mounted on machine table and various gage bore positions are probed. The machine is taken from a cold state to steady warm-up state by cycling it through dry-cycles (programmed similar to a cutting cycle except that no part is present), while intermittently the artifact gage bores are probed. A tolerance specification is established for the deviation of offset compensation values over the test period in order to assess thermal compensation effectiveness. The method has been implemented on several different CNCs and various issues identified and addressed in their thermal compensation schemes and also drove design improvement engineering activities.