Proper judgement on rangeland state needs appropriate sampling plan and accurately estimation of plant characteristics. Sample shape and size are critical issues in vegetation measurement. Thus, decision on appropriate quadrate shape that enables us to determine several parameters accurately and timely would increase sampling efficiency. Several criterias including accuracy, perimeter/area ratio, spent time for measurement, usability and ease of use as well as three variance and or covariance dependent criterias were used to decide on proper plot size to determine multi-variables of rangelands, i.e. different life forms composition, sand and gravels, litter and bare soil coverage. Different quadrate sizes including 1×1, 1×2, 1.4×1.4, 0.4×4, 2×2, 2.8×2.8 and 4×4m were established along a 640 m long transect within a homogeneous vegetation type. Within each quadrate, species canopy cover, litter, sand and gravel coverage in addition to spent time for measurement were estimated and recorded. Life forms composition in each plot was calculated by summing each life form species. An Analytical Hierarchical Process (AHP) was performed to find out appropriate sampling quadrate from aforementioned quadrate sizes. Moreover, 5 repeats of nested quadrates from 25×25cm to 16×16m were established to determine sampling minimal area. Efficacy of minimal area method with this multi criteria method that synchronously incorporated a number of criteria was compared. Results showed that accuracy of multi-variable estimation raises with increasing sampling quadrate size (area) until reaches a 4 square meter quadrate (2×2m) but after that more or less stabilized. Also, on a constant sample size (area), efficiency of square plots is higher than rectangles for estimation of objective variables. AHP results showed that based on accuracy, time, perimeter/area ratio, and usability and ease of application criteria a 4 square meter (i.e., 2×2m) plot is the most appropriate alternative for synchronic measurement of multi-variables. Repeated nested plots showed at least a 32×32m plot for minimal area that is not applicable in practice. We can conclude that 1) minimal area that is estimated using nested plots are not necessarily the most appropriate minimal quadrate size in heterogeneous vegetations for multi-variable measurement, and 2) by considering multi criteria and choosing the best option (alternative) from different quadrates, researcher will increase sampling efficiency.