A three-spring-in-series model is proposed for the nanobelt ͑NB͒ indentation test. Compared with the previous two-spring-in-series model, which considers the bending stiffness of atomic force microscope cantilever and the indenter/NB contact stiffness, this model adds a third spring of the NB/substrate contact stiffness. NB is highly flexural due to its large aspect ratio of length to thickness. The bending and lift-off of NB form a localized contact with substrate, which makes the Oliver-Pharr method ͓W. C. Oliver and G. M. Pharr, J. Mater. Res. 7, 1564 ͑1992͔͒ and Sneddon method ͓I. N. Sneddon, Int. J. Eng. Sci. 3, 47 ͑1965͔͒ inappropriate for NB indentation test. Because the NB/substrate deformation may have significant impact on the force-indentation depth data obtained in experiment, the two-spring-in-series model can lead to erroneous predictions on the NB mechanical properties. NB in indentation test can be susceptible to the adhesion influence because of its large surface area to volume ratio. NB/substrate contact and adhesion can have direct and significant impact on the interpretation of experimental data. Through the three-spring-in-series model, the influence of NB/substrate contact and adhesion is analyzed and methods of reducing such influence are also suggested.