We analyse the CO-dark molecular gas content of simulated molecular clouds from the SILCC-Zoom project. The simulations reach a resolution of 0.1 pc and include H-2 and CO formation, radiative stellar feedback and magnetic fields. CO-dark gas is found in regions with local visual extinctions A(V,3D) similar to 0.2-1.5, number densities of 10-10(3) cm(-3) and gas temperatures of few 10-100 K. CO-bright gas is found at number densities above 300 cm(-3) and temperatures below 50 K. The CO-dark gas fractions range from 40 per cent to 95 per cent and scale inversely with the amount of well-shielded gas ( A(V, 3D) greater than or similar to 1.5), which is smaller in magnetized molecular clouds. We show that the density, chemical abundances and A(V, 3D) along a given line-of-sight cannot be properly determined from projected quantities. As an example, pixels with a projected visual extinction of A(V, 2D) similar or equal to 2.5-5 can be both, CO-bright or CO-dark, which can be attributed to the presence or absence of strong density enhancements along the line-ofsight. By producing synthetic CO(1-0) emission maps of the simulations with RADMC-3D, we show that about 15-65 per cent of the H-2 is in regions with intensities below the detection limit. Our clouds have XCO-factors around 1.5 x 10(20) cm(-2) (K km s(-1))(-1) with a spread of up to a factor similar to 4, implying a similar uncertainty in the derived total H-2 masses and even worse for individual pixels. Based on our results, we suggest a new approach to determine the H-2 mass, which relies on the availability of CO(1-0) emission and A(V, 2D) maps. It reduces the uncertainty of the clouds' overall H-2 mass to a factor of less than or similar to 1.8 and for individual pixels, i.e. on sub-pc scales, to a factor of less than or similar to 3.