Recent data obtained for electron scattering by biomass molecular fragments, indicated that low-energy resonances may have an important role in the de-lignification of biomass through a plasma pre-treatment. To support these findings, we present new experimental evidence of the predicted dissociation pathways on plasma treatment of biomass. An important question is how accurate must the experimental and/or the theoretical data be in order to indicate that plasma modelings can be really useful in understanding plasma applications? In this paper, we initiate a discussion on the role of data accuracy of experimental and theoretical electron-molecule scattering cross sections in plasma modeling. First we review technological motivations for carrying out electron-molecule scattering studies. Then we point out the theoretical and experimental limitations that prevent us from obtaining more accurate cross sections. We present a few examples involving biomass molecular fragments, to illustrate theoretical inaccuracies on: resonances positions and widths, electronic excitation, superelastic cross sections from metastable states and due to multichannel effects on the momentum transfer cross sections. On the experimental side we briefly describe challenges in making absolute cross sections measurements with biomass species and radicals. And finally, through a simulation of a N2 plasma, we illustrate the impact on the simulation due to inaccuracies on the resonance positions and widths and due to multichannel effects on the momentum transfer cross sections.