The literature shows that different possible criteria are used for defining biomass fast pyrolysis. On the basis of a simplified modeling of a cellulose (biomass model compound) particle pyrolysis, the present paper points out that the most often considered parameters (i.e., temperature and heating rate) are inappropriate. They are very difficult to define and measure, and according to their definitions, important errors can be made (kinetic measurements and reactor scaling up). Other possible parameters are also examined such as particle initial size, available heat flux density, heat transfer coefficient, and products elimination efficiency. In order to be able to compare different experimental conditions on a similar basis, it is shown that at the biomass sample level, fast pyrolysis is favoured (enhancement of bio-oil fractions) if two necessary conditions are simultaneously fulfilled. They include high external heat transfer coefficient and efficient products removal.