The concept of CAPE is based on a set of assumptions that make it a bad parameter to calculate the speeds of updrafts. It is assumed that…

...mixing between the parcel and the air in its environment can be neglected, so that there is no exchange of heat and moisture (in other words: no entrainment occurs)

...the weight of water droplets and ice is negligible

...pressure perturbations can be ignored

Those assumptions are generally bad assumptions. This means that CAPE values should never been taken literally to forecast updraft speeds. A consequence of entrainment is that the vertical speeds in updrafts are not only dependent on the total amount of CAPE available to the air entering the updraft, but also on the vertical distribution of CAPE. If, for example, 1000 J/kg of CAPE is released over a 10 km deep layer, entrainment will likely reduce the parcel's excess temperature stronger during its ascent than if the same amount is released over 5 km. In the latter case, entrainment will not have as much of a chance and larger upward speeds can be expected.