GW's experience seems to have proven that the process is indeed more expensive, due to their high rejection rate, which must be considerably higher than the rate seen in stores since I'll give them the benefit of the doubt that their internal QC destroys the majority of the product that comes out, so that we only see the least bad examples. This process may have been oversold to management as a cheaper process looking primarily at material cost and ideal situations; a belief that a process is less expensive does not necessarily mean it is true.
4-10 minutes is considerably longer than a metal spin cast, which can be removed in 2 minutes or less. This means productivity goes down considerably for the same number of workers, dropping 50 to 80% is incredibly expensive. Where some hobbyist manufacturers do not place a value on their own time and cost casting time at zero, GW is not able to do so for obvious reasons. Add to this, an increased requirement for QC personnel, and I do not think metal is so expensive that resin casting is a good solution yet for human sized figures.
Further, let's charitably assume 4 minutes per model cure; again, if this model is no good, then someone, i.e. the customer, has to pay for the lost 4 minutes + demold time + QC inspection + safe destruction (I'm assuming resin at close to 0 value). If this rate is higher than 1 in 2 (based on Wayland's report), and about 1 in 10 or 20 for a metal spin cast, you are looking realistically at best, 4 minutes x 2 + (other processes) x 2 to cast one merchantable figure, anything from 15 minutes to half an hour to produce one model sold at 15 quid or less, compared to 2.5 minutes for a spin cast figure. So, best cast scenario, Finecast is currently 5 to 6 times more expensive in terms of labor than spin casting. I doubt very much the savings in metal in a single man sized 28mm figure (estimated to about $1 or less) can possibly make up for this. Since transport costs for man sized figures will most likely be by volumetric weight, the volume of a packaged metal figure will be the major factor and likely be the same as a resin one.
There are also hidden costs in terms of worker safety when handling resins, which tend to have higher toxicity before they set, especially for very exothermic resins. It is plain that this experiment is off to a shaky start on such a large scale. I'd expect to see even more price rises when the accountants add everything up. The question would be whether or not gamers as a whole will reject the higher priced product and force a change.
I suspect the assumption was that the reject rate would be closer to metal, i.e. 1 in 10, which would probably have made resin casting a sensible thing on the prediction that metal prices would continue to rise.