Topic: About Test Effort Estimation

Just curious about how effort estimations for projects are being carried out. I assume there is no single "standard" way of going about this. My assumption is there is no data from previous projects to fall back on.

I made a small list of the methods I've come across :
1. 40% of development time. Nobody knows where the magic 40 materialized from. Yet blind idolatry prevents a change in this figure upwards or downwards! ;-)

2. Use the FP figures with some modifications. usually comes to 90% of development time. Accuracy ranges from + or - 100% !

3. As long as it takes, until we run out of time or moolah ! ;-)

What method do you use - is there a method to the madness ?