The Stanford “marshmallow experiment” was a series of studies on delayed gratification led by psychologist Walter Mischel, in the late ’60s and early ’70s. A recent revisiting of this experiment can be seen in this short video and, whether you believe in the claims of follow-up studies or not, it’s interesting to watch the different behaviours of different kids (even the difference between two twin boys in this regard).
I was working on a presentation about thinking of testing as being an information service provider when this video came to my attention (thanks to my wife). It got me thinking about the people who make release decisions for the software we work on.
We can provide one marshmallows worth of valuable product information right now and you can release based on that information. Or, we can spend some more time doing really good testing and then give you two marshmallows worth of information, to make an even more informed release decision!
The problem with this analogy is that neither marshmallow necessarily tastes very good – and the second one is likely to taste worse than the first, right?!
Anyway, maybe just give your business folks a marshmallow anyway, it might just make their day.
Rockin' and Testing All Over The World 
…and try to stop managers from cooking the marshmallow prior to the decision makers tasting it (coz they always taste better cooked over a fire).
Don’t even think about putting the marshmallow on a nice plate either, just stick it in the customer’s mouth while they’re trying to ask you a question.
(I’ve been working with SaaS products lately, from the customer’s perspective).