Destructive testing has always been part of every engineering development projects.

When developing new parts it's common to make a lot of test parts and stress them to failure to see how they react.

For innovative design it can take several iterations before finding the right material/design. Each destructive testing is bringing valuable information.

Knowing exactly how a part will fail gives extremely valuable information on how to build a part that will NOT fail and everyone does that including NASA.

SpaceX has just brought this philosophy to whole different level by doing destructive testing on the whole rocket. The best example is that on the last flight they purposefully removed heatshields on some area of the Starship and added sensors in the area to see how it would impact the ship.

The can afford to do that because they focused on building a rocket factory to mass produce starships rather than building a rocket. It means that even if they were not launching it the factory would still produce Starships.

PS: SLS is not 10x the cost of Starship. According to an independent report ( source ) Right now the estimated cost of a Starship launch is estimated around $90 million, one the program is operational the cost of a Starship launch is estimated to be around $10 millions.

A SLS launch is estimated to be around $4.1 billion

So a Starship launch is 40 to 400 time cheaper than a SLS launch