Early peeking of A/B testing p-value

I know it is a statistical sin to early-peek statistical difference. But if actions will not change, and doing the calculation is just purely informational and at most inconsequential, does that violate any statistical principle?

a/b testing b testing change datascience difference p-value statistical testing value will