Why when testing a new drug for blood pressure would we lear

Why, when testing a new drug for blood pressure, would we learn more about its effectiveness by giving it to half of the participants in a group of 1000 than to all 1000 participants?

We should give placebo medicine to half the participants, and real drug to remaining half. The reason is \"placebo effect\", which is the fake treatment effect leading to improvement of condition even if drug has no real impact.