Explain why using either compatible numbers or rounding as a method for estimation would be the same for 32×695?

The question here is why using either compatible numbers or rounding as a method for estimation would be the same?
Well, we use rounding as a method of estimation to get a solution in a certain equation faster yet the same.
For example, in the given equation we have 32 x 695
=> Instead of 32, we can use 30
=> instead of 695, we can use 700
=>  30 x 700
=> the answer is 21 000
Now, let’s multiply the exact number
=> 32 * 695
=> 22 240
Not, really exact yet close.
That’s how estimation works.

Ithink it's the first one ( let me know if it right)