Input vs output metrics
An output metric measures the result you actually want, and usually moves slowly and late. An input metric measures something you can move this week that you believe drives it. Good Key Results mostly target inputs, because a quarter is rarely long enough for an output to give useful feedback.
The trade-off
Outputs are what matter and inputs are what you can steer. Revenue is the output; qualified conversations, activation rate and time to value are the inputs.
Aim only at outputs and you will spend a quarter watching a number that responds too slowly to teach you anything. Aim only at inputs and you can move every one of them and still make no difference, because the belief connecting them to the output was wrong.
The link is a hypothesis
Every input metric carries an assumption: that moving this moves that. Write the assumption down when you choose the metric.
It is the assumption, not the metric, that you are really testing. A quarter where the input moved and the output did not is not a failed quarter. It is an answer, and a valuable one, provided somebody wrote down what was being assumed.
A practical pattern
Put the output in the Objective and the inputs in the Key Results. The Objective says what change you are after; the Key Results are the levers you believe will produce it and can actually check weekly.
Then keep the output on the scorecard, watched continuously, so you can see whether the theory is holding up over a longer arc than one quarter.
What is the difference between an input and an output metric? +
An output measures the result you want and usually moves slowly. An input measures something you can move now that you believe drives the output. Inputs give you feedback inside a quarter; outputs usually do not.
Should Key Results use input or output metrics? +
Mostly inputs, with the output stated in the Objective and monitored on the scorecard. Write down the assumption linking the two, because that assumption is what the quarter is really testing.