Well, not really that stupid a question. Pretty long and involved answer though. It starts with knowing your danger space and range error brackets. What are we shooting and how are we ranging the target? What are the conditions and target size/quality? This lets us evaluate elevation error potential. Next, we need to evaluate the target and background. We do a full class with photos in a PowerPoint, so this is an overview.
How likely are we to see impact?
If we do see impact, how does t h at point relate to the target?
If there is no impact, can we see trace?
If no trace, what does our target evaluation and error bracket, combined with our DOPE tell us is the most likely correction?
Can the target's reaction tell us anything about the location of the shot?
Using all the above information, we can now setup for the shot and anticipate what we might need for a correction.
For example, fully 80% of wind calls by less experienced shooters are short, missing on the downwind side. We should know the target size, and have a correction planned. The absolute MINIMUM correction is half the target width. That's if you scrape the edge of the target.
How did we get the range, and what's our history with that method? Are w e commonly long or short of actual?
Did the target react in any way? People will tend to look towards the bullet passage or impact point.
Plan the shot, anticipate the most likely errror(s) and make an educated guess when you have minimal impact/trace to go on.