Improvement fails when players try to fix everything at once. Counter-strafe, crosshair height, sensitivity, spray pattern, and pre-aim are all real things, and chasing all of them in the same week usually moves none of them. A demo analysis is most useful when it produces one priority, one drill focus, and one measurable retest.
This page is the workflow that connects a demo to a focused training week. For the broader fundamentals see the CS2 aim training guide.
Upload a demo"Practice your aim" is not a plan. It is a category. Without a target metric, hours in an aim trainer don't transfer cleanly into matchmaking, and you cannot tell whether the time was spent well. Worse, mixing too many drills in one session means none of them get reps deep enough to build a habit.
Pick one. Move it. Then pick the next one.
Use a recent competitive demo where you played roughly your normal way. Don't pick a smurf game. Don't pick a tilt game. The point is to see your real habits, not your best or worst case.
If your sensitivity is the suspect leak, a deathmatch demo gives more flick data per minute, which makes the recommendation more reliable. See the CS2 sensitivity guide.
"Reliable" matters here. A scary number computed from twelve engagements is not actually scary. Look at metrics with enough sample size to be decision-grade. The demo analysis limitations page explains why this matters.
If two metrics look bad and both are reliable, pick the upstream one. Movement is upstream of placement, placement is upstream of first-bullet, first-bullet is upstream of spray. Fix upstream first.
One. Not three. The five most useful priorities a demo can hand you:
Short, daily, specific. 15–25 minutes of the chosen drill is more useful than a two-hour session you do once. Avoid switching focus mid-week — even if a different leak feels worse on a given night.
The honest test is whether the metric moved in a real match, not whether the drill felt better in an empty server. Same map type, same role, same general skill level. Compare the targeted metric against the baseline demo from Step 1.