Generate a follow-up sub-lesson on any aspect of this topic
Guide complete
U-Substitution in Integration, Made Intuitive
Generate a follow-up sub-lesson on any aspect of this topic
U-Substitution in Integration, Made Intuitive
Get fluent at recognizing when substitution in integration works, picking a smart , and rewriting the whole integrand cleanly. You will see the pattern as a reverse chain rule, practice the derivative-nearby cue, and leave with a checklist that prevents mixed variables and missed bounds.
You are staring at an integral like and it feels close to something you know, but not quite. The trick is that many messy-looking integrals are really a simple chain rule derivative in disguise, just running backward. U-substitution is the method for turning the disguise back into a basic integral you can actually do, without guessing antiderivatives from scratch.
If you remember one idea, make it this. The chain rule says that if , then . U-substitution looks for that same structure inside an integral and reverses it by naming the inside.
You pick . Then . That lets you swap a complicated integrand in for a simpler one in .
A good mental model is inside and outside.
See the chain rule and its reverse side by side.
Same pattern
U-sub does not invent a new method. It just renames the inside so the chain rule pattern becomes obvious and usable.
U-substitution tends to work when the integrand is a composite function with its inner derivative sitting nearby as a factor.
These are common green lights.
For example, works because the inside has derivative , and a constant multiple is easy to fix.
But is missing the . That does not mean impossible, just that plain u-sub does not immediately click.
Play with how close the inner derivative needs to be.
Choosing is usually simpler than it feels. Pick the expression that is being fed into another function. Then compute and see whether the rest of the integrand can become part of after a constant adjustment.
A clean workflow is:
If you see , the inside is . If you see , the inside is .
Compute .
If , then , so .
If , then , so .
That last line is the whole game. You are trying to find a chunk of the original integrand that can be replaced by .
Compare a couple plausible choices and notice which one collapses the integral fastest.
After you choose and compute , the goal is strict. Rewrite the integral so it contains only and . No left anywhere.
Start from the template
Two common cleanup moves do most of the work.
Example.
Let , so . Then the whole becomes and you get
If instead you had , you would still let , but then you need to solve for whatever remains. gives . That is how you replace leftover terms with terms.
No mixing
If you see and in the same integrand after substituting, stop. Go back and rewrite until the variable is fully eliminated.
For definite integrals, you can avoid back-substitution by changing limits.
If , then when , the new lower bound is . When , the new upper bound is .
Example.
Let , . Bounds become , .
Let , so and .
Let , then .
Let , so and .
Let , .
Explore optional notes that address common traps and alternate choices.
Sometimes the integrand has a composite shape but the missing piece is not something you can produce with a constant factor. Classic example.
If , then , but there is no factor available. Plain u-sub stalls.
Other times there are two competing insides and neither derivative appears, like . That is a job for integration by parts.
A quick set of alternatives to try when u-sub fails.
Bring an integral you are unsure about and check whether substitution applies.
When you are stuck, run this mental model in order.
Practice turns this from a trick into pattern recognition. Pick a few integrals each day and force yourself to write the substitution step cleanly, including the to rewrite. That is where most errors either disappear or get exposed fast.