8 topics. Each one opens with its key idea, then the detail you need to use it.
1. Interpreting the derivative in context
Key idea: Every context answer names the instant, the direction of change, and the rate in output units per input unit.
When a problem gives a function with units, the exam wants a sentence, not just a number. Suppose C(t) is the temperature of coffee in degrees Fahrenheit t minutes after it is poured, and C'(10) = −3. That means at t = 10 minutes, the temperature is decreasing at a rate of 3°F per minute. Notice three parts: the instant, the direction, and the rate with units.
The units come straight from the definition. A derivative is change in output over change in input. If V(t) is liters of water after t minutes, V'(t) is in liters per minute. V''(t) is in liters per minute per minute, because it is the rate at which the flow rate changes.
Two misreadings show up again and again. The first reads C'(10) = −3 as "the coffee is 3°F at 10 minutes", which describes C(10), the amount itself. The second reads it as "the coffee cooled 3°F over the first 10 minutes", which is an average change over an interval. The derivative is instantaneous, so it describes one moment.
In economics, the derivative is called marginal. If C(x) is the cost of making x widgets and C'(200) = 4, then making about the 201st widget adds roughly $4 to the total cost. Say "about" or "approximately", because the derivative only estimates the cost of one more unit.
2. Straight-line motion
Key idea: Velocity sign gives direction; speeding up means v and a share a sign; total distance needs a break at every turn.
A particle moving on a line has position s(t). Velocity is v(t) = s'(t), and acceleration is a(t) = v'(t) = s''(t). Velocity has a sign, which gives direction. Positive means moving right or up, and negative means moving left or down. Speed is the absolute value, |v(t)|, so speed is never negative.
Take s(t) = t³ − 9t² + 24t for t ≥ 0. Then v(t) = 3t² − 18t + 24 = 3(t − 2)(t − 4), and a(t) = 6t − 18. Velocity is positive before t = 2, negative between 2 and 4, and positive after 4. So the particle changes direction at t = 2 and t = 4. A zero of velocity alone is not enough; the sign must flip.
The particle is speeding up when velocity and acceleration have the same sign, and slowing down when they differ. At t = 1, v = 9 and a = −12, so it is moving right but slowing. At t = 5, v = 9 and a = 12, so it is moving right and speeding up. Students who only check whether a is positive get this wrong.
Total distance is not the same as displacement. Here s(0) = 0, s(2) = 20, s(4) = 16 and s(5) = 20. Displacement from 0 to 5 is 20. Total distance adds each leg: 20 + 4 + 4 = 28. Break the trip at every direction change.
3. Related rates
Key idea: Write the equation, differentiate with respect to t, and only then substitute the values at the instant asked.
In a related-rates problem, two or more quantities change over time and are tied by one equation. You differentiate that equation with respect to t. Every variable that changes picks up its own rate through the chain rule, so x becomes dx/dt and y² becomes 2y·dy/dt.
The procedure is fixed. First, draw the picture and name the variables. Second, write an equation true at every moment. Third, differentiate with respect to t. Fourth, plug in the values at the given instant and solve. Plug in numbers only after you differentiate. If you substitute x = 5 first, x becomes a constant, and its rate wrongly disappears.
Take a 13-foot ladder whose bottom slides away from a wall at 2 ft/s. Find how fast the top falls when the bottom is 5 feet out. The equation is x² + y² = 169. Differentiating gives 2x·dx/dt + 2y·dy/dt = 0. At x = 5, the Pythagorean theorem gives y = 12. So 2(5)(2) + 2(12)·dy/dt = 0, and dy/dt = −20/24 = −5/6 ft/s. The negative sign means the top is moving down.
Two common errors are worth naming. The rates are not equal just because the ladder is rigid; they scale by the ratio x/y. Many students also flip that ratio, giving −12/5 instead of −5/12 times the bottom's speed. Solving dy/dt = −(x/y)·dx/dt explicitly prevents the flip.
4. Critical points, extreme values and the first derivative test
Key idea: Critical points come from f'; classify them by sign change; on a closed interval, compare f at critical points and both endpoints.
A critical point is an x in the domain where f'(x) = 0 or f'(x) does not exist. You find them from the derivative, never from f. For f(x) = x³ − 3x, f'(x) = 3(x − 1)(x + 1), so the critical points are x = ±1. Setting f(x) = 0 instead gives the zeros, x = 0 and ±√3, which is a different question. Points where f' is undefined count too. For f(x) = x^(2/3), the cusp at x = 0 is a critical point.
The first derivative test classifies a critical point by the sign of f' on each side. If f' changes from positive to negative at c, f has a relative maximum at c. Negative to positive gives a relative minimum. No sign change means no extremum. For f'(x) = (x − 1)²(x + 3), the sign changes at x = −3, giving a minimum. At x = 1 the squared factor keeps f' positive on both sides, so x = 1 is neither.
For an absolute extremum on a closed interval [a, b], use the candidates test. The Extreme Value Theorem promises a maximum and minimum when f is continuous there. Evaluate f at every critical point and at both endpoints, then compare. For f(x) = x³ − 12x on [−3, 5], the values are f(−3) = 9, f(−2) = 16, f(2) = −16 and f(5) = 65. The absolute maximum is 65 at the endpoint x = 5, not at the critical point. Skipping endpoints is a classic way to lose this question.
5. Concavity and the second derivative test
Key idea: Inflection needs a sign change in f''; the second derivative test gives min when f'' > 0, max when f'' < 0, and nothing when f'' = 0.
Concavity describes how the slope changes. If f'' > 0 on an interval, f' is increasing there, so the curve bends upward like a cup. If f'' < 0, f' is decreasing, and the curve bends downward. Concavity is independent of direction. A curve with f' > 0 and f'' < 0 is increasing and concave down. Mixing up the two signs accounts for most of the wrong choices on a typical question.
An inflection point is where the concavity changes, which means f'' changes sign. f''(c) = 0 is not enough. For f(x) = x⁴, f''(0) = 0, but f'' = 12x² is positive on both sides, so there is no inflection point at x = 0.
The second derivative test is a shortcut for classifying a critical point c where f'(c) = 0. If f''(c) > 0, the curve is a cup there, so c is a relative minimum. If f''(c) < 0, it is a cap, so c is a relative maximum. If f''(c) = 0, the test tells you nothing.
Take f(x) = x⁴ − 4x³. Then f'(x) = 4x²(x − 3) and f''(x) = 12x(x − 2). At x = 3, f''(3) = 36 > 0, so there is a relative minimum. At x = 0, f''(0) = 0, so the test fails. The first derivative test settles it: f' is negative on both sides of 0, so there is no extremum. But f'' does change sign at x = 0 and at x = 2, so both are inflection points.
6. Connecting f, f' and f'' from graphs
Key idea: On a graph of f', zeros that cross mark extrema of f, and peaks or valleys mark inflection points of f.
Many questions show you the graph of f' and ask about f. This is where students lose the most points, because they read the graph as if it were f. Before you answer, write at the top of your work which function the picture shows.
Use a ladder of translations. Where the graph of f' lies above the x-axis, f is increasing. Where f' crosses the axis from above to below, f has a relative maximum. Where it crosses from below to above, f has a relative minimum. Where f' merely touches the axis and turns back, f has no extremum.
One level up, the slope of the f' graph is f''. Where the f' graph rises, f is concave up. Where it falls, f is concave down. So the inflection points of f sit at the peaks and valleys of the f' graph, not at its zeros. That is the single most tested link in this topic.
Curve sketching runs the ladder in reverse. Find where f' is positive or negative and where f'' is positive or negative. Then split the number line at every critical point and every possible inflection point. In each interval, combine the two signs into one of four shapes: rising cup, rising cap, falling cup or falling cap. Plot the key points and join the shapes. A sign table is fine for your own thinking, but on a free response you still write the sentence.
7. Optimization
Key idea: Reduce to one variable with the constraint, set its domain, find critical points, and justify the absolute extremum.
An optimization problem asks for the largest or smallest value of some quantity under a limit. The method has five steps. Name the quantity to optimize and write it as a formula. Use the constraint to reduce it to one variable. State the domain that makes physical sense. Find the critical points. Then justify that your answer is the maximum or minimum.
Take a 12-inch square sheet. You cut a square of side x from each corner and fold up the sides to make an open box. The volume is V(x) = x(12 − 2x)², with domain 0 < x < 6. Differentiating with the product rule and factoring gives V'(x) = (12 − 2x)(12 − 6x). The only critical point inside the domain is x = 2, where V(2) = 2(8)² = 128 cubic inches. V' is positive on (0, 2) and negative on (2, 6). Because this is the only critical point and V changes from increasing to decreasing, V has its absolute maximum at x = 2.
The constraint is where most wrong answers start. In a fence-against-a-barn problem, only three sides need fencing, so 200 feet gives 2x + y = 200, not 2x + 2y = 200. Treating it as four sides gives a square and an area far too small. Splitting the fence into three equal pieces is also wrong, because the best shape here is not equal-sided. Always let the derivative choose the dimensions.
8. L'Hôpital's Rule
Key idea: Confirm 0/0 or ∞/∞ in writing, then differentiate the top and bottom separately and take the limit again.
Some limits come out as 0/0 or ∞/∞ when you substitute. These are indeterminate forms, because the answer depends on how fast the top and bottom approach their limits. L'Hôpital's Rule says that, in these cases, the limit of f(x)/g(x) equals the limit of f'(x)/g'(x), if that second limit exists.
Differentiate the top and bottom separately. Do not use the quotient rule. For the limit as x → 0 of (eˣ − 1)/x, the top and bottom both go to 0. Their derivatives are eˣ and 1, and eˣ approaches e⁰ = 1, so the limit is 1. The wrong answers on this item come from stopping early. Answering 0 looks only at the numerator. Answering ∞ treats 0/0 as if it blows up. Answering e plugs in x = 1 instead of x = 0.
You may need the rule more than once. For (1 − cos x)/x² as x → 0, one round gives sin x/(2x), still 0/0. A second round gives cos x/2, which approaches 1/2.
Check the form before you apply the rule. As x → 1, (x² + 1)/(x + 1) is 2/2 = 1 by substitution. Applying L'Hôpital's anyway gives 2x/1 → 2, which is wrong. On a free response, show the check in writing: state that the limits of the numerator and of the denominator are both 0, or both infinite. Readers look for that line before they award the answer.