Finding an Exponential Function from a Graph

An exponential function has the form \( y = a \cdot b^x \), where \(a\) is the starting value (the \(y\)-intercept) and \(b\) is the base that controls how fast the function grows or decays. If you are given a picture of the curve instead of a table or a formula, you can still recover \(a\) and \(b\) just by reading coordinates off the graph. This lesson walks through that process step by step. If you haven't yet seen how the shape of these curves is built, it helps to first look at graphing exponential functions.

Before pulling numbers off a graph, it's worth knowing exactly what each piece of \( y = a \cdot b^x \) means:

• \(a\) is the value of \(y\) when \(x = 0\), so it's always the \(y\)-intercept of the graph.
• \(b\) is the base, and it must be positive and not equal to 1. If \(b > 1\), the graph rises from left to right (growth). If \(0 < b < 1\), the graph falls from left to right (decay).
• The curve never touches the \(x\)-axis; it gets closer and closer to \(y = 0\), which is the horizontal asymptote for this basic form.

To go from a graph to an equation, follow these steps:

• Step 1: Find the point where the graph crosses the \(y\)-axis. That \(y\)-value is \(a\).
• Step 2: Pick a second point on the graph where both coordinates are easy to read.
• Step 3: Substitute the second point's \(x\) and \(y\) values, along with the value of \(a\) you already found, into \( y = a \cdot b^x \), then solve for \(b\).
• Step 4: Write the final equation \( y = a \cdot b^x \) using the values you found, and check it against a third point on the graph if one is available.

Suppose a graph passes through \((0, 2)\) and \((1, 6)\), and the curve rises as \(x\) increases.

Since the graph crosses the \(y\)-axis at \((0, 2)\), we know \(a = 2\). Now use the second point \((1, 6)\):

\( 6 = 2 \cdot b^1 \)

Dividing both sides by 2 gives \( b = 3 \). So the equation of this graph is \( y = 2 \cdot 3^x \). Because \(b = 3 > 1\), this confirms the curve represents growth, which matches what the picture shows. For more on how the growth factor itself is interpreted, see exponential growth and decay by a factor.

Now suppose a graph crosses the \(y\)-axis at \((0, 5)\) and also passes through \((2, 1.25)\), with the curve falling as \(x\) increases.

Here \(a = 5\), since that's the \(y\)-intercept. Substituting the point \((2, 1.25)\):

\( 1.25 = 5 \cdot b^2 \)

Dividing both sides by 5 gives \( 0.25 = b^2 \), so \( b = 0.5 \) (we take the positive root, since the base of an exponential function can't be negative). The equation is \( y = 5 \cdot (0.5)^x \). Since \(0 < b < 1\), this is decay, matching the falling shape of the curve. This kind of problem connects closely to a half life formula for exponential decay, where the base is tied to how much of a quantity remains after each fixed time interval.

Sometimes a graph is zoomed in or shifted so you can't clearly read the \(y\)-intercept, but you can still identify two other points, say \((x_1, y_1)\) and \((x_2, y_2)\). In that case, write two equations:

\( y_1 = a \cdot b^{x_1} \) and \( y_2 = a \cdot b^{x_2} \)

Dividing the second equation by the first cancels out \(a\), leaving an equation in \(b\) alone: \( \dfrac{y_2}{y_1} = b^{x_2 - x_1} \). Solve this for \(b\), then substitute back into either original equation to solve for \(a\). This two-point method works no matter where the intercept falls, and it's the same idea used when a problem gives you coordinates directly instead of a picture.

• Mixing up \(a\) and \(b\): remember \(a\) is read straight from the \(y\)-intercept, while \(b\) comes from solving an equation with a second point.
• Forgetting that \(b\) must be positive: even if algebra seems to allow a negative base, exponential functions only use positive bases other than 1.
• Misreading the direction of the curve: a rising graph should give \(b > 1\) and a falling graph should give \(0 < b < 1\). If your answer doesn't match the shape, recheck your points.
• Choosing a point that's hard to read precisely off the grid; whenever possible, pick a point that lands exactly on a grid intersection.

Finding an exponential function from its graph always comes down to identifying \(a\) from the \(y\)-intercept and solving for \(b\) using one more clear point. Once you're comfortable with the basic form \( y = a \cdot b^x \), the same reasoning extends naturally to graphs that have been shifted or stretched, which is covered in graphing transformations of exponential functions, as well as to real-world situations like population growth or interest rates.

Related lessons

• Algebra Help